Oncology

A global analysis challenged the notion that a rise in cancer is disproportionately affecting younger adults, finding instead that several cancer types previously seen rising in younger adults are also increasing in older adults.

More specifically, the analysis found that incidence rates for thyroid cancer, breast cancer, kidney cancer, endometrial cancer, and leukemia increased similarly in both younger and older adults in most countries over a 15-year period. Colorectal cancer (CRC) was the exception, where incidence rates increased in younger adults in most countries but only increased slightly in older adults in about half and decreased in about one quarter.

“Our findings suggest that whatever is triggering the rise in these cancers is more likely to be common across all age groups, rather than specific to cancers in the under 50s, since there were similar increases in younger and older adults,” Amy Berrington de González, DPhil, The Institute of Cancer Research, London, England, who led the study, said in a statement.

The authors of an editorial agreed, adding that the growing “concern about increasing cancer rates should recognize that this increase is not restricted to young adults but affects all generations.”

The study and editorial were published recently in Annals of Internal Medicine.

 

Data Defy Early-Onset Cancer Epidemic Narrative

A growing body of evidence suggests that cancer incidence rates are increasing among younger adults in many countries. However, studies tracking international trends have largely evaluated cancer incidence in younger adults without comparing these trends in older adults or analyses have focused the age comparison in individual countries, Berrington de González and colleagues explained.

To better understand cancer incidence trends across countries and age groups, the researchers evaluated cancer trends in 42 countries between 2003 and 2017, focusing on 13 cancer types previously reported to be climbing in adults younger than age 50 years.

The researchers found that incidence rates for six of the 13 cancer types increased among younger adults (aged 20-49 years) in more than three quarters of the countries studied.

The largest increase was in thyroid cancer (median average annual percentage change [AAPC], 3.57%), followed by kidney cancer (median AAPC, 2.21%), endometrial cancer (median AAPC, 1.66%), CRC (median AAPC, 1.45%), breast cancer (median AAPC, 0.89%), and leukemia (median AAPC, 0.78%).

But with the exception of CRC, incidence rates for these cancers increased to a similar degree in adults aged 50 years or older — with median AAPCs of 3% (vs 3.57%) for thyroid cancer, 1.65% (vs 2.21%) for kidney cancer, 1.20% (vs 1.66%) for endometrial cancer, 0.86% (vs 0.89%) for breast cancer, and 0.61% (vs 0.78%) for leukemia.

In older adults, CRC incidence rates only increased in about half the countries (median AAPC, 0.37%), and the annual percentage change was much greater in younger than older adults in nearly 70% of countries. CRC incidence rates in older individuals also decreased in nearly 25% of countries.

Why is CRC an apparent outlier?

“Bowel cancer screening not only helps detect cancer at earlier stages but also helps prevent cancer through the removal of premalignant lesions,” Berrington de González said. “This could be why bowel cancer cases seem to be rising faster in younger adults — we’re getting better at preventing them developing in older adults.”

The incidence of certain cancers also declined in younger adults. Specifically, rates of liver, oral, esophageal, and stomach cancers decreased in younger adults in more than half of countries assessed, with median AAPCs of -0.14% for liver, -0.42% for oral, -0.92% for esophageal, and -1.62% for stomach cancers.

Over half of countries also saw declining rates of stomach (median AAPC, -2.05%) and esophageal (median AAPC, -0.25%) cancers among older adults, while rates of liver and oral cancers increased in older individuals (median AAPC, 2.17% and 0.49%, respectively).

For gallbladder, pancreatic, and prostate cancers — three other cancers previously found to be increasing in younger adults — the researchers reported that incidence rates increased in younger adults in just over half of countries (median AAPCs, 3.2% for prostate cancer, 0.49% for gallbladder cancer, and 1% for pancreatic cancer). Incidence rates also often increased in older adults but to a lesser extent (median AAPCs, 0.75% for prostate cancer, -0.10% for gallbladder, and 0.96% for pancreatic cancer).

 

True Rise or Increased Scrutiny?

Why are cancer rates increasing?

“Understanding factors that contribute to the increase in incidence across the age spectrum was beyond the scope of the study,” editorialists Christopher Cann, MD, Fox Chase Cancer Center, and Efrat Dotan, MD, University of Pennsylvania Health System, both in Philadelphia, wrote.

Several studies have suggested that rising rates of obesity could help explain increasing cancer incidence, particularly in younger adults. In fact, “the cancers that we identified as increasing are all obesity-related cancers, including endometrial and kidney cancer,” Berrington de González said. However, so far, the evidence on this link remains unclear, she acknowledged.

Weighing in on the study, Gilbert Welch, MD, Brigham and Women’s Hospital, Boston, told this news organization that it’s “critically important” to distinguish between two explanations for rising cancer incidence.

There may be an increase in the true occurrence of clinically meaningful cancer, which “warrants investigation into biologic explanations, better treatment, and perhaps more testing,” Welch said.

But it may instead reflect changes in diagnostic scrutiny. “Simply put, whenever we doctors look harder for cancer, we find more,” Welch said. “And there are lots of ways to look harder: testing more people, testing people more frequently, using tests with increasing ability to detect small irregularities, and using lower diagnostic thresholds for labeling these as cancer.”

If increased incidence is the result of greater diagnostic scrutiny, searching for biologic causes is bound to be unproductive and more testing will only aggravate the problem, he explained.

Welch pointed out that the fastest rising cancer in both younger and older adults was thyroid cancer (AAPC, ≥ 3%), which is “exquisitely sensitive” to diagnostic scrutiny.

Take what happened in South Korea. Around 2000, the government of South Korea started a national screening program for breast, colon, and stomach cancers. Doctors and hospitals often added on ultrasound scans for thyroid cancer for a small additional fee.

“A decade later the rate of thyroid cancer diagnosis had increased 15-fold, turning what was once a rare cancer into the most common cancer in Korea,” Welch said. “But the death rate from thyroid cancer did not change. This was not an epidemic of disease; this was an epidemic of diagnosis.”

Welch also noted that the study authors and editorialists put the finding in perspective by explaining that, despite the rising rates of certain cancers in younger adults, cancer remains rare in these adults.

Welch highlighted that, for younger adults in the US, cancer death rates in young adults have cut in half over the last 30 years. “Cancer accounts for only 10% of deaths in young people in the US — and that number is falling,” Welch said.

The study was funded by the Institute of Cancer Research and the National Institutes of Health Intramural Research Program. Disclosures for authors and editorial writers are available with the original articles. Welch reported receiving royalties from three books including “Should I be tested for cancer?”

A version of this article first appeared on Medscape.com.

A global analysis challenged the notion that a rise in cancer is disproportionately affecting younger adults, finding instead that several cancer types previously seen rising in younger adults are also increasing in older adults.

More specifically, the analysis found that incidence rates for thyroid cancer, breast cancer, kidney cancer, endometrial cancer, and leukemia increased similarly in both younger and older adults in most countries over a 15-year period. Colorectal cancer (CRC) was the exception, where incidence rates increased in younger adults in most countries but only increased slightly in older adults in about half and decreased in about one quarter.

“Our findings suggest that whatever is triggering the rise in these cancers is more likely to be common across all age groups, rather than specific to cancers in the under 50s, since there were similar increases in younger and older adults,” Amy Berrington de González, DPhil, The Institute of Cancer Research, London, England, who led the study, said in a statement.

The authors of an editorial agreed, adding that the growing “concern about increasing cancer rates should recognize that this increase is not restricted to young adults but affects all generations.”

The study and editorial were published recently in Annals of Internal Medicine.

 

Data Defy Early-Onset Cancer Epidemic Narrative

A growing body of evidence suggests that cancer incidence rates are increasing among younger adults in many countries. However, studies tracking international trends have largely evaluated cancer incidence in younger adults without comparing these trends in older adults or analyses have focused the age comparison in individual countries, Berrington de González and colleagues explained.

To better understand cancer incidence trends across countries and age groups, the researchers evaluated cancer trends in 42 countries between 2003 and 2017, focusing on 13 cancer types previously reported to be climbing in adults younger than age 50 years.

The researchers found that incidence rates for six of the 13 cancer types increased among younger adults (aged 20-49 years) in more than three quarters of the countries studied.

The largest increase was in thyroid cancer (median average annual percentage change [AAPC], 3.57%), followed by kidney cancer (median AAPC, 2.21%), endometrial cancer (median AAPC, 1.66%), CRC (median AAPC, 1.45%), breast cancer (median AAPC, 0.89%), and leukemia (median AAPC, 0.78%).

But with the exception of CRC, incidence rates for these cancers increased to a similar degree in adults aged 50 years or older — with median AAPCs of 3% (vs 3.57%) for thyroid cancer, 1.65% (vs 2.21%) for kidney cancer, 1.20% (vs 1.66%) for endometrial cancer, 0.86% (vs 0.89%) for breast cancer, and 0.61% (vs 0.78%) for leukemia.

In older adults, CRC incidence rates only increased in about half the countries (median AAPC, 0.37%), and the annual percentage change was much greater in younger than older adults in nearly 70% of countries. CRC incidence rates in older individuals also decreased in nearly 25% of countries.

Why is CRC an apparent outlier?

“Bowel cancer screening not only helps detect cancer at earlier stages but also helps prevent cancer through the removal of premalignant lesions,” Berrington de González said. “This could be why bowel cancer cases seem to be rising faster in younger adults — we’re getting better at preventing them developing in older adults.”

The incidence of certain cancers also declined in younger adults. Specifically, rates of liver, oral, esophageal, and stomach cancers decreased in younger adults in more than half of countries assessed, with median AAPCs of -0.14% for liver, -0.42% for oral, -0.92% for esophageal, and -1.62% for stomach cancers.

Over half of countries also saw declining rates of stomach (median AAPC, -2.05%) and esophageal (median AAPC, -0.25%) cancers among older adults, while rates of liver and oral cancers increased in older individuals (median AAPC, 2.17% and 0.49%, respectively).

For gallbladder, pancreatic, and prostate cancers — three other cancers previously found to be increasing in younger adults — the researchers reported that incidence rates increased in younger adults in just over half of countries (median AAPCs, 3.2% for prostate cancer, 0.49% for gallbladder cancer, and 1% for pancreatic cancer). Incidence rates also often increased in older adults but to a lesser extent (median AAPCs, 0.75% for prostate cancer, -0.10% for gallbladder, and 0.96% for pancreatic cancer).

 

True Rise or Increased Scrutiny?

Why are cancer rates increasing?

“Understanding factors that contribute to the increase in incidence across the age spectrum was beyond the scope of the study,” editorialists Christopher Cann, MD, Fox Chase Cancer Center, and Efrat Dotan, MD, University of Pennsylvania Health System, both in Philadelphia, wrote.

Several studies have suggested that rising rates of obesity could help explain increasing cancer incidence, particularly in younger adults. In fact, “the cancers that we identified as increasing are all obesity-related cancers, including endometrial and kidney cancer,” Berrington de González said. However, so far, the evidence on this link remains unclear, she acknowledged.

Weighing in on the study, Gilbert Welch, MD, Brigham and Women’s Hospital, Boston, told this news organization that it’s “critically important” to distinguish between two explanations for rising cancer incidence.

There may be an increase in the true occurrence of clinically meaningful cancer, which “warrants investigation into biologic explanations, better treatment, and perhaps more testing,” Welch said.

But it may instead reflect changes in diagnostic scrutiny. “Simply put, whenever we doctors look harder for cancer, we find more,” Welch said. “And there are lots of ways to look harder: testing more people, testing people more frequently, using tests with increasing ability to detect small irregularities, and using lower diagnostic thresholds for labeling these as cancer.”

If increased incidence is the result of greater diagnostic scrutiny, searching for biologic causes is bound to be unproductive and more testing will only aggravate the problem, he explained.

Welch pointed out that the fastest rising cancer in both younger and older adults was thyroid cancer (AAPC, ≥ 3%), which is “exquisitely sensitive” to diagnostic scrutiny.

Take what happened in South Korea. Around 2000, the government of South Korea started a national screening program for breast, colon, and stomach cancers. Doctors and hospitals often added on ultrasound scans for thyroid cancer for a small additional fee.

“A decade later the rate of thyroid cancer diagnosis had increased 15-fold, turning what was once a rare cancer into the most common cancer in Korea,” Welch said. “But the death rate from thyroid cancer did not change. This was not an epidemic of disease; this was an epidemic of diagnosis.”

Welch also noted that the study authors and editorialists put the finding in perspective by explaining that, despite the rising rates of certain cancers in younger adults, cancer remains rare in these adults.

Welch highlighted that, for younger adults in the US, cancer death rates in young adults have cut in half over the last 30 years. “Cancer accounts for only 10% of deaths in young people in the US — and that number is falling,” Welch said.

The study was funded by the Institute of Cancer Research and the National Institutes of Health Intramural Research Program. Disclosures for authors and editorial writers are available with the original articles. Welch reported receiving royalties from three books including “Should I be tested for cancer?”

A version of this article first appeared on Medscape.com.

A global analysis challenged the notion that a rise in cancer is disproportionately affecting younger adults, finding instead that several cancer types previously seen rising in younger adults are also increasing in older adults.

