TBD Meeting (5421-23)

A tool that incorporates lupus-related variables with traditional risk factors provides a much more accurate assessment of cardiovascular (CV) risk in patients with systemic lupus erythematosus (SLE), according to data presented at the annual meeting of the Canadian Rheumatology Association.

In the initial clinical assessment of this tool, called the SLECRISK, “it identified high-risk lupus patients who would otherwise be missed by traditional methods of CV risk assessment,” reported May Y. Choi, MD, associate director of translational research at the University of Calgary’s (Alta.) Lupus Centre of Excellence.

bowdenimages/iStock/Getty Images

It is well known that patients with SLE face an increased risk of CV events starting at an age long before risk begins climbing in the general population, according to Dr. Choi. She cited one study that showed women aged 35-44 years have a 50-fold greater risk of myocardial infarction than healthy individuals.

All major guidelines recognize this increased risk and recommend CV risk assessment in patients with SLE, even though Dr. Choi pointed out that traditional tools, such as the American College of Cardiology atherosclerotic cardiovascular disease (ASCVD) risk calculator or the Framingham Risk Score (FRS) have a limited ability to detect the patients with SLE who are most likely to have an event.
 

In SLE, current tools are inadequate

“These risk assessment tools perform poorly in SLE patients because they do not capture SLE-related inflammation,” Dr. Choi said. Of several examples, Dr. Choi cited a study showing “seven times more MIs and strokes observed than expected in SLE patients on the basis of the FRS.”

The disparity between expected and observed MIs and strokes is worse with increasing severity of SLE. In a study she presented 3 years ago, rates of CV events were 12 times higher in those with inactive or mild SLE, rising to a 16-fold increase among those with moderate disease and jumping to a 32-fold increase in those with severe SLE.

The SLECRISK tool was developed from the Brigham and Women’s Hospital SLE Registry, which was initiated in 1992. Patients without a history of CV disease were evaluated for traditional CV risk factors and for SLE-specific characteristics such as disease activity, levels of the complement proteins C3 and C4, kidney function, the presence of nephritis, and SLE duration. The value of these characteristics as predictors of CV events were then assessed over a 10-year follow-up period before being assembled into the SLECRISK tool.

In an example of the risk equation, Dr. Choi described a 50-year-old patient with SLE and a 5% 10-year ASCVD risk score, which is low. After adjustment for SLE risks, which included 10 years disease duration, high disease activity, elevated creatinine, and positive anti–double stranded DNA status, the 10-year CV risk score climbed to 16.2%, which is moderate.

The performance of the SLECRISK was evaluated in 1,243 patients providing 8,946.51 person-years of follow-up. During this period, there were 90 major adverse cardiac events (MACE), of which 82% were adjudicated by cardiologists, and 211 secondary events.

Relative to the ASCVD risk score, the SLECRISK identified about twice as many patients with SLE as having moderate risk and 3.5-fold more patients as having high risk. Among patients who experienced CV events, traditional CV risk factors were more common but so were SLE-specific risk factors, including greater disease severity, a greater likelihood of lupus nephritis, increased complement levels, and greater exposure to glucocorticoids, according to Dr. Choi.

Specificities for CV events higher on SLECRISK

In predicting CV events, the differences in specificities were in the same general range, although somewhat higher for the ASCVD risk score in regard to predicting MACE (83% vs. 72%) and MACE plus secondary events (90% vs. 79%). However, the sensitivities were much higher for SLECRISK relative to the ASCVD risk score for MACE alone (64% vs. 41%) and for MACE plus secondary events (58% vs. 35%).

When comparing those who had an MI or stroke, the ASCVD risk score identified 8 (7%) patients missed by SLECRISK, whereas SLECRISK identified 89 (73%) missed by the ASCVD risk score. The remaining 25 patients (20%) were identified by both. The advantage of SLECRISK was similar for MACE plus secondary outcomes.

Dr. Choi noted that all of the SLE-specific variables in SLECRISK are readily obtained and often already available in patient charts. She said that there is a plan to validate the tool in larger groups, but with a goal of creating a tool available online for clinicians and their patients to use. There is also an even more ambitious plan for the future.

“We have funding to look at machine learning to evaluate predictive variables in SLE patients,” Dr. Choi said. Rather than adding SLE-specific variables to traditional risks, the plan is to “start from scratch,” letting artificial intelligence assemble predictors without prejudice to what might or might not be relevant.

