Can a nomogram foretell invasive pulmonary adenocarcinoma?

The diagnosis of solitary peripheral subsolid nodule carries with it an undefined risk of invasive pulmonary carcinoma, but clinicians have not had a tool that can help guide their planning for surgery. However, researchers in China have developed a nomogram that they said may aid clinicians to predict the risk of invasive pulmonary adenocarcinoma in these patients.

“Validation by the use of bootstrap resampling revealed optimal discrimination and calibration, indicating that the nomogram may have clinical utility,” said Chenghua Jin, MD, and Jinlin Cao, MD, of Zhejiang University, Hangzhou, China, and coauthors. They reported their findings in the February issue of the Journal of Thoracic and Cardiovascular Surgery (2017;153:42-9).

The nomogram accounts for the following factors: computed tomography attenuation; nodule size; spiculation; signs of vascular convergence; pleural tags; and solid proportion. “The nomogram showed a robust discrimination with an area under the receiver operating characteristic curve of 0.894,” Dr. Jin and coauthors reported. An area under the curve of 1 is equivalent to 100%, so the area under the curve this study reported shows close to 90% accuracy.

The study involved a retrospective analysis of 273 consecutive patients who had resection of a solitary peripheral subsolid nodule at Zhejiang University School of Medicine from January 2013 to December 2014. Subsolid pulmonary nodules include pure ground-glass nodules and part-solid nodules that feature both solid and ground-glass components. “The optimal management of patients with a subsolid nodule is of growing clinical concern, because the most common diagnosis for resected subsolid nodules is lung adenocarcinoma,” Dr. Jin and colleagues indicated.

Of the study population, 58% were diagnosed with invasive pulmonary adenocarcinoma. Other diagnoses within the group were benign (13%), atypical adenomatous hyperplasia (1%), adenocarcinoma in situ (6.5%) and minimally invasive adenocarcinoma (21%).

Results of the multivariable analyses showed that invasive pulmonary adenocarcinoma correlated with the following characteristics: lesion size; spiculation; vascular convergence; and pleural tag. Factors that were not significant included age, family history of lung cancer, CT attenuation, and solid proportion. However, the researchers did include CT attenuation, along with solid proportion, in the final regression analysis based on their contributions to the statistical analysis.

For the model, CT attenuation of –500 to –200 Hounsfield units carried an odds ratio of 1.690 (P = .228) while CT attenuation greater than –200 HU had an OR of 1.791 (P = .645). Positive spiculation had an OR of 3.312 (no P value given) and negative vascular convergence an OR of 0.300 (no P value given).

While a number of prediction models have been devised and validated to evaluate the likelihood of malignancy in pulmonary nodules, they have not given subsolid nodules “specific or detailed consideration,” Dr. Jin and and coauthors said. “To our knowledge, this study was the first to construct a quantitative nomogram to predict the probability of invasive pulmonary adenocarcinoma in patients with subsolid nodules,” the researchers wrote.

One limitation of the study is its selection bias toward patients with a greater probability of having a malignancy. Also, validation of the nomogram requires external analysis with additional databases from other countries and with more diverse ethnic groups. Another shortcoming is the retrospective nature of the study and a small number of patients who had positron emission tomography. “Further data collection, wider geographic recruitment, and incorporation of positron emission tomography results and some molecular factors could improve this model for future use,” Dr. Jin and coauthors concluded.

Dr. Jin and Dr. Cao had no relevant financial disclosures. The study received funding from the Zhejiang Province Science and Technology Plan.

The diagnosis of solitary peripheral subsolid nodule carries with it an undefined risk of invasive pulmonary carcinoma, but clinicians have not had a tool that can help guide their planning for surgery. However, researchers in China have developed a nomogram that they said may aid clinicians to predict the risk of invasive pulmonary adenocarcinoma in these patients.

“Validation by the use of bootstrap resampling revealed optimal discrimination and calibration, indicating that the nomogram may have clinical utility,” said Chenghua Jin, MD, and Jinlin Cao, MD, of Zhejiang University, Hangzhou, China, and coauthors. They reported their findings in the February issue of the Journal of Thoracic and Cardiovascular Surgery (2017;153:42-9).

The nomogram accounts for the following factors: computed tomography attenuation; nodule size; spiculation; signs of vascular convergence; pleural tags; and solid proportion. “The nomogram showed a robust discrimination with an area under the receiver operating characteristic curve of 0.894,” Dr. Jin and coauthors reported. An area under the curve of 1 is equivalent to 100%, so the area under the curve this study reported shows close to 90% accuracy.

The study involved a retrospective analysis of 273 consecutive patients who had resection of a solitary peripheral subsolid nodule at Zhejiang University School of Medicine from January 2013 to December 2014. Subsolid pulmonary nodules include pure ground-glass nodules and part-solid nodules that feature both solid and ground-glass components. “The optimal management of patients with a subsolid nodule is of growing clinical concern, because the most common diagnosis for resected subsolid nodules is lung adenocarcinoma,” Dr. Jin and colleagues indicated.