More specifically, the analysis found that incidence rates for thyroid cancer, breast cancer, kidney cancer, endometrial cancer, and leukemia increased similarly in both younger and older adults in most countries over a 15-year period. Colorectal cancer (CRC) was the exception, where incidence rates increased in younger adults in most countries but only increased slightly in older adults in about half and decreased in about one quarter.

“Our findings suggest that whatever is triggering the rise in these cancers is more likely to be common across all age groups, rather than specific to cancers in the under 50s, since there were similar increases in younger and older adults,” Amy Berrington de González, DPhil, The Institute of Cancer Research, London, England, who led the study, said in a statement.

The authors of an editorial agreed, adding that the growing “concern about increasing cancer rates should recognize that this increase is not restricted to young adults but affects all generations.”

The study and editorial were published recently in Annals of Internal Medicine.

 

Data Defy Early-Onset Cancer Epidemic Narrative

A growing body of evidence suggests that cancer incidence rates are increasing among younger adults in many countries. However, studies tracking international trends have largely evaluated cancer incidence in younger adults without comparing these trends in older adults or analyses have focused the age comparison in individual countries, Berrington de González and colleagues explained.

To better understand cancer incidence trends across countries and age groups, the researchers evaluated cancer trends in 42 countries between 2003 and 2017, focusing on 13 cancer types previously reported to be climbing in adults younger than age 50 years.

The researchers found that incidence rates for six of the 13 cancer types increased among younger adults (aged 20-49 years) in more than three quarters of the countries studied.

The largest increase was in thyroid cancer (median average annual percentage change [AAPC], 3.57%), followed by kidney cancer (median AAPC, 2.21%), endometrial cancer (median AAPC, 1.66%), CRC (median AAPC, 1.45%), breast cancer (median AAPC, 0.89%), and leukemia (median AAPC, 0.78%).

But with the exception of CRC, incidence rates for these cancers increased to a similar degree in adults aged 50 years or older — with median AAPCs of 3% (vs 3.57%) for thyroid cancer, 1.65% (vs 2.21%) for kidney cancer, 1.20% (vs 1.66%) for endometrial cancer, 0.86% (vs 0.89%) for breast cancer, and 0.61% (vs 0.78%) for leukemia.

In older adults, CRC incidence rates only increased in about half the countries (median AAPC, 0.37%), and the annual percentage change was much greater in younger than older adults in nearly 70% of countries. CRC incidence rates in older individuals also decreased in nearly 25% of countries.

Why is CRC an apparent outlier?

“Bowel cancer screening not only helps detect cancer at earlier stages but also helps prevent cancer through the removal of premalignant lesions,” Berrington de González said. “This could be why bowel cancer cases seem to be rising faster in younger adults — we’re getting better at preventing them developing in older adults.”

The incidence of certain cancers also declined in younger adults. Specifically, rates of liver, oral, esophageal, and stomach cancers decreased in younger adults in more than half of countries assessed, with median AAPCs of -0.14% for liver, -0.42% for oral, -0.92% for esophageal, and -1.62% for stomach cancers.

Over half of countries also saw declining rates of stomach (median AAPC, -2.05%) and esophageal (median AAPC, -0.25%) cancers among older adults, while rates of liver and oral cancers increased in older individuals (median AAPC, 2.17% and 0.49%, respectively).

For gallbladder, pancreatic, and prostate cancers — three other cancers previously found to be increasing in younger adults — the researchers reported that incidence rates increased in younger adults in just over half of countries (median AAPCs, 3.2% for prostate cancer, 0.49% for gallbladder cancer, and 1% for pancreatic cancer). Incidence rates also often increased in older adults but to a lesser extent (median AAPCs, 0.75% for prostate cancer, -0.10% for gallbladder, and 0.96% for pancreatic cancer).

 

True Rise or Increased Scrutiny?

Why are cancer rates increasing?

“Understanding factors that contribute to the increase in incidence across the age spectrum was beyond the scope of the study,” editorialists Christopher Cann, MD, Fox Chase Cancer Center, and Efrat Dotan, MD, University of Pennsylvania Health System, both in Philadelphia, wrote.

Several studies have suggested that rising rates of obesity could help explain increasing cancer incidence, particularly in younger adults. In fact, “the cancers that we identified as increasing are all obesity-related cancers, including endometrial and kidney cancer,” Berrington de González said. However, so far, the evidence on this link remains unclear, she acknowledged.

Weighing in on the study, Gilbert Welch, MD, Brigham and Women’s Hospital, Boston, told this news organization that it’s “critically important” to distinguish between two explanations for rising cancer incidence.

There may be an increase in the true occurrence of clinically meaningful cancer, which “warrants investigation into biologic explanations, better treatment, and perhaps more testing,” Welch said.

But it may instead reflect changes in diagnostic scrutiny. “Simply put, whenever we doctors look harder for cancer, we find more,” Welch said. “And there are lots of ways to look harder: testing more people, testing people more frequently, using tests with increasing ability to detect small irregularities, and using lower diagnostic thresholds for labeling these as cancer.”

If increased incidence is the result of greater diagnostic scrutiny, searching for biologic causes is bound to be unproductive and more testing will only aggravate the problem, he explained.

Welch pointed out that the fastest rising cancer in both younger and older adults was thyroid cancer (AAPC, ≥ 3%), which is “exquisitely sensitive” to diagnostic scrutiny.

Take what happened in South Korea. Around 2000, the government of South Korea started a national screening program for breast, colon, and stomach cancers. Doctors and hospitals often added on ultrasound scans for thyroid cancer for a small additional fee.

“A decade later the rate of thyroid cancer diagnosis had increased 15-fold, turning what was once a rare cancer into the most common cancer in Korea,” Welch said. “But the death rate from thyroid cancer did not change. This was not an epidemic of disease; this was an epidemic of diagnosis.”

Welch also noted that the study authors and editorialists put the finding in perspective by explaining that, despite the rising rates of certain cancers in younger adults, cancer remains rare in these adults.

Welch highlighted that, for younger adults in the US, cancer death rates in young adults have cut in half over the last 30 years. “Cancer accounts for only 10% of deaths in young people in the US — and that number is falling,” Welch said.

The study was funded by the Institute of Cancer Research and the National Institutes of Health Intramural Research Program. Disclosures for authors and editorial writers are available with the original articles. Welch reported receiving royalties from three books including “Should I be tested for cancer?”

A version of this article first appeared on Medscape.com.

TOPLINE: Large Language Models (LLMs) achieve more than 95% accuracy in extracting colorectal cancer and dysplasia diagnoses from Veterans Health Administration (VHA) pathology reports, including patients with Million Veteran Program (MVP) genomic data. The validated approach using publicly available LLMs demonstrates excellent performance across both Inflammatory Bowel Disease (IBD) and non-IBD populations.

METHODOLOGY: 

Researchers analyzed 116,373 pathology reports generated in the VHA between 1999 and 2024, utilizing search term filtering followed by simple yes/no question prompts for identifying colorectal dysplasia, high-grade dysplasia and/or colorectal adenocarcinoma, and invasive colorectal cancer.

• Results were compared to blinded manual chart review of 200 to 300 pathology reports for each patient cohort and diagnostic task, totaling 3,816 reviewed reports, to validate the LLM approach.

• Validation was performed independently in IBD and non-IBD populations using Gemma-2 and Llama-3 LLMs without any task-specific training or fine-tuning.

• Performance metrics included F1 scores, positive predictive value, negative predictive value, sensitivity, specificity, and Matthew's correlation coefficient to evaluate accuracy across different tasks.

TAKEAWAY:

• In patients with IBD in the MVP, the LLM achieved (F1-score, 96.9%; 95% confidence interval [CI], 94.0%-99.6%) for identifying dysplasia, (F1-score, 93.7%; 95% CI, 88.2%-98.4%) for identifying high-grade dysplasia/colorectal cancer, and (F1-score, 98%; 95% CI, 96.3%-99.4%) for identifying colorectal cancer.

• In non-IBD MVP patients, the LLM demonstrated (F1-score, 99.2%; 95% CI, 98.2%-100%) for identifying colorectal dysplasia, (F1-score, 96.5%; 95% CI, 93.0%-99.2%) for high-grade dysplasia/colorectal cancer, and (F1-score, 95%; 95% CI, 92.8%-97.2%) for identifying colorectal cancer.

• Agreement between reviewers was excellent across tasks, with (Cohen's kappa, 89%-97%) for main tasks, and (Cohen's kappa, 78.1%-93.1%) for indefinite for dysplasia in IBD cohort.

• The LLM approach maintained high accuracy when applied to full pathology reports, with (F1-score, 97.1%; 95% CI, 93.5%-100%) for dysplasia detection in IBD patients.

IN PRACTICE: “We have shown that LLMs are powerful, potentially generalizable tools for accurately extracting important information from clinical semistructured and unstructured text and which require little human-led development.” the authors of the study wrote

SOURCE: The study was based on data from the Million Veteran Program and supported by the Office of Research and Development, Veterans Health Administration, and the US Department of Veterans Affairs Biomedical Laboratory. It was published online in BMJ Open Gastroenterology.

LIMITATIONS:  According to the authors, this research may be specific to the VHA system and the LLM models used. The authors did not test larger models. The authors acknowledge that without long-term access to graphics processing units, they could not feasibly test larger models, which may overcome some of the shortcomings seen in smaller models. Additionally, the researchers could not rule out overlap between Million Veteran Program and Corporate Data Warehouse reports, though they state that results in either cohort alone are sufficient validation compared with previously published work.

DISCLOSURES: The study was supported by Merit Review Award from the United States Department of Veterans Affairs Biomedical Laboratory Research and Development Service, AGA Research Foundation, National Institutes of Health grants, and the National Library of Medicine Training Grant. Kit Curtius reported receiving an investigator-led research grant from Phathom Pharmaceuticals. Shailja C Shah disclosed being a paid consultant for RedHill Biopharma and Phathom Pharmaceuticals, and an unpaid scientific advisory board member for Ilico Genetics, Inc.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.

TOPLINE: Large Language Models (LLMs) achieve more than 95% accuracy in extracting colorectal cancer and dysplasia diagnoses from Veterans Health Administration (VHA) pathology reports, including patients with Million Veteran Program (MVP) genomic data. The validated approach using publicly available LLMs demonstrates excellent performance across both Inflammatory Bowel Disease (IBD) and non-IBD populations.

METHODOLOGY: 

Researchers analyzed 116,373 pathology reports generated in the VHA between 1999 and 2024, utilizing search term filtering followed by simple yes/no question prompts for identifying colorectal dysplasia, high-grade dysplasia and/or colorectal adenocarcinoma, and invasive colorectal cancer.

• Results were compared to blinded manual chart review of 200 to 300 pathology reports for each patient cohort and diagnostic task, totaling 3,816 reviewed reports, to validate the LLM approach.

• Validation was performed independently in IBD and non-IBD populations using Gemma-2 and Llama-3 LLMs without any task-specific training or fine-tuning.

• Performance metrics included F1 scores, positive predictive value, negative predictive value, sensitivity, specificity, and Matthew's correlation coefficient to evaluate accuracy across different tasks.

TAKEAWAY:

• In patients with IBD in the MVP, the LLM achieved (F1-score, 96.9%; 95% confidence interval [CI], 94.0%-99.6%) for identifying dysplasia, (F1-score, 93.7%; 95% CI, 88.2%-98.4%) for identifying high-grade dysplasia/colorectal cancer, and (F1-score, 98%; 95% CI, 96.3%-99.4%) for identifying colorectal cancer.

• In non-IBD MVP patients, the LLM demonstrated (F1-score, 99.2%; 95% CI, 98.2%-100%) for identifying colorectal dysplasia, (F1-score, 96.5%; 95% CI, 93.0%-99.2%) for high-grade dysplasia/colorectal cancer, and (F1-score, 95%; 95% CI, 92.8%-97.2%) for identifying colorectal cancer.

• Agreement between reviewers was excellent across tasks, with (Cohen's kappa, 89%-97%) for main tasks, and (Cohen's kappa, 78.1%-93.1%) for indefinite for dysplasia in IBD cohort.

• The LLM approach maintained high accuracy when applied to full pathology reports, with (F1-score, 97.1%; 95% CI, 93.5%-100%) for dysplasia detection in IBD patients.

IN PRACTICE: “We have shown that LLMs are powerful, potentially generalizable tools for accurately extracting important information from clinical semistructured and unstructured text and which require little human-led development.” the authors of the study wrote

SOURCE: The study was based on data from the Million Veteran Program and supported by the Office of Research and Development, Veterans Health Administration, and the US Department of Veterans Affairs Biomedical Laboratory. It was published online in BMJ Open Gastroenterology.

LIMITATIONS:  According to the authors, this research may be specific to the VHA system and the LLM models used. The authors did not test larger models. The authors acknowledge that without long-term access to graphics processing units, they could not feasibly test larger models, which may overcome some of the shortcomings seen in smaller models. Additionally, the researchers could not rule out overlap between Million Veteran Program and Corporate Data Warehouse reports, though they state that results in either cohort alone are sufficient validation compared with previously published work.

DISCLOSURES: The study was supported by Merit Review Award from the United States Department of Veterans Affairs Biomedical Laboratory Research and Development Service, AGA Research Foundation, National Institutes of Health grants, and the National Library of Medicine Training Grant. Kit Curtius reported receiving an investigator-led research grant from Phathom Pharmaceuticals. Shailja C Shah disclosed being a paid consultant for RedHill Biopharma and Phathom Pharmaceuticals, and an unpaid scientific advisory board member for Ilico Genetics, Inc.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.