A SLE-specific tool for evaluating CV risk is an important “unmet need,” according to Karen H. Costenbader, MD, professor in the division of rheumatology, inflammation, and immunity at Brigham and Women’s Hospital and Harvard Medical School, both in Boston. In an interview, she reiterated that measuring CV risk in SLE is already guideline recommended, but conventional tools have been shown to be inaccurate.

“I can envision it being used in clinical encounters to help guide shared decision-making with patients,” explained Dr. Costenbader, who was not involved in the presentation at the CRA meeting but worked with Dr. Choi in developing SLECRISK. “It would give us more precise estimates, allowing us to risk stratify our patients and informing us as to which modifiable SLE-specific and nonspecific factors are contributing most to CV risk.’

The problem of using conventional risk assessments in SLE has been well recognized. Of those who have written on this subject, Maureen McMahon, MD, site director of the Lupus Clinical Trials Network at the University of California, Los Angeles, said: “There is a critical need for the development of SLE-specific risk assessment tools like SLECRISK.”

Author of several studies looking at alternatives for CV risk assessment in SLE, including a study looking at a panel of biomarkers that was published in ACR Open Rheumatology, Dr. McMahon said in an interview that CV risk in SLE is high but conventional risk assessments are flawed.

“Multiple previous studies have demonstrated that these currently available calculators are not adequate for identifying risk in the lupus patient population,” she said. According to Dr. McMahon, the fact that rheumatologists remain “dependent upon [these conventional] cardiovascular risk calculators” is a well-recognized problem that needs resolution.

Dr. Choi has financial relationships with AstraZeneca, GlaxoSmithKline, Mallinckrodt. MitogenDx, Organon, and Werfen International. Dr. Costenbader reports no potential conflicts of interest. Dr. McMahon has financial relationships with AstraZeneca, Aurinia Pharmaceuticals, Eli Lilly, and GlaxoSmithKline.

A tool that incorporates lupus-related variables with traditional risk factors provides a much more accurate assessment of cardiovascular (CV) risk in patients with systemic lupus erythematosus (SLE), according to data presented at the annual meeting of the Canadian Rheumatology Association.

In the initial clinical assessment of this tool, called the SLECRISK, “it identified high-risk lupus patients who would otherwise be missed by traditional methods of CV risk assessment,” reported May Y. Choi, MD, associate director of translational research at the University of Calgary’s (Alta.) Lupus Centre of Excellence.

bowdenimages/iStock/Getty Images

It is well known that patients with SLE face an increased risk of CV events starting at an age long before risk begins climbing in the general population, according to Dr. Choi. She cited one study that showed women aged 35-44 years have a 50-fold greater risk of myocardial infarction than healthy individuals.

All major guidelines recognize this increased risk and recommend CV risk assessment in patients with SLE, even though Dr. Choi pointed out that traditional tools, such as the American College of Cardiology atherosclerotic cardiovascular disease (ASCVD) risk calculator or the Framingham Risk Score (FRS) have a limited ability to detect the patients with SLE who are most likely to have an event.
 

In SLE, current tools are inadequate

“These risk assessment tools perform poorly in SLE patients because they do not capture SLE-related inflammation,” Dr. Choi said. Of several examples, Dr. Choi cited a study showing “seven times more MIs and strokes observed than expected in SLE patients on the basis of the FRS.”

The disparity between expected and observed MIs and strokes is worse with increasing severity of SLE. In a study she presented 3 years ago, rates of CV events were 12 times higher in those with inactive or mild SLE, rising to a 16-fold increase among those with moderate disease and jumping to a 32-fold increase in those with severe SLE.

The SLECRISK tool was developed from the Brigham and Women’s Hospital SLE Registry, which was initiated in 1992. Patients without a history of CV disease were evaluated for traditional CV risk factors and for SLE-specific characteristics such as disease activity, levels of the complement proteins C3 and C4, kidney function, the presence of nephritis, and SLE duration. The value of these characteristics as predictors of CV events were then assessed over a 10-year follow-up period before being assembled into the SLECRISK tool.

In an example of the risk equation, Dr. Choi described a 50-year-old patient with SLE and a 5% 10-year ASCVD risk score, which is low. After adjustment for SLE risks, which included 10 years disease duration, high disease activity, elevated creatinine, and positive anti–double stranded DNA status, the 10-year CV risk score climbed to 16.2%, which is moderate.

The performance of the SLECRISK was evaluated in 1,243 patients providing 8,946.51 person-years of follow-up. During this period, there were 90 major adverse cardiac events (MACE), of which 82% were adjudicated by cardiologists, and 211 secondary events.