Of the study population, 58% were diagnosed with invasive pulmonary adenocarcinoma. Other diagnoses within the group were benign (13%), atypical adenomatous hyperplasia (1%), adenocarcinoma in situ (6.5%) and minimally invasive adenocarcinoma (21%).

Results of the multivariable analyses showed that invasive pulmonary adenocarcinoma correlated with the following characteristics: lesion size; spiculation; vascular convergence; and pleural tag. Factors that were not significant included age, family history of lung cancer, CT attenuation, and solid proportion. However, the researchers did include CT attenuation, along with solid proportion, in the final regression analysis based on their contributions to the statistical analysis.

For the model, CT attenuation of –500 to –200 Hounsfield units carried an odds ratio of 1.690 (P = .228) while CT attenuation greater than –200 HU had an OR of 1.791 (P = .645). Positive spiculation had an OR of 3.312 (no P value given) and negative vascular convergence an OR of 0.300 (no P value given).

While a number of prediction models have been devised and validated to evaluate the likelihood of malignancy in pulmonary nodules, they have not given subsolid nodules “specific or detailed consideration,” Dr. Jin and and coauthors said. “To our knowledge, this study was the first to construct a quantitative nomogram to predict the probability of invasive pulmonary adenocarcinoma in patients with subsolid nodules,” the researchers wrote.

One limitation of the study is its selection bias toward patients with a greater probability of having a malignancy. Also, validation of the nomogram requires external analysis with additional databases from other countries and with more diverse ethnic groups. Another shortcoming is the retrospective nature of the study and a small number of patients who had positron emission tomography. “Further data collection, wider geographic recruitment, and incorporation of positron emission tomography results and some molecular factors could improve this model for future use,” Dr. Jin and coauthors concluded.

Dr. Jin and Dr. Cao had no relevant financial disclosures. The study received funding from the Zhejiang Province Science and Technology Plan.

The diagnosis of solitary peripheral subsolid nodule carries with it an undefined risk of invasive pulmonary carcinoma, but clinicians have not had a tool that can help guide their planning for surgery. However, researchers in China have developed a nomogram that they said may aid clinicians to predict the risk of invasive pulmonary adenocarcinoma in these patients.

“Validation by the use of bootstrap resampling revealed optimal discrimination and calibration, indicating that the nomogram may have clinical utility,” said Chenghua Jin, MD, and Jinlin Cao, MD, of Zhejiang University, Hangzhou, China, and coauthors. They reported their findings in the February issue of the Journal of Thoracic and Cardiovascular Surgery (2017;153:42-9).

The nomogram accounts for the following factors: computed tomography attenuation; nodule size; spiculation; signs of vascular convergence; pleural tags; and solid proportion. “The nomogram showed a robust discrimination with an area under the receiver operating characteristic curve of 0.894,” Dr. Jin and coauthors reported. An area under the curve of 1 is equivalent to 100%, so the area under the curve this study reported shows close to 90% accuracy.

The study involved a retrospective analysis of 273 consecutive patients who had resection of a solitary peripheral subsolid nodule at Zhejiang University School of Medicine from January 2013 to December 2014. Subsolid pulmonary nodules include pure ground-glass nodules and part-solid nodules that feature both solid and ground-glass components. “The optimal management of patients with a subsolid nodule is of growing clinical concern, because the most common diagnosis for resected subsolid nodules is lung adenocarcinoma,” Dr. Jin and colleagues indicated.

Of the study population, 58% were diagnosed with invasive pulmonary adenocarcinoma. Other diagnoses within the group were benign (13%), atypical adenomatous hyperplasia (1%), adenocarcinoma in situ (6.5%) and minimally invasive adenocarcinoma (21%).

Results of the multivariable analyses showed that invasive pulmonary adenocarcinoma correlated with the following characteristics: lesion size; spiculation; vascular convergence; and pleural tag. Factors that were not significant included age, family history of lung cancer, CT attenuation, and solid proportion. However, the researchers did include CT attenuation, along with solid proportion, in the final regression analysis based on their contributions to the statistical analysis.

For the model, CT attenuation of –500 to –200 Hounsfield units carried an odds ratio of 1.690 (P = .228) while CT attenuation greater than –200 HU had an OR of 1.791 (P = .645). Positive spiculation had an OR of 3.312 (no P value given) and negative vascular convergence an OR of 0.300 (no P value given).

While a number of prediction models have been devised and validated to evaluate the likelihood of malignancy in pulmonary nodules, they have not given subsolid nodules “specific or detailed consideration,” Dr. Jin and and coauthors said. “To our knowledge, this study was the first to construct a quantitative nomogram to predict the probability of invasive pulmonary adenocarcinoma in patients with subsolid nodules,” the researchers wrote.

One limitation of the study is its selection bias toward patients with a greater probability of having a malignancy. Also, validation of the nomogram requires external analysis with additional databases from other countries and with more diverse ethnic groups. Another shortcoming is the retrospective nature of the study and a small number of patients who had positron emission tomography. “Further data collection, wider geographic recruitment, and incorporation of positron emission tomography results and some molecular factors could improve this model for future use,” Dr. Jin and coauthors concluded.

Dr. Jin and Dr. Cao had no relevant financial disclosures. The study received funding from the Zhejiang Province Science and Technology Plan.