TOPLINE: Large Language Models (LLMs) achieve more than 95% accuracy in extracting colorectal cancer and dysplasia diagnoses from Veterans Health Administration (VHA) pathology reports, including patients with Million Veteran Program (MVP) genomic data. The validated approach using publicly available LLMs demonstrates excellent performance across both Inflammatory Bowel Disease (IBD) and non-IBD populations.

METHODOLOGY: 

Researchers analyzed 116,373 pathology reports generated in the VHA between 1999 and 2024, utilizing search term filtering followed by simple yes/no question prompts for identifying colorectal dysplasia, high-grade dysplasia and/or colorectal adenocarcinoma, and invasive colorectal cancer.

• Results were compared to blinded manual chart review of 200 to 300 pathology reports for each patient cohort and diagnostic task, totaling 3,816 reviewed reports, to validate the LLM approach.

• Validation was performed independently in IBD and non-IBD populations using Gemma-2 and Llama-3 LLMs without any task-specific training or fine-tuning.

• Performance metrics included F1 scores, positive predictive value, negative predictive value, sensitivity, specificity, and Matthew's correlation coefficient to evaluate accuracy across different tasks.

TAKEAWAY:

• In patients with IBD in the MVP, the LLM achieved (F1-score, 96.9%; 95% confidence interval [CI], 94.0%-99.6%) for identifying dysplasia, (F1-score, 93.7%; 95% CI, 88.2%-98.4%) for identifying high-grade dysplasia/colorectal cancer, and (F1-score, 98%; 95% CI, 96.3%-99.4%) for identifying colorectal cancer.

• In non-IBD MVP patients, the LLM demonstrated (F1-score, 99.2%; 95% CI, 98.2%-100%) for identifying colorectal dysplasia, (F1-score, 96.5%; 95% CI, 93.0%-99.2%) for high-grade dysplasia/colorectal cancer, and (F1-score, 95%; 95% CI, 92.8%-97.2%) for identifying colorectal cancer.

• Agreement between reviewers was excellent across tasks, with (Cohen's kappa, 89%-97%) for main tasks, and (Cohen's kappa, 78.1%-93.1%) for indefinite for dysplasia in IBD cohort.

• The LLM approach maintained high accuracy when applied to full pathology reports, with (F1-score, 97.1%; 95% CI, 93.5%-100%) for dysplasia detection in IBD patients.

IN PRACTICE: “We have shown that LLMs are powerful, potentially generalizable tools for accurately extracting important information from clinical semistructured and unstructured text and which require little human-led development.” the authors of the study wrote

SOURCE: The study was based on data from the Million Veteran Program and supported by the Office of Research and Development, Veterans Health Administration, and the US Department of Veterans Affairs Biomedical Laboratory. It was published online in BMJ Open Gastroenterology.

LIMITATIONS:  According to the authors, this research may be specific to the VHA system and the LLM models used. The authors did not test larger models. The authors acknowledge that without long-term access to graphics processing units, they could not feasibly test larger models, which may overcome some of the shortcomings seen in smaller models. Additionally, the researchers could not rule out overlap between Million Veteran Program and Corporate Data Warehouse reports, though they state that results in either cohort alone are sufficient validation compared with previously published work.

DISCLOSURES: The study was supported by Merit Review Award from the United States Department of Veterans Affairs Biomedical Laboratory Research and Development Service, AGA Research Foundation, National Institutes of Health grants, and the National Library of Medicine Training Grant. Kit Curtius reported receiving an investigator-led research grant from Phathom Pharmaceuticals. Shailja C Shah disclosed being a paid consultant for RedHill Biopharma and Phathom Pharmaceuticals, and an unpaid scientific advisory board member for Ilico Genetics, Inc.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.

PHOENIX — Patients with head and neck cancer face high rates of malnutrition during treatment, and oral supplements are often recommended. But they are not the entire answer, a dietician told colleagues at the Association of Veterans Affairs (VA) Hematology/Oncology annual meeting.

“Patients should consume the most liberal diet possible throughout treatment,” said advanced practice oncology dietician Brittany Leneweaver, RD, CSO, CES, at the VA Washington DC Healthcare System. “This means not solely relying on oral nutrition supplements like Ensure if possible.”

While Leneweaver said many patients will need supplements, she stressed these products “are meant to supplement the diet and not be the sole source of nutrition, ideally.” Encouraging the intake of whole foods “is really key to make the transition back to solid foods after they’re done with treatment. This makes it so much easier when they’re already swallowing those thicker textures, rather than just liquid the entire time.”

Malnutrition: Common and Damaging

As Leneweaver noted, malnutrition is common in patients with head and neck cancer, and can lead to “increased treatment toxicity, increased risk of infection, decreased survival, increased surgical complication, delayed healing, decreased physical function, and decreased quality of life.”

Malnutrition data in patients with head and neck cancer in the US is sparse. However, a 2024 study found malnutrition in 20% of patients undergoing head and neck cancer surgery and linked the condition to increased length of stay (β, 5.20 additional days), higher costs (β, $15,722) higher odds of potentially preventable complications (adjusted odds ratio [aOR], 2.04), and lower odds of discharge to home (aOR, 0.34).

Leneweaver said her role involves addressing “nutrition impact symptoms” that reduce veteran food intake such as difficulty swallowing, taste disorders, dry mouth, and inflammation of the mucus membranes.

“I can’t tell you how much time I spend just talking to the patient about their medication regimens, making sure they have antiemetics on board, letting the radiation oncologist know, ‘Hey, it’s probably time for medicine,’” she said. “We’re constantly looking at side effects and addressing to alert the team as quickly as possible so that we can prevent further weight loss.”

Better Diets Lead to Better Outcomes

Leneweaver noted that “many times, patients will continue to rely on oral supplements as their primary source of nutrition over the long term. They may be missing out on several health benefits as a result.”

Research shows that high-quality diets matter in this patient group, she said. They’re associated with “decreased symptoms during treatment, reduced head and neck cancer risk, and reduced risk of those chronic nutrition impact symptoms,” Leneweaver said.

Diets before and after cancer diagnosis can make a difference. A 2019 study examined patient diets prior to diagnosis of head and neck cancer. It found that patients with better diet quality were less likely to experience overall nutrition impact symptoms (OR 0.45). However, “studies have found that the majority of our patients with head and neck cancer have an inadequate diet prior to diagnosis,” Leneweaver said. 

As for postdiagnosis nutrition, a 2022 study linked healthier diets in patients with head and neck cancer to 93% lower 3-year risk of all-cause mortality and 85% lower risk of cancer-specific mortality. 

What’s in a High-Quality Diet?

Regarding specific food recommendations, Leneweaver prefers the American Institute for Cancer Research (AICR) nutrition guidelines over the US Department of Agriculture’s Dietary Guidelines for Americans. The AICR “more clearly recommends plant-based diet with at least two-thirds of each meal coming from a variety of plant sources” and recommends avoiding alcohol entirely and limiting red meat, she said. 

Leneweaver said she recognizes that dietary change can be gradual.

“It’s not going to happen overnight,” she said. “We know that lifestyle change takes a lot of work.”

Basic interventions can be effective, she said: “This can be just as simple as recommending a plant-based diet to your patient or recommending they eat the rainbow. And I don’t mean Skittles, I mean actual plants. If you just mention these couple of things to the patients, this can really go a long way, especially if they’re hearing that consistent messaging.”

Team-Based Follow-Up Is Key

Leneweaver emphasized the importance of following up over time even if patients do not initially accept referrals to nutritional services. Dieticians ideally see patients before or during initial treatment and then weekly during radiation therapy. Posttreatment follow-up continues “until they’re nutritionally stable. This can be anywhere from weekly to monthly.”

Leneweaver emphasized collaborating with other team members. For example, she works with a speech pathologist at joint visits, either weekly or monthly, “so that they can get off of that feeding tube or get back to a solid consistency diet, typically before that 3-month PET scan.”

It is also important to understand barriers to healthy eating in the veteran population, including transportation challenges and poor access to healthy food, Leneweaver said.

“Make sure you’re utilizing your social worker, your psychologist, other resources, and food pantries, if you have them.” 

Even when the most ideal choices are not available, she said, “if they only have access to canned vegetables, I’d much rather them eat that than have nothing.”

No disclosures for Leneweaver were provided. 

PHOENIX — Patients with head and neck cancer face high rates of malnutrition during treatment, and oral supplements are often recommended. But they are not the entire answer, a dietician told colleagues at the Association of Veterans Affairs (VA) Hematology/Oncology annual meeting.

“Patients should consume the most liberal diet possible throughout treatment,” said advanced practice oncology dietician Brittany Leneweaver, RD, CSO, CES, at the VA Washington DC Healthcare System. “This means not solely relying on oral nutrition supplements like Ensure if possible.”

While Leneweaver said many patients will need supplements, she stressed these products “are meant to supplement the diet and not be the sole source of nutrition, ideally.” Encouraging the intake of whole foods “is really key to make the transition back to solid foods after they’re done with treatment. This makes it so much easier when they’re already swallowing those thicker textures, rather than just liquid the entire time.”

Malnutrition: Common and Damaging

As Leneweaver noted, malnutrition is common in patients with head and neck cancer, and can lead to “increased treatment toxicity, increased risk of infection, decreased survival, increased surgical complication, delayed healing, decreased physical function, and decreased quality of life.”

Malnutrition data in patients with head and neck cancer in the US is sparse. However, a 2024 study found malnutrition in 20% of patients undergoing head and neck cancer surgery and linked the condition to increased length of stay (β, 5.20 additional days), higher costs (β, $15,722) higher odds of potentially preventable complications (adjusted odds ratio [aOR], 2.04), and lower odds of discharge to home (aOR, 0.34).

Leneweaver said her role involves addressing “nutrition impact symptoms” that reduce veteran food intake such as difficulty swallowing, taste disorders, dry mouth, and inflammation of the mucus membranes.

“I can’t tell you how much time I spend just talking to the patient about their medication regimens, making sure they have antiemetics on board, letting the radiation oncologist know, ‘Hey, it’s probably time for medicine,’” she said. “We’re constantly looking at side effects and addressing to alert the team as quickly as possible so that we can prevent further weight loss.”

Better Diets Lead to Better Outcomes

Leneweaver noted that “many times, patients will continue to rely on oral supplements as their primary source of nutrition over the long term. They may be missing out on several health benefits as a result.”

Research shows that high-quality diets matter in this patient group, she said. They’re associated with “decreased symptoms during treatment, reduced head and neck cancer risk, and reduced risk of those chronic nutrition impact symptoms,” Leneweaver said.

Diets before and after cancer diagnosis can make a difference. A 2019 study examined patient diets prior to diagnosis of head and neck cancer. It found that patients with better diet quality were less likely to experience overall nutrition impact symptoms (OR 0.45). However, “studies have found that the majority of our patients with head and neck cancer have an inadequate diet prior to diagnosis,” Leneweaver said. 

As for postdiagnosis nutrition, a 2022 study linked healthier diets in patients with head and neck cancer to 93% lower 3-year risk of all-cause mortality and 85% lower risk of cancer-specific mortality. 

What’s in a High-Quality Diet?

Regarding specific food recommendations, Leneweaver prefers the American Institute for Cancer Research (AICR) nutrition guidelines over the US Department of Agriculture’s Dietary Guidelines for Americans. The AICR “more clearly recommends plant-based diet with at least two-thirds of each meal coming from a variety of plant sources” and recommends avoiding alcohol entirely and limiting red meat, she said. 

Leneweaver said she recognizes that dietary change can be gradual.

“It’s not going to happen overnight,” she said. “We know that lifestyle change takes a lot of work.”

Basic interventions can be effective, she said: “This can be just as simple as recommending a plant-based diet to your patient or recommending they eat the rainbow. And I don’t mean Skittles, I mean actual plants. If you just mention these couple of things to the patients, this can really go a long way, especially if they’re hearing that consistent messaging.”

Team-Based Follow-Up Is Key

Leneweaver emphasized the importance of following up over time even if patients do not initially accept referrals to nutritional services. Dieticians ideally see patients before or during initial treatment and then weekly during radiation therapy. Posttreatment follow-up continues “until they’re nutritionally stable. This can be anywhere from weekly to monthly.”

Leneweaver emphasized collaborating with other team members. For example, she works with a speech pathologist at joint visits, either weekly or monthly, “so that they can get off of that feeding tube or get back to a solid consistency diet, typically before that 3-month PET scan.”

It is also important to understand barriers to healthy eating in the veteran population, including transportation challenges and poor access to healthy food, Leneweaver said.

“Make sure you’re utilizing your social worker, your psychologist, other resources, and food pantries, if you have them.” 

Even when the most ideal choices are not available, she said, “if they only have access to canned vegetables, I’d much rather them eat that than have nothing.”

No disclosures for Leneweaver were provided. 

PHOENIX — Patients with head and neck cancer face high rates of malnutrition during treatment, and oral supplements are often recommended. But they are not the entire answer, a dietician told colleagues at the Association of Veterans Affairs (VA) Hematology/Oncology annual meeting.