Relative to the ASCVD risk score, the SLECRISK identified about twice as many patients with SLE as having moderate risk and 3.5-fold more patients as having high risk. Among patients who experienced CV events, traditional CV risk factors were more common but so were SLE-specific risk factors, including greater disease severity, a greater likelihood of lupus nephritis, increased complement levels, and greater exposure to glucocorticoids, according to Dr. Choi.

Specificities for CV events higher on SLECRISK

In predicting CV events, the differences in specificities were in the same general range, although somewhat higher for the ASCVD risk score in regard to predicting MACE (83% vs. 72%) and MACE plus secondary events (90% vs. 79%). However, the sensitivities were much higher for SLECRISK relative to the ASCVD risk score for MACE alone (64% vs. 41%) and for MACE plus secondary events (58% vs. 35%).

When comparing those who had an MI or stroke, the ASCVD risk score identified 8 (7%) patients missed by SLECRISK, whereas SLECRISK identified 89 (73%) missed by the ASCVD risk score. The remaining 25 patients (20%) were identified by both. The advantage of SLECRISK was similar for MACE plus secondary outcomes.

Dr. Choi noted that all of the SLE-specific variables in SLECRISK are readily obtained and often already available in patient charts. She said that there is a plan to validate the tool in larger groups, but with a goal of creating a tool available online for clinicians and their patients to use. There is also an even more ambitious plan for the future.

“We have funding to look at machine learning to evaluate predictive variables in SLE patients,” Dr. Choi said. Rather than adding SLE-specific variables to traditional risks, the plan is to “start from scratch,” letting artificial intelligence assemble predictors without prejudice to what might or might not be relevant.

A SLE-specific tool for evaluating CV risk is an important “unmet need,” according to Karen H. Costenbader, MD, professor in the division of rheumatology, inflammation, and immunity at Brigham and Women’s Hospital and Harvard Medical School, both in Boston. In an interview, she reiterated that measuring CV risk in SLE is already guideline recommended, but conventional tools have been shown to be inaccurate.

“I can envision it being used in clinical encounters to help guide shared decision-making with patients,” explained Dr. Costenbader, who was not involved in the presentation at the CRA meeting but worked with Dr. Choi in developing SLECRISK. “It would give us more precise estimates, allowing us to risk stratify our patients and informing us as to which modifiable SLE-specific and nonspecific factors are contributing most to CV risk.’

The problem of using conventional risk assessments in SLE has been well recognized. Of those who have written on this subject, Maureen McMahon, MD, site director of the Lupus Clinical Trials Network at the University of California, Los Angeles, said: “There is a critical need for the development of SLE-specific risk assessment tools like SLECRISK.”

Author of several studies looking at alternatives for CV risk assessment in SLE, including a study looking at a panel of biomarkers that was published in ACR Open Rheumatology, Dr. McMahon said in an interview that CV risk in SLE is high but conventional risk assessments are flawed.

“Multiple previous studies have demonstrated that these currently available calculators are not adequate for identifying risk in the lupus patient population,” she said. According to Dr. McMahon, the fact that rheumatologists remain “dependent upon [these conventional] cardiovascular risk calculators” is a well-recognized problem that needs resolution.

Dr. Choi has financial relationships with AstraZeneca, GlaxoSmithKline, Mallinckrodt. MitogenDx, Organon, and Werfen International. Dr. Costenbader reports no potential conflicts of interest. Dr. McMahon has financial relationships with AstraZeneca, Aurinia Pharmaceuticals, Eli Lilly, and GlaxoSmithKline.

A tool that incorporates lupus-related variables with traditional risk factors provides a much more accurate assessment of cardiovascular (CV) risk in patients with systemic lupus erythematosus (SLE), according to data presented at the annual meeting of the Canadian Rheumatology Association.

In the initial clinical assessment of this tool, called the SLECRISK, “it identified high-risk lupus patients who would otherwise be missed by traditional methods of CV risk assessment,” reported May Y. Choi, MD, associate director of translational research at the University of Calgary’s (Alta.) Lupus Centre of Excellence.

bowdenimages/iStock/Getty Images

It is well known that patients with SLE face an increased risk of CV events starting at an age long before risk begins climbing in the general population, according to Dr. Choi. She cited one study that showed women aged 35-44 years have a 50-fold greater risk of myocardial infarction than healthy individuals.

All major guidelines recognize this increased risk and recommend CV risk assessment in patients with SLE, even though Dr. Choi pointed out that traditional tools, such as the American College of Cardiology atherosclerotic cardiovascular disease (ASCVD) risk calculator or the Framingham Risk Score (FRS) have a limited ability to detect the patients with SLE who are most likely to have an event.
 