“Patients should consume the most liberal diet possible throughout treatment,” said advanced practice oncology dietician Brittany Leneweaver, RD, CSO, CES, at the VA Washington DC Healthcare System. “This means not solely relying on oral nutrition supplements like Ensure if possible.”

While Leneweaver said many patients will need supplements, she stressed these products “are meant to supplement the diet and not be the sole source of nutrition, ideally.” Encouraging the intake of whole foods “is really key to make the transition back to solid foods after they’re done with treatment. This makes it so much easier when they’re already swallowing those thicker textures, rather than just liquid the entire time.”

Malnutrition: Common and Damaging

As Leneweaver noted, malnutrition is common in patients with head and neck cancer, and can lead to “increased treatment toxicity, increased risk of infection, decreased survival, increased surgical complication, delayed healing, decreased physical function, and decreased quality of life.”

Malnutrition data in patients with head and neck cancer in the US is sparse. However, a 2024 study found malnutrition in 20% of patients undergoing head and neck cancer surgery and linked the condition to increased length of stay (β, 5.20 additional days), higher costs (β, $15,722) higher odds of potentially preventable complications (adjusted odds ratio [aOR], 2.04), and lower odds of discharge to home (aOR, 0.34).

Leneweaver said her role involves addressing “nutrition impact symptoms” that reduce veteran food intake such as difficulty swallowing, taste disorders, dry mouth, and inflammation of the mucus membranes.

“I can’t tell you how much time I spend just talking to the patient about their medication regimens, making sure they have antiemetics on board, letting the radiation oncologist know, ‘Hey, it’s probably time for medicine,’” she said. “We’re constantly looking at side effects and addressing to alert the team as quickly as possible so that we can prevent further weight loss.”

Better Diets Lead to Better Outcomes

Leneweaver noted that “many times, patients will continue to rely on oral supplements as their primary source of nutrition over the long term. They may be missing out on several health benefits as a result.”

Research shows that high-quality diets matter in this patient group, she said. They’re associated with “decreased symptoms during treatment, reduced head and neck cancer risk, and reduced risk of those chronic nutrition impact symptoms,” Leneweaver said.

Diets before and after cancer diagnosis can make a difference. A 2019 study examined patient diets prior to diagnosis of head and neck cancer. It found that patients with better diet quality were less likely to experience overall nutrition impact symptoms (OR 0.45). However, “studies have found that the majority of our patients with head and neck cancer have an inadequate diet prior to diagnosis,” Leneweaver said. 

As for postdiagnosis nutrition, a 2022 study linked healthier diets in patients with head and neck cancer to 93% lower 3-year risk of all-cause mortality and 85% lower risk of cancer-specific mortality. 

What’s in a High-Quality Diet?

Regarding specific food recommendations, Leneweaver prefers the American Institute for Cancer Research (AICR) nutrition guidelines over the US Department of Agriculture’s Dietary Guidelines for Americans. The AICR “more clearly recommends plant-based diet with at least two-thirds of each meal coming from a variety of plant sources” and recommends avoiding alcohol entirely and limiting red meat, she said. 

Leneweaver said she recognizes that dietary change can be gradual.

“It’s not going to happen overnight,” she said. “We know that lifestyle change takes a lot of work.”

Basic interventions can be effective, she said: “This can be just as simple as recommending a plant-based diet to your patient or recommending they eat the rainbow. And I don’t mean Skittles, I mean actual plants. If you just mention these couple of things to the patients, this can really go a long way, especially if they’re hearing that consistent messaging.”

Team-Based Follow-Up Is Key

Leneweaver emphasized the importance of following up over time even if patients do not initially accept referrals to nutritional services. Dieticians ideally see patients before or during initial treatment and then weekly during radiation therapy. Posttreatment follow-up continues “until they’re nutritionally stable. This can be anywhere from weekly to monthly.”

Leneweaver emphasized collaborating with other team members. For example, she works with a speech pathologist at joint visits, either weekly or monthly, “so that they can get off of that feeding tube or get back to a solid consistency diet, typically before that 3-month PET scan.”

It is also important to understand barriers to healthy eating in the veteran population, including transportation challenges and poor access to healthy food, Leneweaver said.

“Make sure you’re utilizing your social worker, your psychologist, other resources, and food pantries, if you have them.” 

Even when the most ideal choices are not available, she said, “if they only have access to canned vegetables, I’d much rather them eat that than have nothing.”

No disclosures for Leneweaver were provided. 

TOPLINE:

A comprehensive review of 92 studies found that 15% to 20% of lung cancers occurred among nonsmokers and were associated with environmental and germline risk factors. These cancers frequently harbored actionable genomic drivers, and targeted EGFR and ALK therapies produced significant diseasefree survival (DFS) and overall survival benefits.

METHODOLOGY:

• Lung cancer continues to be the leading cause of cancer death worldwide, causing about 1.8 million deaths in 2022, with smoking remaining the predominant risk factor. However, the incidence of lung cancer among nonsmokers (those who have smoked less than 100 cigarettes in their lifetime) is rising, varies by sex and geography, and is linked to environmental exposures and family history. The misperception that lung cancer is almost invariably caused by smoking may delay assessment and diagnosis.
• Researchers conducted a review of 92 studies on lung cancer in nonsmokers: 6 meta-analyses or systematic reviews, 16 randomized clinical trials, eight prospective cohort studies, seven retrospective cohort studies, three cross-sectional studies, four observational or case-control studies, 13 genomic studies, and 35 other studies.
• Overall, lung cancer among nonsmokers accounted for 15% to 20% of all lung cancer cases. Most lung cancers in nonsmokers were adenocarcinomas (60% to 80%), with a median age at diagnosis of 67 years in this group compared with 70 years in people with a history of smoking.
• Data analysis from three US hospital networks showed that the proportion of lung cancer among nonsmokers increased from 8.0% to 14.9% between 1990 and 2013. A pooled analysis of seven Finnish cohorts reported an absolute increase in lung cancer among nonsmokers from 6.9 per 100,000 person-years in 1972 to 12.9 per 100,000 person-years in 2015.
• The age-adjusted incidence rate of lung cancer in the US between 2000 and 2013 was 17.5 per 100,000 individuals among Asian female nonsmokers compared with 10.1 per 100,000 among non-Hispanic White female nonsmokers.

TAKEAWAY:

• Environmental and occupational risk factors were secondhand smoke, residential radon, outdoor and household air pollution (PM_2.5), asbestos and silica exposure, and prior thoracic radiotherapy. Having a first-degree relative with lung cancer increased the risk of developing lung cancer, and genome-wide association studies identified susceptibility loci associated with lung cancer risk in nonsmokers.
• Family history and inherited susceptibility increased lung cancer risk in never smokers (odds ratio [OR] for lung cancer in those with a first–degree relative, 1.51), and clonal hematopoiesis was also associated with higher risk (OR, 1.43). Importantly, tumors in nonsmokers were frequently driven by actionable somatic alterations (EGFR mutations, 40% to 60% in nonsmokers compared with 10% in smokers) and enrichment of ALK/ROS1/RET/ERBB2/NTRK/NRG1 fusions; 78% to 92% of adenocarcinomas in nonsmokers harbored actionable drivers (compared with 49.5% in ever smokers), and nonsmokers had a substantially lower tumor mutational burden (10–fold lower).
• Similar to individuals with a history of smoking, nonsmokers with lung cancer presented with cough, pain, dyspnea, or weight loss or had disease detected incidentally. Surgical resection remained the preferred treatment for anatomically resectable lung cancer (stages I-III) in medically eligible patients, with follow-up CT screening recommended every 6 months for 2 to 3 years and then annually.
• Targeted adjuvant therapy substantially improved outcomes for resected EGFR–mutant or ALK–rearranged non-small cell lung cancer (NSCLC). Four-year DFS was increased to 70% with osimertinib compared with 29% with placebo (hazard ratio [HR], 0.23) and 5–year overall survival was increased to 85% compared with 73% (HR, 0.49). Two–year DFS was 93.8% with alectinib compared with 63% with placebo (HR, 0.24). In unresectable EGFR-mutated stage III NSCLC, median progression-free survival was 39.1 months with adjuvant osimertinib compared with 5.6 months with placebo. For resected ALKpositive disease, 2–year DFS was 93.8% with adjuvant alectinib compared with 63.0% with chemotherapy (HR, 0.24).
• However, singleagent single agent programmed cell death protein 1 inhibitors or programmed death-ligand 1 inhibitors demonstrated limited efficacy in EGFR or ALK–driven tumors, and benefit was attenuated in never smokers. Regarding screening and early detection, the US Preventive Services Task Force did not recommend lowdose CT screening for nonsmokers, whereas Taiwan implemented a biennial screening program for selected nonsmoking high–risk groups.

IN PRACTICE:

“Among patients with lung cancer, nonsmoking individuals are more likely to have genomic alterations, such as EGFR mutations or ALK gene rearrangements, and these patients have improved survival when treated with TKIs compared with chemotherapy,” the authors of the study wrote.

SOURCE:

The study, led by Cian Murphy, PhD, Cancer Evolution and Genome Instability Laboratory, Francis Crick Institute, London, England, was published online in JAMA.

LIMITATIONS:

Becausesmoking history was often not included in many databases, cancer registries, and trials, the incidence and prevalence of lung cancer in nonsmokers could not be accurately determined. Additionally, accurate quantification of environmental exposures, such as air pollution, presented significant challenges. The quality of the evidence was not formally evaluated, and some relevant articles may have been missed in the literature review.

DISCLOSURES:

The study received support from multiple organizations, including the Rosetrees Trust, Ruth Strauss Foundation, Cancer Research UK, and the National Health and Medical Research Council. Several authors reported receiving grants or personal fees from and having other ties with various sources. Full disclosures are noted in the original article.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.

A version of this article first appeared on Medscape.com.

TOPLINE:

A comprehensive review of 92 studies found that 15% to 20% of lung cancers occurred among nonsmokers and were associated with environmental and germline risk factors. These cancers frequently harbored actionable genomic drivers, and targeted EGFR and ALK therapies produced significant diseasefree survival (DFS) and overall survival benefits.

METHODOLOGY:

• Lung cancer continues to be the leading cause of cancer death worldwide, causing about 1.8 million deaths in 2022, with smoking remaining the predominant risk factor. However, the incidence of lung cancer among nonsmokers (those who have smoked less than 100 cigarettes in their lifetime) is rising, varies by sex and geography, and is linked to environmental exposures and family history. The misperception that lung cancer is almost invariably caused by smoking may delay assessment and diagnosis.
• Researchers conducted a review of 92 studies on lung cancer in nonsmokers: 6 meta-analyses or systematic reviews, 16 randomized clinical trials, eight prospective cohort studies, seven retrospective cohort studies, three cross-sectional studies, four observational or case-control studies, 13 genomic studies, and 35 other studies.
• Overall, lung cancer among nonsmokers accounted for 15% to 20% of all lung cancer cases. Most lung cancers in nonsmokers were adenocarcinomas (60% to 80%), with a median age at diagnosis of 67 years in this group compared with 70 years in people with a history of smoking.
• Data analysis from three US hospital networks showed that the proportion of lung cancer among nonsmokers increased from 8.0% to 14.9% between 1990 and 2013. A pooled analysis of seven Finnish cohorts reported an absolute increase in lung cancer among nonsmokers from 6.9 per 100,000 person-years in 1972 to 12.9 per 100,000 person-years in 2015.
• The age-adjusted incidence rate of lung cancer in the US between 2000 and 2013 was 17.5 per 100,000 individuals among Asian female nonsmokers compared with 10.1 per 100,000 among non-Hispanic White female nonsmokers.

TAKEAWAY:

• Environmental and occupational risk factors were secondhand smoke, residential radon, outdoor and household air pollution (PM_2.5), asbestos and silica exposure, and prior thoracic radiotherapy. Having a first-degree relative with lung cancer increased the risk of developing lung cancer, and genome-wide association studies identified susceptibility loci associated with lung cancer risk in nonsmokers.
• Family history and inherited susceptibility increased lung cancer risk in never smokers (odds ratio [OR] for lung cancer in those with a first–degree relative, 1.51), and clonal hematopoiesis was also associated with higher risk (OR, 1.43). Importantly, tumors in nonsmokers were frequently driven by actionable somatic alterations (EGFR mutations, 40% to 60% in nonsmokers compared with 10% in smokers) and enrichment of ALK/ROS1/RET/ERBB2/NTRK/NRG1 fusions; 78% to 92% of adenocarcinomas in nonsmokers harbored actionable drivers (compared with 49.5% in ever smokers), and nonsmokers had a substantially lower tumor mutational burden (10–fold lower).
• Similar to individuals with a history of smoking, nonsmokers with lung cancer presented with cough, pain, dyspnea, or weight loss or had disease detected incidentally. Surgical resection remained the preferred treatment for anatomically resectable lung cancer (stages I-III) in medically eligible patients, with follow-up CT screening recommended every 6 months for 2 to 3 years and then annually.
• Targeted adjuvant therapy substantially improved outcomes for resected EGFR–mutant or ALK–rearranged non-small cell lung cancer (NSCLC). Four-year DFS was increased to 70% with osimertinib compared with 29% with placebo (hazard ratio [HR], 0.23) and 5–year overall survival was increased to 85% compared with 73% (HR, 0.49). Two–year DFS was 93.8% with alectinib compared with 63% with placebo (HR, 0.24). In unresectable EGFR-mutated stage III NSCLC, median progression-free survival was 39.1 months with adjuvant osimertinib compared with 5.6 months with placebo. For resected ALKpositive disease, 2–year DFS was 93.8% with adjuvant alectinib compared with 63.0% with chemotherapy (HR, 0.24).
• However, singleagent single agent programmed cell death protein 1 inhibitors or programmed death-ligand 1 inhibitors demonstrated limited efficacy in EGFR or ALK–driven tumors, and benefit was attenuated in never smokers. Regarding screening and early detection, the US Preventive Services Task Force did not recommend lowdose CT screening for nonsmokers, whereas Taiwan implemented a biennial screening program for selected nonsmoking high–risk groups.

IN PRACTICE:

“Among patients with lung cancer, nonsmoking individuals are more likely to have genomic alterations, such as EGFR mutations or ALK gene rearrangements, and these patients have improved survival when treated with TKIs compared with chemotherapy,” the authors of the study wrote.

SOURCE:

The study, led by Cian Murphy, PhD, Cancer Evolution and Genome Instability Laboratory, Francis Crick Institute, London, England, was published online in JAMA.

LIMITATIONS:

Becausesmoking history was often not included in many databases, cancer registries, and trials, the incidence and prevalence of lung cancer in nonsmokers could not be accurately determined. Additionally, accurate quantification of environmental exposures, such as air pollution, presented significant challenges. The quality of the evidence was not formally evaluated, and some relevant articles may have been missed in the literature review.

DISCLOSURES:

The study received support from multiple organizations, including the Rosetrees Trust, Ruth Strauss Foundation, Cancer Research UK, and the National Health and Medical Research Council. Several authors reported receiving grants or personal fees from and having other ties with various sources. Full disclosures are noted in the original article.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.

A version of this article first appeared on Medscape.com.

TOPLINE:

A comprehensive review of 92 studies found that 15% to 20% of lung cancers occurred among nonsmokers and were associated with environmental and germline risk factors. These cancers frequently harbored actionable genomic drivers, and targeted EGFR and ALK therapies produced significant diseasefree survival (DFS) and overall survival benefits.

METHODOLOGY:

• Lung cancer continues to be the leading cause of cancer death worldwide, causing about 1.8 million deaths in 2022, with smoking remaining the predominant risk factor. However, the incidence of lung cancer among nonsmokers (those who have smoked less than 100 cigarettes in their lifetime) is rising, varies by sex and geography, and is linked to environmental exposures and family history. The misperception that lung cancer is almost invariably caused by smoking may delay assessment and diagnosis.
• Researchers conducted a review of 92 studies on lung cancer in nonsmokers: 6 meta-analyses or systematic reviews, 16 randomized clinical trials, eight prospective cohort studies, seven retrospective cohort studies, three cross-sectional studies, four observational or case-control studies, 13 genomic studies, and 35 other studies.
• Overall, lung cancer among nonsmokers accounted for 15% to 20% of all lung cancer cases. Most lung cancers in nonsmokers were adenocarcinomas (60% to 80%), with a median age at diagnosis of 67 years in this group compared with 70 years in people with a history of smoking.
• Data analysis from three US hospital networks showed that the proportion of lung cancer among nonsmokers increased from 8.0% to 14.9% between 1990 and 2013. A pooled analysis of seven Finnish cohorts reported an absolute increase in lung cancer among nonsmokers from 6.9 per 100,000 person-years in 1972 to 12.9 per 100,000 person-years in 2015.
• The age-adjusted incidence rate of lung cancer in the US between 2000 and 2013 was 17.5 per 100,000 individuals among Asian female nonsmokers compared with 10.1 per 100,000 among non-Hispanic White female nonsmokers.

TAKEAWAY:

• Environmental and occupational risk factors were secondhand smoke, residential radon, outdoor and household air pollution (PM_2.5), asbestos and silica exposure, and prior thoracic radiotherapy. Having a first-degree relative with lung cancer increased the risk of developing lung cancer, and genome-wide association studies identified susceptibility loci associated with lung cancer risk in nonsmokers.
• Family history and inherited susceptibility increased lung cancer risk in never smokers (odds ratio [OR] for lung cancer in those with a first–degree relative, 1.51), and clonal hematopoiesis was also associated with higher risk (OR, 1.43). Importantly, tumors in nonsmokers were frequently driven by actionable somatic alterations (EGFR mutations, 40% to 60% in nonsmokers compared with 10% in smokers) and enrichment of ALK/ROS1/RET/ERBB2/NTRK/NRG1 fusions; 78% to 92% of adenocarcinomas in nonsmokers harbored actionable drivers (compared with 49.5% in ever smokers), and nonsmokers had a substantially lower tumor mutational burden (10–fold lower).
• Similar to individuals with a history of smoking, nonsmokers with lung cancer presented with cough, pain, dyspnea, or weight loss or had disease detected incidentally. Surgical resection remained the preferred treatment for anatomically resectable lung cancer (stages I-III) in medically eligible patients, with follow-up CT screening recommended every 6 months for 2 to 3 years and then annually.
• Targeted adjuvant therapy substantially improved outcomes for resected EGFR–mutant or ALK–rearranged non-small cell lung cancer (NSCLC). Four-year DFS was increased to 70% with osimertinib compared with 29% with placebo (hazard ratio [HR], 0.23) and 5–year overall survival was increased to 85% compared with 73% (HR, 0.49). Two–year DFS was 93.8% with alectinib compared with 63% with placebo (HR, 0.24). In unresectable EGFR-mutated stage III NSCLC, median progression-free survival was 39.1 months with adjuvant osimertinib compared with 5.6 months with placebo. For resected ALKpositive disease, 2–year DFS was 93.8% with adjuvant alectinib compared with 63.0% with chemotherapy (HR, 0.24).
• However, singleagent single agent programmed cell death protein 1 inhibitors or programmed death-ligand 1 inhibitors demonstrated limited efficacy in EGFR or ALK–driven tumors, and benefit was attenuated in never smokers. Regarding screening and early detection, the US Preventive Services Task Force did not recommend lowdose CT screening for nonsmokers, whereas Taiwan implemented a biennial screening program for selected nonsmoking high–risk groups.

IN PRACTICE:

“Among patients with lung cancer, nonsmoking individuals are more likely to have genomic alterations, such as EGFR mutations or ALK gene rearrangements, and these patients have improved survival when treated with TKIs compared with chemotherapy,” the authors of the study wrote.

SOURCE:

The study, led by Cian Murphy, PhD, Cancer Evolution and Genome Instability Laboratory, Francis Crick Institute, London, England, was published online in JAMA.

LIMITATIONS:

Becausesmoking history was often not included in many databases, cancer registries, and trials, the incidence and prevalence of lung cancer in nonsmokers could not be accurately determined. Additionally, accurate quantification of environmental exposures, such as air pollution, presented significant challenges. The quality of the evidence was not formally evaluated, and some relevant articles may have been missed in the literature review.

DISCLOSURES:

The study received support from multiple organizations, including the Rosetrees Trust, Ruth Strauss Foundation, Cancer Research UK, and the National Health and Medical Research Council. Several authors reported receiving grants or personal fees from and having other ties with various sources. Full disclosures are noted in the original article.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.

A version of this article first appeared on Medscape.com.

A faster, oral alternative to docetaxel is set to reach NHS clinics after the National Institute for Health and Care Excellence (NICE) recommended darolutamide (Nubeqa, Bayer) in combination with androgen deprivation therapy (ADT) for men with metastatic hormone-sensitive prostate cancer who are unable to receive or tolerate chemotherapy.

Detailed in NICE’s final draft guidance, the decision will make darolutamide available through the NHS in England and Wales to approximately 6000 patients, offering a new oral therapy for those who with limited alternatives to docetaxel or other androgen-receptor inhibitors.

 

New Option for Chemo-Ineligible Patients

Darolutamide functions by blocking hormones that fuel cancer growth, specifically depriving prostate cancer cells of testosterone required for multiplication and spread. Patients take two tablets twice daily alongside standard ADT. 

Peter Johnson, national clinical director for cancer at NHS England, welcomed the decision and expects this approval to give clinicians and their patients “more flexibility to choose the approach best suited to individual circumstances and clinical needs.”

The guidance was finalised 5 weeks ahead of the standard review timeline, underscoring NICE’s commitment to accelerating access to effective prostate cancer treatments.

 

Clinical Trial Evidence

The NICE’s decision was supported by evidence from the phase 3 ARASENS trial (N = 1306). 

The results showed that adding darolutamide to ADT and docetaxel significantly improved overall survival in metastatic hormone-sensitive prostate cancer, reducing the risk for death by 32% compared with ADT and docetaxel alone. Progression-free outcomes, measured by time to castration-resistant disease or death, also favoured darolutamide. 

A NICE network meta-analysis of the TITAN, ARCHES, LATITUDE, and STAMPEDE trials suggested that combining ADT with androgen-receptor pathway inhibitors such as apalutamide, enzalutamide, and abiraterone provides comparable survival benefits in this disease setting.

 

Cost and Implementation

NICE determined that darolutamide plus ADT delivers similar or lower overall costs to the NHS compared with apalutamide plus ADT. The list price is £4040.00 for a 28-day supply (112 × 300-mg tablets), though Bayer has agreed to a confidential commercial discount.

The guidance requires healthcare providers to use the least expensive suitable treatment option, considering administration costs, dosages, price per dose, and commercial arrangements when choosing between darolutamide plus ADT and apalutamide plus ADT. 

NHS England and integrated care boards must provide funding within 30 days of final publication, with routine commissioning beginning after this interim period.

A version of this article first appeared on Medscape.com.

A faster, oral alternative to docetaxel is set to reach NHS clinics after the National Institute for Health and Care Excellence (NICE) recommended darolutamide (Nubeqa, Bayer) in combination with androgen deprivation therapy (ADT) for men with metastatic hormone-sensitive prostate cancer who are unable to receive or tolerate chemotherapy.

Detailed in NICE’s final draft guidance, the decision will make darolutamide available through the NHS in England and Wales to approximately 6000 patients, offering a new oral therapy for those who with limited alternatives to docetaxel or other androgen-receptor inhibitors.

 

New Option for Chemo-Ineligible Patients

Darolutamide functions by blocking hormones that fuel cancer growth, specifically depriving prostate cancer cells of testosterone required for multiplication and spread. Patients take two tablets twice daily alongside standard ADT. 

Peter Johnson, national clinical director for cancer at NHS England, welcomed the decision and expects this approval to give clinicians and their patients “more flexibility to choose the approach best suited to individual circumstances and clinical needs.”

The guidance was finalised 5 weeks ahead of the standard review timeline, underscoring NICE’s commitment to accelerating access to effective prostate cancer treatments.

 

Clinical Trial Evidence

The NICE’s decision was supported by evidence from the phase 3 ARASENS trial (N = 1306). 

The results showed that adding darolutamide to ADT and docetaxel significantly improved overall survival in metastatic hormone-sensitive prostate cancer, reducing the risk for death by 32% compared with ADT and docetaxel alone. Progression-free outcomes, measured by time to castration-resistant disease or death, also favoured darolutamide. 

A NICE network meta-analysis of the TITAN, ARCHES, LATITUDE, and STAMPEDE trials suggested that combining ADT with androgen-receptor pathway inhibitors such as apalutamide, enzalutamide, and abiraterone provides comparable survival benefits in this disease setting.

 

Cost and Implementation

NICE determined that darolutamide plus ADT delivers similar or lower overall costs to the NHS compared with apalutamide plus ADT. The list price is £4040.00 for a 28-day supply (112 × 300-mg tablets), though Bayer has agreed to a confidential commercial discount.

The guidance requires healthcare providers to use the least expensive suitable treatment option, considering administration costs, dosages, price per dose, and commercial arrangements when choosing between darolutamide plus ADT and apalutamide plus ADT. 

NHS England and integrated care boards must provide funding within 30 days of final publication, with routine commissioning beginning after this interim period.

A version of this article first appeared on Medscape.com.

A faster, oral alternative to docetaxel is set to reach NHS clinics after the National Institute for Health and Care Excellence (NICE) recommended darolutamide (Nubeqa, Bayer) in combination with androgen deprivation therapy (ADT) for men with metastatic hormone-sensitive prostate cancer who are unable to receive or tolerate chemotherapy.

Detailed in NICE’s final draft guidance, the decision will make darolutamide available through the NHS in England and Wales to approximately 6000 patients, offering a new oral therapy for those who with limited alternatives to docetaxel or other androgen-receptor inhibitors.

 

New Option for Chemo-Ineligible Patients

Darolutamide functions by blocking hormones that fuel cancer growth, specifically depriving prostate cancer cells of testosterone required for multiplication and spread. Patients take two tablets twice daily alongside standard ADT. 

Peter Johnson, national clinical director for cancer at NHS England, welcomed the decision and expects this approval to give clinicians and their patients “more flexibility to choose the approach best suited to individual circumstances and clinical needs.”

The guidance was finalised 5 weeks ahead of the standard review timeline, underscoring NICE’s commitment to accelerating access to effective prostate cancer treatments.

 

Clinical Trial Evidence

The NICE’s decision was supported by evidence from the phase 3 ARASENS trial (N = 1306). 

The results showed that adding darolutamide to ADT and docetaxel significantly improved overall survival in metastatic hormone-sensitive prostate cancer, reducing the risk for death by 32% compared with ADT and docetaxel alone. Progression-free outcomes, measured by time to castration-resistant disease or death, also favoured darolutamide. 

A NICE network meta-analysis of the TITAN, ARCHES, LATITUDE, and STAMPEDE trials suggested that combining ADT with androgen-receptor pathway inhibitors such as apalutamide, enzalutamide, and abiraterone provides comparable survival benefits in this disease setting.

 

Cost and Implementation

NICE determined that darolutamide plus ADT delivers similar or lower overall costs to the NHS compared with apalutamide plus ADT. The list price is £4040.00 for a 28-day supply (112 × 300-mg tablets), though Bayer has agreed to a confidential commercial discount.

The guidance requires healthcare providers to use the least expensive suitable treatment option, considering administration costs, dosages, price per dose, and commercial arrangements when choosing between darolutamide plus ADT and apalutamide plus ADT. 

NHS England and integrated care boards must provide funding within 30 days of final publication, with routine commissioning beginning after this interim period.

A version of this article first appeared on Medscape.com.

In this Practical AI column, we’ve explored everything from large language models to the nuances of trial matching, but one of the most immediate and impactful applications of AI is unfolding right now in breast imaging. For oncologists, this isn’t an abstract future — with new screening guidelines, dense-breast mandates, and a shrinking radiology workforce, it’s the imaging reports and patient questions landing in your clinic today.

Here is what oncologists need to know, and how to put it to work for their patients.

 

Why AI in Mammography Matters

More than 200 million women undergo breast cancer screening each year. In the US alone, 10% of the 40 million women screened annually require additional diagnostic imaging, and 4%–5% of these women are eventually diagnosed with breast cancer.

Two major shifts are redefining breast cancer screening in the US: The US Preventive Services Task Force (USPSTF) now recommends biennial screening from age 40 to 74 years, and notifying patients of breast density is a federal requirement as of September 10, 2024. That means more mammograms, more patient questions, and more downstream oncology decisions. Patients will increasingly ask about “dense” breast results and what to do next. Add a national radiologist shortage into the mix, and the pressure on timely callbacks, biopsies, and treatment planning will only grow.

 

Can AI Help Without Compromising Care?

The short answer is yes. With AI, we may be able to transform these rate-limiting steps into opportunities for earlier detection, decentralized screening, and smarter triage and save hundreds of thousands of women from an unnecessary diagnostic procedure, if implemented deliberately.

Don’t Confuse Today’s AI With Yesterday’s CAD 

Think of older computer-aided detection (CAD) like a 1990s chemotherapy drug: It sometimes helped, but it came with significant toxicity and rarely delivered consistent survival benefits. Today’s deep-learning AI is closer to targeted therapy — trained on millions of “trial participants” (mammograms), more precise, and applied in specific contexts where it adds value. If you once dismissed CAD as noise, it’s time to revisit what AI can now offer.

The role of AI is broader than drawing boxes. It provides second readings, worklist triage, risk prediction, density assessment, and decision support. FDA has cleared several AI tools for both 2D and digital breast tomosynthesis (DBT), which include iCAD ProFound (DBT), ScreenPoint Transpara (2D/DBT), and Lunit INSIGHT DBT. 

Some of the strongest evidence for AI in mammography is as a second reader during screening. Large trials show that AI plus one radiologist can match reading from two radiologists, cutting workload by about 40%. For example, the MASAI randomized trial showed that AI-supported screening achieved similar cancer detection but cut human screen-reading workload about 44% vs standard double reading (39,996 vs 40,024 participants). The primary interval cancer outcomes are maturing, but the safety analysis is reassuring.

Reducing second reads and arbitration time are important for clinicians because it frees capacity for callbacks and diagnostic workups. This will be especially key given that screening now starts at age 40. That will mean about 21 to 22 million more women are newly eligible, translating to about 10 to 11 million additional mammograms each year under biennial screening.

Another important area where AI can make its mark in mammography is triage and time to diagnosis. The results from a randomized implementation study showed that AI-prioritized worklists accelerated time to additional imaging and biopsy diagnosis without harming efficiency for others — exactly the kind of outcome patients feel.

Multiple studies have demonstrated improved diagnostic performance and shorter reading times when AI supports DBT interpretation, which is important because DBT can otherwise be time intensive.

We are also seeing rapid advancement in risk-based screening, moving beyond a single dense vs not dense approach. Deep-learning risk models, such as Mirai, predict 1- to 5-year breast cancer risk directly from the mammogram, and these tools are now being assessed prospectively to guide supplemental MRI. Cost-effectiveness modeling supports risk-stratified intervals vs one-size-fits-all schedules.

Finally, automated density tools, such as Transpara Density and Volpara, offer objective, reproducible volumetric measures that map to the Breast Imaging-Reporting and Data System, which is useful for Mammography Quality Standards Act-required reporting and as inputs to risk calculators.

While early evidence suggests AI may help surface future or interval cancers earlier, including more invasive tumors, the definitive impacts on interval cancer rates and mortality require longitudinal follow-up, which is now in progress.

 

Pitfalls to Watch For

Bias is real. Studies show false-positive differences by race, age, and density. AI can even infer racial identity from images, potentially amplifying disparities. Performance can also shift by vendor, demographics, and prevalence.

A Radiology study of 4855 DBT exams showed that an algorithm produced more false-positive case scores in Black patients and older patients (aged 71-80 years) patients and in women with extremely dense breasts. This can happen because AI can infer proxies for race directly from images, even when humans cannot, and this can propagate disparities if not addressed. External validations and reviews emphasize that performance can shift with device manufacturer, demographics, and prevalence, which is why all tools need to undergo local validation and calibration. 

Here’s a pragmatic adoption checklist before going live with an AI tool.

• Confirm FDA clearance: Verify the name and version of the algorithm, imaging modes (2D vs DBT), and operating points. Confirm 510(k) numbers.
• Local validation: Test on your patient mix and vendor stack (Hologic, GE, Siemens, Fuji). Compare this to your baseline recall rate, positive predictive value of recall (PPV1), cancer detection rate, and reading time. Commit to recalibration if drift occurs.
• Equity plan: Monitor false-positive and negative false-rates by age, race/ethnicity, and density; document corrective actions if disparities emerge. (This isn’t optional.)
• Workflow clarity: Is AI a second reader, an additional reader, or a triage tool? Who arbitrates discordance? What’s the escalation path for high-risk or interval cancer-like patterns?
• Regulatory strategy: Confirm whether the vendor has (or will file) a Predetermined Change Control Plan so models can be updated safely without repeated submissions. Also confirm how you’ll be notified about performance-relevant changes.
• Data governance: Audit logs of AI outputs, retention, protected health information handling, and the patient communication policy for AI-assisted reads.

After going live, set up a quarterly dashboard. It should include cancer detection rate per 1000 patients, recall rate, PPV1, interval cancer rate (as it matures), reading time, and turnaround time to diagnostic imaging or biopsy — all stratified by age, race/ethnicity, and density.

Here, I dissect what this discussion means through the lens of Moravec’s paradox (machines excel at what clinicians find hard, and vice versa) and offer a possible playbook for putting these tools to work.

 

What to Tell Patients

When speaking with patients, emphasize that a radiologist still reads their mammogram. AI helps with consistency and efficiency; it doesn’t replace human oversight. Patients with dense breasts should still expect a standard notice; discussion of individualized risk factors, such as family history, genetics, and prior biopsies; and consideration of supplemental imaging if risk warrants. But it’s also important to tell these patients that while dense breasts are common, they do not automatically mean high cancer risk.

As for screening schedules, remind patients that screening is at least biennial from 40 to 74 years of age per the USPSTF guidelines; however, specialty groups may recommend starting on an annual schedule at 40.

 

What You Can Implement Now

There are multiple practical use cases you can introduce now. One is to use AI as a second reader or an additional reader safety net to preserve detection while reducing human workload. This helps your breast center absorb screening expansion to age 40 without diluting quality. Another is to turn on AI triage to shorten the time to callback and biopsy for the few who need it most — patients notice and appreciate faster answers. You can also begin adopting automated density plus risk models to move beyond “dense/not dense.” For selected patients, AI-informed risk can justify MRI or tailored intervals. 

Here’s a quick cheat sheet (for your next leadership or tumor-board meeting).

 

Do:

• Use AI as a second or additional reader or triage tool, not as a black box.
• Track cancer detection rate, recall, PPV1, interval cancers, and reading time, stratified by age, race, and breast density.
• Pair automated density with AI risk to personalize screening and supplemental imaging.
• Enroll patients in future clinical trials, such as PRISM, the first large-scale randomized controlled trial of AI for screening mammography. This US-based, $16 million, seven-site study is funded by the Patient-Centered Outcomes Research Institute.

Don’t:

• Assume “AI = CAD.” The 2015 CAD story is over; modern deep learning systems are different and require different oversight.
• Go live without a local validation and equity plan or without clarity on software updates.
• Forget to remind patients that screening starts at age 40, and dense breast notifications are now universal. Use the visit to discuss risk, supplemental imaging, and why a human still directs their care.

The Bottom Line

AI won’t replace radiologists or read mammograms for us — just as PET scans didn’t replace oncologists and stethoscopes didn’t make cardiologists obsolete. What it will do is catch what the tired human eye might miss, shave days off anxious waiting, and turn breast density into data instead of doubt. For oncologists, that means staging sooner, enrolling smarter, and spending more time talking with patients instead of chasing callbacks.

In short, AI may not take the picture, but it helps us frame the story, making it sharper, faster, and with fewer blind spots. By pairing this powerful technology with rigorous, equity-focused local validation and transparent governance under the FDA’s emerging Predetermined Change Control Plan framework, we can realize the tangible benefits of practical AI for our patients without widening disparities. 

Now, during Breast Cancer Awareness Month, how about we add on AI to that pink ribbon — how cool would that be?

Thoughts? Drop me a line at Arturo.AI.MedTech@gmail.com. Let’s keep the conversation — and pink ribbons — going.

Arturo Loaiza-Bonilla, MD, MSEd, is the co-founder and chief medical AI officer at Massive Bio, a company connecting patients to clinical trials using artificial intelligence. His research and professional interests focus on precision medicine, clinical trial design, digital health, entrepreneurship, and patient advocacy. Dr Loaiza-Bonilla serves as Systemwide Chief of Hematology and Oncology at St. Luke’s University Health Network, where he maintains a connection to patient care by attending to patients 2 days a week.

A version of this article first appeared on Medscape.com.

In this Practical AI column, we’ve explored everything from large language models to the nuances of trial matching, but one of the most immediate and impactful applications of AI is unfolding right now in breast imaging. For oncologists, this isn’t an abstract future — with new screening guidelines, dense-breast mandates, and a shrinking radiology workforce, it’s the imaging reports and patient questions landing in your clinic today.

Here is what oncologists need to know, and how to put it to work for their patients.

 

Why AI in Mammography Matters

More than 200 million women undergo breast cancer screening each year. In the US alone, 10% of the 40 million women screened annually require additional diagnostic imaging, and 4%–5% of these women are eventually diagnosed with breast cancer.

Two major shifts are redefining breast cancer screening in the US: The US Preventive Services Task Force (USPSTF) now recommends biennial screening from age 40 to 74 years, and notifying patients of breast density is a federal requirement as of September 10, 2024. That means more mammograms, more patient questions, and more downstream oncology decisions. Patients will increasingly ask about “dense” breast results and what to do next. Add a national radiologist shortage into the mix, and the pressure on timely callbacks, biopsies, and treatment planning will only grow.

 

Can AI Help Without Compromising Care?

The short answer is yes. With AI, we may be able to transform these rate-limiting steps into opportunities for earlier detection, decentralized screening, and smarter triage and save hundreds of thousands of women from an unnecessary diagnostic procedure, if implemented deliberately.

Don’t Confuse Today’s AI With Yesterday’s CAD 

Think of older computer-aided detection (CAD) like a 1990s chemotherapy drug: It sometimes helped, but it came with significant toxicity and rarely delivered consistent survival benefits. Today’s deep-learning AI is closer to targeted therapy — trained on millions of “trial participants” (mammograms), more precise, and applied in specific contexts where it adds value. If you once dismissed CAD as noise, it’s time to revisit what AI can now offer.

The role of AI is broader than drawing boxes. It provides second readings, worklist triage, risk prediction, density assessment, and decision support. FDA has cleared several AI tools for both 2D and digital breast tomosynthesis (DBT), which include iCAD ProFound (DBT), ScreenPoint Transpara (2D/DBT), and Lunit INSIGHT DBT. 

Some of the strongest evidence for AI in mammography is as a second reader during screening. Large trials show that AI plus one radiologist can match reading from two radiologists, cutting workload by about 40%. For example, the MASAI randomized trial showed that AI-supported screening achieved similar cancer detection but cut human screen-reading workload about 44% vs standard double reading (39,996 vs 40,024 participants). The primary interval cancer outcomes are maturing, but the safety analysis is reassuring.

Reducing second reads and arbitration time are important for clinicians because it frees capacity for callbacks and diagnostic workups. This will be especially key given that screening now starts at age 40. That will mean about 21 to 22 million more women are newly eligible, translating to about 10 to 11 million additional mammograms each year under biennial screening.

Another important area where AI can make its mark in mammography is triage and time to diagnosis. The results from a randomized implementation study showed that AI-prioritized worklists accelerated time to additional imaging and biopsy diagnosis without harming efficiency for others — exactly the kind of outcome patients feel.

Multiple studies have demonstrated improved diagnostic performance and shorter reading times when AI supports DBT interpretation, which is important because DBT can otherwise be time intensive.

We are also seeing rapid advancement in risk-based screening, moving beyond a single dense vs not dense approach. Deep-learning risk models, such as Mirai, predict 1- to 5-year breast cancer risk directly from the mammogram, and these tools are now being assessed prospectively to guide supplemental MRI. Cost-effectiveness modeling supports risk-stratified intervals vs one-size-fits-all schedules.

Finally, automated density tools, such as Transpara Density and Volpara, offer objective, reproducible volumetric measures that map to the Breast Imaging-Reporting and Data System, which is useful for Mammography Quality Standards Act-required reporting and as inputs to risk calculators.

While early evidence suggests AI may help surface future or interval cancers earlier, including more invasive tumors, the definitive impacts on interval cancer rates and mortality require longitudinal follow-up, which is now in progress.

 

Pitfalls to Watch For

Bias is real. Studies show false-positive differences by race, age, and density. AI can even infer racial identity from images, potentially amplifying disparities. Performance can also shift by vendor, demographics, and prevalence.

A Radiology study of 4855 DBT exams showed that an algorithm produced more false-positive case scores in Black patients and older patients (aged 71-80 years) patients and in women with extremely dense breasts. This can happen because AI can infer proxies for race directly from images, even when humans cannot, and this can propagate disparities if not addressed. External validations and reviews emphasize that performance can shift with device manufacturer, demographics, and prevalence, which is why all tools need to undergo local validation and calibration. 

Here’s a pragmatic adoption checklist before going live with an AI tool.

• Confirm FDA clearance: Verify the name and version of the algorithm, imaging modes (2D vs DBT), and operating points. Confirm 510(k) numbers.
• Local validation: Test on your patient mix and vendor stack (Hologic, GE, Siemens, Fuji). Compare this to your baseline recall rate, positive predictive value of recall (PPV1), cancer detection rate, and reading time. Commit to recalibration if drift occurs.
• Equity plan: Monitor false-positive and negative false-rates by age, race/ethnicity, and density; document corrective actions if disparities emerge. (This isn’t optional.)
• Workflow clarity: Is AI a second reader, an additional reader, or a triage tool? Who arbitrates discordance? What’s the escalation path for high-risk or interval cancer-like patterns?
• Regulatory strategy: Confirm whether the vendor has (or will file) a Predetermined Change Control Plan so models can be updated safely without repeated submissions. Also confirm how you’ll be notified about performance-relevant changes.
• Data governance: Audit logs of AI outputs, retention, protected health information handling, and the patient communication policy for AI-assisted reads.

After going live, set up a quarterly dashboard. It should include cancer detection rate per 1000 patients, recall rate, PPV1, interval cancer rate (as it matures), reading time, and turnaround time to diagnostic imaging or biopsy — all stratified by age, race/ethnicity, and density.

Here, I dissect what this discussion means through the lens of Moravec’s paradox (machines excel at what clinicians find hard, and vice versa) and offer a possible playbook for putting these tools to work.

 

What to Tell Patients

When speaking with patients, emphasize that a radiologist still reads their mammogram. AI helps with consistency and efficiency; it doesn’t replace human oversight. Patients with dense breasts should still expect a standard notice; discussion of individualized risk factors, such as family history, genetics, and prior biopsies; and consideration of supplemental imaging if risk warrants. But it’s also important to tell these patients that while dense breasts are common, they do not automatically mean high cancer risk.

As for screening schedules, remind patients that screening is at least biennial from 40 to 74 years of age per the USPSTF guidelines; however, specialty groups may recommend starting on an annual schedule at 40.

 

What You Can Implement Now

There are multiple practical use cases you can introduce now. One is to use AI as a second reader or an additional reader safety net to preserve detection while reducing human workload. This helps your breast center absorb screening expansion to age 40 without diluting quality. Another is to turn on AI triage to shorten the time to callback and biopsy for the few who need it most — patients notice and appreciate faster answers. You can also begin adopting automated density plus risk models to move beyond “dense/not dense.” For selected patients, AI-informed risk can justify MRI or tailored intervals. 

Here’s a quick cheat sheet (for your next leadership or tumor-board meeting).

 

Do:

• Use AI as a second or additional reader or triage tool, not as a black box.
• Track cancer detection rate, recall, PPV1, interval cancers, and reading time, stratified by age, race, and breast density.
• Pair automated density with AI risk to personalize screening and supplemental imaging.
• Enroll patients in future clinical trials, such as PRISM, the first large-scale randomized controlled trial of AI for screening mammography. This US-based, $16 million, seven-site study is funded by the Patient-Centered Outcomes Research Institute.

Don’t:

• Assume “AI = CAD.” The 2015 CAD story is over; modern deep learning systems are different and require different oversight.
• Go live without a local validation and equity plan or without clarity on software updates.
• Forget to remind patients that screening starts at age 40, and dense breast notifications are now universal. Use the visit to discuss risk, supplemental imaging, and why a human still directs their care.

The Bottom Line

AI won’t replace radiologists or read mammograms for us — just as PET scans didn’t replace oncologists and stethoscopes didn’t make cardiologists obsolete. What it will do is catch what the tired human eye might miss, shave days off anxious waiting, and turn breast density into data instead of doubt. For oncologists, that means staging sooner, enrolling smarter, and spending more time talking with patients instead of chasing callbacks.

In short, AI may not take the picture, but it helps us frame the story, making it sharper, faster, and with fewer blind spots. By pairing this powerful technology with rigorous, equity-focused local validation and transparent governance under the FDA’s emerging Predetermined Change Control Plan framework, we can realize the tangible benefits of practical AI for our patients without widening disparities. 

Now, during Breast Cancer Awareness Month, how about we add on AI to that pink ribbon — how cool would that be?

Thoughts? Drop me a line at Arturo.AI.MedTech@gmail.com. Let’s keep the conversation — and pink ribbons — going.

Arturo Loaiza-Bonilla, MD, MSEd, is the co-founder and chief medical AI officer at Massive Bio, a company connecting patients to clinical trials using artificial intelligence. His research and professional interests focus on precision medicine, clinical trial design, digital health, entrepreneurship, and patient advocacy. Dr Loaiza-Bonilla serves as Systemwide Chief of Hematology and Oncology at St. Luke’s University Health Network, where he maintains a connection to patient care by attending to patients 2 days a week.

A version of this article first appeared on Medscape.com.

In this Practical AI column, we’ve explored everything from large language models to the nuances of trial matching, but one of the most immediate and impactful applications of AI is unfolding right now in breast imaging. For oncologists, this isn’t an abstract future — with new screening guidelines, dense-breast mandates, and a shrinking radiology workforce, it’s the imaging reports and patient questions landing in your clinic today.

Here is what oncologists need to know, and how to put it to work for their patients.

 

Why AI in Mammography Matters

More than 200 million women undergo breast cancer screening each year. In the US alone, 10% of the 40 million women screened annually require additional diagnostic imaging, and 4%–5% of these women are eventually diagnosed with breast cancer.

Two major shifts are redefining breast cancer screening in the US: The US Preventive Services Task Force (USPSTF) now recommends biennial screening from age 40 to 74 years, and notifying patients of breast density is a federal requirement as of September 10, 2024. That means more mammograms, more patient questions, and more downstream oncology decisions. Patients will increasingly ask about “dense” breast results and what to do next. Add a national radiologist shortage into the mix, and the pressure on timely callbacks, biopsies, and treatment planning will only grow.

 

Can AI Help Without Compromising Care?

The short answer is yes. With AI, we may be able to transform these rate-limiting steps into opportunities for earlier detection, decentralized screening, and smarter triage and save hundreds of thousands of women from an unnecessary diagnostic procedure, if implemented deliberately.

Don’t Confuse Today’s AI With Yesterday’s CAD 

Think of older computer-aided detection (CAD) like a 1990s chemotherapy drug: It sometimes helped, but it came with significant toxicity and rarely delivered consistent survival benefits. Today’s deep-learning AI is closer to targeted therapy — trained on millions of “trial participants” (mammograms), more precise, and applied in specific contexts where it adds value. If you once dismissed CAD as noise, it’s time to revisit what AI can now offer.

The role of AI is broader than drawing boxes. It provides second readings, worklist triage, risk prediction, density assessment, and decision support. FDA has cleared several AI tools for both 2D and digital breast tomosynthesis (DBT), which include iCAD ProFound (DBT), ScreenPoint Transpara (2D/DBT), and Lunit INSIGHT DBT. 

Some of the strongest evidence for AI in mammography is as a second reader during screening. Large trials show that AI plus one radiologist can match reading from two radiologists, cutting workload by about 40%. For example, the MASAI randomized trial showed that AI-supported screening achieved similar cancer detection but cut human screen-reading workload about 44% vs standard double reading (39,996 vs 40,024 participants). The primary interval cancer outcomes are maturing, but the safety analysis is reassuring.

Reducing second reads and arbitration time are important for clinicians because it frees capacity for callbacks and diagnostic workups. This will be especially key given that screening now starts at age 40. That will mean about 21 to 22 million more women are newly eligible, translating to about 10 to 11 million additional mammograms each year under biennial screening.

Another important area where AI can make its mark in mammography is triage and time to diagnosis. The results from a randomized implementation study showed that AI-prioritized worklists accelerated time to additional imaging and biopsy diagnosis without harming efficiency for others — exactly the kind of outcome patients feel.

Multiple studies have demonstrated improved diagnostic performance and shorter reading times when AI supports DBT interpretation, which is important because DBT can otherwise be time intensive.

We are also seeing rapid advancement in risk-based screening, moving beyond a single dense vs not dense approach. Deep-learning risk models, such as Mirai, predict 1- to 5-year breast cancer risk directly from the mammogram, and these tools are now being assessed prospectively to guide supplemental MRI. Cost-effectiveness modeling supports risk-stratified intervals vs one-size-fits-all schedules.

Finally, automated density tools, such as Transpara Density and Volpara, offer objective, reproducible volumetric measures that map to the Breast Imaging-Reporting and Data System, which is useful for Mammography Quality Standards Act-required reporting and as inputs to risk calculators.

While early evidence suggests AI may help surface future or interval cancers earlier, including more invasive tumors, the definitive impacts on interval cancer rates and mortality require longitudinal follow-up, which is now in progress.

 

Pitfalls to Watch For

Bias is real. Studies show false-positive differences by race, age, and density. AI can even infer racial identity from images, potentially amplifying disparities. Performance can also shift by vendor, demographics, and prevalence.

A Radiology study of 4855 DBT exams showed that an algorithm produced more false-positive case scores in Black patients and older patients (aged 71-80 years) patients and in women with extremely dense breasts. This can happen because AI can infer proxies for race directly from images, even when humans cannot, and this can propagate disparities if not addressed. External validations and reviews emphasize that performance can shift with device manufacturer, demographics, and prevalence, which is why all tools need to undergo local validation and calibration. 

Here’s a pragmatic adoption checklist before going live with an AI tool.

• Confirm FDA clearance: Verify the name and version of the algorithm, imaging modes (2D vs DBT), and operating points. Confirm 510(k) numbers.
• Local validation: Test on your patient mix and vendor stack (Hologic, GE, Siemens, Fuji). Compare this to your baseline recall rate, positive predictive value of recall (PPV1), cancer detection rate, and reading time. Commit to recalibration if drift occurs.
• Equity plan: Monitor false-positive and negative false-rates by age, race/ethnicity, and density; document corrective actions if disparities emerge. (This isn’t optional.)
• Workflow clarity: Is AI a second reader, an additional reader, or a triage tool? Who arbitrates discordance? What’s the escalation path for high-risk or interval cancer-like patterns?
• Regulatory strategy: Confirm whether the vendor has (or will file) a Predetermined Change Control Plan so models can be updated safely without repeated submissions. Also confirm how you’ll be notified about performance-relevant changes.
• Data governance: Audit logs of AI outputs, retention, protected health information handling, and the patient communication policy for AI-assisted reads.

After going live, set up a quarterly dashboard. It should include cancer detection rate per 1000 patients, recall rate, PPV1, interval cancer rate (as it matures), reading time, and turnaround time to diagnostic imaging or biopsy — all stratified by age, race/ethnicity, and density.

Here, I dissect what this discussion means through the lens of Moravec’s paradox (machines excel at what clinicians find hard, and vice versa) and offer a possible playbook for putting these tools to work.

 

What to Tell Patients

When speaking with patients, emphasize that a radiologist still reads their mammogram. AI helps with consistency and efficiency; it doesn’t replace human oversight. Patients with dense breasts should still expect a standard notice; discussion of individualized risk factors, such as family history, genetics, and prior biopsies; and consideration of supplemental imaging if risk warrants. But it’s also important to tell these patients that while dense breasts are common, they do not automatically mean high cancer risk.

As for screening schedules, remind patients that screening is at least biennial from 40 to 74 years of age per the USPSTF guidelines; however, specialty groups may recommend starting on an annual schedule at 40.

 

What You Can Implement Now

There are multiple practical use cases you can introduce now. One is to use AI as a second reader or an additional reader safety net to preserve detection while reducing human workload. This helps your breast center absorb screening expansion to age 40 without diluting quality. Another is to turn on AI triage to shorten the time to callback and biopsy for the few who need it most — patients notice and appreciate faster answers. You can also begin adopting automated density plus risk models to move beyond “dense/not dense.” For selected patients, AI-informed risk can justify MRI or tailored intervals. 

Here’s a quick cheat sheet (for your next leadership or tumor-board meeting).

 

Do:

• Use AI as a second or additional reader or triage tool, not as a black box.
• Track cancer detection rate, recall, PPV1, interval cancers, and reading time, stratified by age, race, and breast density.
• Pair automated density with AI risk to personalize screening and supplemental imaging.
• Enroll patients in future clinical trials, such as PRISM, the first large-scale randomized controlled trial of AI for screening mammography. This US-based, $16 million, seven-site study is funded by the Patient-Centered Outcomes Research Institute.

Don’t:

• Assume “AI = CAD.” The 2015 CAD story is over; modern deep learning systems are different and require different oversight.
• Go live without a local validation and equity plan or without clarity on software updates.
• Forget to remind patients that screening starts at age 40, and dense breast notifications are now universal. Use the visit to discuss risk, supplemental imaging, and why a human still directs their care.

The Bottom Line

AI won’t replace radiologists or read mammograms for us — just as PET scans didn’t replace oncologists and stethoscopes didn’t make cardiologists obsolete. What it will do is catch what the tired human eye might miss, shave days off anxious waiting, and turn breast density into data instead of doubt. For oncologists, that means staging sooner, enrolling smarter, and spending more time talking with patients instead of chasing callbacks.

In short, AI may not take the picture, but it helps us frame the story, making it sharper, faster, and with fewer blind spots. By pairing this powerful technology with rigorous, equity-focused local validation and transparent governance under the FDA’s emerging Predetermined Change Control Plan framework, we can realize the tangible benefits of practical AI for our patients without widening disparities. 

Now, during Breast Cancer Awareness Month, how about we add on AI to that pink ribbon — how cool would that be?

Thoughts? Drop me a line at Arturo.AI.MedTech@gmail.com. Let’s keep the conversation — and pink ribbons — going.

Arturo Loaiza-Bonilla, MD, MSEd, is the co-founder and chief medical AI officer at Massive Bio, a company connecting patients to clinical trials using artificial intelligence. His research and professional interests focus on precision medicine, clinical trial design, digital health, entrepreneurship, and patient advocacy. Dr Loaiza-Bonilla serves as Systemwide Chief of Hematology and Oncology at St. Luke’s University Health Network, where he maintains a connection to patient care by attending to patients 2 days a week.

A version of this article first appeared on Medscape.com.

PHOENIX – Medications used in oncology clinical trials pose unique challenges in areas such as labeling, packaging, and administration, a US Department of Veterans Affairs (VA) pharmacist cautioned colleagues, and placebos have special needs too.

Even basic safety protections can be lacking when a drug is investigational, said Emily Hennes, PharmD, BCOP, clinical pharmacy specialist for research at William S. Middleton Memorial Veterans Hospital in Shorewood Hills, Wisconsin, in a presentation at the annual meeting of the Association of VA Hematology/Oncology.

“All of the safety features that we have come to know and love in dispensing commercial drugs are absent. There’s no Tall Man lettering, there's no color differentiation, and there's no barcoding, because these are not registered drugs," she said.

A 2017 report found that 81% of pharmacists surveyed indicated some level of concern regarding the safety risk in using  investigational drugs. At the same time, Hennes noted, the Joint Commission has mandated that pharmacists must control the storage, dispensing, labeling, and distribution of investigational medications.

Here are things to know about the use of investigational cancer drugs:

Drug Interactions Are Common

Hennes highlighted a 2023 study of medication reconciliation of 501 patients in 79 clinical trials that found alarming levels of drug interactions: 

• 360 clinically relevant drug-drug interactions were identified among 189 patients, including 158 therapies that were prohibited by protocols. Of these, 57.7% involved cytochrome P450 enzymes, which are involved in metabolism. 

• Reconciliation revealed that 35.2% of medications were not otherwise known or documented.

• A median of 2 previously unknown therapies per patient was discovered in 74% of patients.

• Alternative medicine products such as supplements and over-the-counter drugs were implicated in 60% of identified drug interactions.

• Only 41% of oncologists discussed alternative medicine use with patients, which Hennes attributed to “lack of familiarity with many alternative medicine products or insufficient training.”

To make things more complicated, “We sometimes don’t know the full pharmacokinetic and pharmacodynamic profile of an investigational agent,” she said. 

Naming and Labeling May Not Be Standard

Investigational products may not have genetic names and instead have an alphanumeric identifier such as INV54826 that can be quite similar to other products, she said. Investigational drugs may even go through name changes, forcing pharmacists to be alerted to protect patients. 

In addition, labeling may not be standardized. Drugs may arrive unlabeled, with the wrong volume and size, and lack of barcoding. In some cases, pharmacists choose to put new, patient-friendly labels on these products, Hennes said. 

Information Distribution is Key

“Something that comes up in our practice quite a bit is that there’s no standard drug reference regarding investigational drugs,” Hennes said. “Finding ways to get key information to staff at the point of care is really critical to make sure we’re able to safely treat our patients.”

Precautions May Be Needed to Maintain Blinding Protocols 

Hennes explained that pharmacists must use opaque brown bag covers to maintain blinding when parenteral products have distinctive colors. Lines may have to be covered too, which can create challenges during administration. 

“Pumps aren’t meant to run lines that are covered,” she said, which can lead to jams. “If you don’t do education with your point of care staff, it can cause a lot of confusion.” 

It’s also important for blinding purposes to keep an eye on how long it takes to prepare a treatment, she said. A study’s integrity, for example, could be violated if a complex investigational product takes an hour to equilibrate to room temperature and 20-30 minutes to prepare, while a placebo only requires “drawing a few mils of saline out of a bag and labeling it.”

Education for Patients Can Be Useful 

Hennes urged colleagues to remind patients to save investigational medication at the end of each cycle and return it to the clinic site for accountability.

She also suggested creating treatment calendars/reminders for patients and discussing adherence strategies with them. 

Hennes reported no disclosures. 

PHOENIX – Medications used in oncology clinical trials pose unique challenges in areas such as labeling, packaging, and administration, a US Department of Veterans Affairs (VA) pharmacist cautioned colleagues, and placebos have special needs too.

Even basic safety protections can be lacking when a drug is investigational, said Emily Hennes, PharmD, BCOP, clinical pharmacy specialist for research at William S. Middleton Memorial Veterans Hospital in Shorewood Hills, Wisconsin, in a presentation at the annual meeting of the Association of VA Hematology/Oncology.

“All of the safety features that we have come to know and love in dispensing commercial drugs are absent. There’s no Tall Man lettering, there's no color differentiation, and there's no barcoding, because these are not registered drugs," she said.

A 2017 report found that 81% of pharmacists surveyed indicated some level of concern regarding the safety risk in using  investigational drugs. At the same time, Hennes noted, the Joint Commission has mandated that pharmacists must control the storage, dispensing, labeling, and distribution of investigational medications.

Here are things to know about the use of investigational cancer drugs:

Drug Interactions Are Common

Hennes highlighted a 2023 study of medication reconciliation of 501 patients in 79 clinical trials that found alarming levels of drug interactions: 

• 360 clinically relevant drug-drug interactions were identified among 189 patients, including 158 therapies that were prohibited by protocols. Of these, 57.7% involved cytochrome P450 enzymes, which are involved in metabolism. 

• Reconciliation revealed that 35.2% of medications were not otherwise known or documented.

• A median of 2 previously unknown therapies per patient was discovered in 74% of patients.

• Alternative medicine products such as supplements and over-the-counter drugs were implicated in 60% of identified drug interactions.

• Only 41% of oncologists discussed alternative medicine use with patients, which Hennes attributed to “lack of familiarity with many alternative medicine products or insufficient training.”

To make things more complicated, “We sometimes don’t know the full pharmacokinetic and pharmacodynamic profile of an investigational agent,” she said. 

Naming and Labeling May Not Be Standard

Investigational products may not have genetic names and instead have an alphanumeric identifier such as INV54826 that can be quite similar to other products, she said. Investigational drugs may even go through name changes, forcing pharmacists to be alerted to protect patients. 

In addition, labeling may not be standardized. Drugs may arrive unlabeled, with the wrong volume and size, and lack of barcoding. In some cases, pharmacists choose to put new, patient-friendly labels on these products, Hennes said. 

Information Distribution is Key

“Something that comes up in our practice quite a bit is that there’s no standard drug reference regarding investigational drugs,” Hennes said. “Finding ways to get key information to staff at the point of care is really critical to make sure we’re able to safely treat our patients.”

Precautions May Be Needed to Maintain Blinding Protocols 

Hennes explained that pharmacists must use opaque brown bag covers to maintain blinding when parenteral products have distinctive colors. Lines may have to be covered too, which can create challenges during administration. 

“Pumps aren’t meant to run lines that are covered,” she said, which can lead to jams. “If you don’t do education with your point of care staff, it can cause a lot of confusion.” 

It’s also important for blinding purposes to keep an eye on how long it takes to prepare a treatment, she said. A study’s integrity, for example, could be violated if a complex investigational product takes an hour to equilibrate to room temperature and 20-30 minutes to prepare, while a placebo only requires “drawing a few mils of saline out of a bag and labeling it.”

Education for Patients Can Be Useful 

Hennes urged colleagues to remind patients to save investigational medication at the end of each cycle and return it to the clinic site for accountability.

She also suggested creating treatment calendars/reminders for patients and discussing adherence strategies with them. 

Hennes reported no disclosures. 

PHOENIX – Medications used in oncology clinical trials pose unique challenges in areas such as labeling, packaging, and administration, a US Department of Veterans Affairs (VA) pharmacist cautioned colleagues, and placebos have special needs too.

Even basic safety protections can be lacking when a drug is investigational, said Emily Hennes, PharmD, BCOP, clinical pharmacy specialist for research at William S. Middleton Memorial Veterans Hospital in Shorewood Hills, Wisconsin, in a presentation at the annual meeting of the Association of VA Hematology/Oncology.

“All of the safety features that we have come to know and love in dispensing commercial drugs are absent. There’s no Tall Man lettering, there's no color differentiation, and there's no barcoding, because these are not registered drugs," she said.

A 2017 report found that 81% of pharmacists surveyed indicated some level of concern regarding the safety risk in using  investigational drugs. At the same time, Hennes noted, the Joint Commission has mandated that pharmacists must control the storage, dispensing, labeling, and distribution of investigational medications.

Here are things to know about the use of investigational cancer drugs:

Drug Interactions Are Common

Hennes highlighted a 2023 study of medication reconciliation of 501 patients in 79 clinical trials that found alarming levels of drug interactions: 

• 360 clinically relevant drug-drug interactions were identified among 189 patients, including 158 therapies that were prohibited by protocols. Of these, 57.7% involved cytochrome P450 enzymes, which are involved in metabolism. 

• Reconciliation revealed that 35.2% of medications were not otherwise known or documented.

• A median of 2 previously unknown therapies per patient was discovered in 74% of patients.

• Alternative medicine products such as supplements and over-the-counter drugs were implicated in 60% of identified drug interactions.

• Only 41% of oncologists discussed alternative medicine use with patients, which Hennes attributed to “lack of familiarity with many alternative medicine products or insufficient training.”

To make things more complicated, “We sometimes don’t know the full pharmacokinetic and pharmacodynamic profile of an investigational agent,” she said. 

Naming and Labeling May Not Be Standard

Investigational products may not have genetic names and instead have an alphanumeric identifier such as INV54826 that can be quite similar to other products, she said. Investigational drugs may even go through name changes, forcing pharmacists to be alerted to protect patients. 

In addition, labeling may not be standardized. Drugs may arrive unlabeled, with the wrong volume and size, and lack of barcoding. In some cases, pharmacists choose to put new, patient-friendly labels on these products, Hennes said. 

Information Distribution is Key

“Something that comes up in our practice quite a bit is that there’s no standard drug reference regarding investigational drugs,” Hennes said. “Finding ways to get key information to staff at the point of care is really critical to make sure we’re able to safely treat our patients.”

Precautions May Be Needed to Maintain Blinding Protocols 

Hennes explained that pharmacists must use opaque brown bag covers to maintain blinding when parenteral products have distinctive colors. Lines may have to be covered too, which can create challenges during administration. 

“Pumps aren’t meant to run lines that are covered,” she said, which can lead to jams. “If you don’t do education with your point of care staff, it can cause a lot of confusion.” 

It’s also important for blinding purposes to keep an eye on how long it takes to prepare a treatment, she said. A study’s integrity, for example, could be violated if a complex investigational product takes an hour to equilibrate to room temperature and 20-30 minutes to prepare, while a placebo only requires “drawing a few mils of saline out of a bag and labeling it.”

Education for Patients Can Be Useful 

Hennes urged colleagues to remind patients to save investigational medication at the end of each cycle and return it to the clinic site for accountability.

She also suggested creating treatment calendars/reminders for patients and discussing adherence strategies with them. 

Hennes reported no disclosures.