In SLE, current tools are inadequate

“These risk assessment tools perform poorly in SLE patients because they do not capture SLE-related inflammation,” Dr. Choi said. Of several examples, Dr. Choi cited a study showing “seven times more MIs and strokes observed than expected in SLE patients on the basis of the FRS.”

The disparity between expected and observed MIs and strokes is worse with increasing severity of SLE. In a study she presented 3 years ago, rates of CV events were 12 times higher in those with inactive or mild SLE, rising to a 16-fold increase among those with moderate disease and jumping to a 32-fold increase in those with severe SLE.

The SLECRISK tool was developed from the Brigham and Women’s Hospital SLE Registry, which was initiated in 1992. Patients without a history of CV disease were evaluated for traditional CV risk factors and for SLE-specific characteristics such as disease activity, levels of the complement proteins C3 and C4, kidney function, the presence of nephritis, and SLE duration. The value of these characteristics as predictors of CV events were then assessed over a 10-year follow-up period before being assembled into the SLECRISK tool.

In an example of the risk equation, Dr. Choi described a 50-year-old patient with SLE and a 5% 10-year ASCVD risk score, which is low. After adjustment for SLE risks, which included 10 years disease duration, high disease activity, elevated creatinine, and positive anti–double stranded DNA status, the 10-year CV risk score climbed to 16.2%, which is moderate.

The performance of the SLECRISK was evaluated in 1,243 patients providing 8,946.51 person-years of follow-up. During this period, there were 90 major adverse cardiac events (MACE), of which 82% were adjudicated by cardiologists, and 211 secondary events.

Relative to the ASCVD risk score, the SLECRISK identified about twice as many patients with SLE as having moderate risk and 3.5-fold more patients as having high risk. Among patients who experienced CV events, traditional CV risk factors were more common but so were SLE-specific risk factors, including greater disease severity, a greater likelihood of lupus nephritis, increased complement levels, and greater exposure to glucocorticoids, according to Dr. Choi.

Specificities for CV events higher on SLECRISK

In predicting CV events, the differences in specificities were in the same general range, although somewhat higher for the ASCVD risk score in regard to predicting MACE (83% vs. 72%) and MACE plus secondary events (90% vs. 79%). However, the sensitivities were much higher for SLECRISK relative to the ASCVD risk score for MACE alone (64% vs. 41%) and for MACE plus secondary events (58% vs. 35%).

When comparing those who had an MI or stroke, the ASCVD risk score identified 8 (7%) patients missed by SLECRISK, whereas SLECRISK identified 89 (73%) missed by the ASCVD risk score. The remaining 25 patients (20%) were identified by both. The advantage of SLECRISK was similar for MACE plus secondary outcomes.

Dr. Choi noted that all of the SLE-specific variables in SLECRISK are readily obtained and often already available in patient charts. She said that there is a plan to validate the tool in larger groups, but with a goal of creating a tool available online for clinicians and their patients to use. There is also an even more ambitious plan for the future.

“We have funding to look at machine learning to evaluate predictive variables in SLE patients,” Dr. Choi said. Rather than adding SLE-specific variables to traditional risks, the plan is to “start from scratch,” letting artificial intelligence assemble predictors without prejudice to what might or might not be relevant.

A SLE-specific tool for evaluating CV risk is an important “unmet need,” according to Karen H. Costenbader, MD, professor in the division of rheumatology, inflammation, and immunity at Brigham and Women’s Hospital and Harvard Medical School, both in Boston. In an interview, she reiterated that measuring CV risk in SLE is already guideline recommended, but conventional tools have been shown to be inaccurate.

“I can envision it being used in clinical encounters to help guide shared decision-making with patients,” explained Dr. Costenbader, who was not involved in the presentation at the CRA meeting but worked with Dr. Choi in developing SLECRISK. “It would give us more precise estimates, allowing us to risk stratify our patients and informing us as to which modifiable SLE-specific and nonspecific factors are contributing most to CV risk.’

The problem of using conventional risk assessments in SLE has been well recognized. Of those who have written on this subject, Maureen McMahon, MD, site director of the Lupus Clinical Trials Network at the University of California, Los Angeles, said: “There is a critical need for the development of SLE-specific risk assessment tools like SLECRISK.”

Author of several studies looking at alternatives for CV risk assessment in SLE, including a study looking at a panel of biomarkers that was published in ACR Open Rheumatology, Dr. McMahon said in an interview that CV risk in SLE is high but conventional risk assessments are flawed.

“Multiple previous studies have demonstrated that these currently available calculators are not adequate for identifying risk in the lupus patient population,” she said. According to Dr. McMahon, the fact that rheumatologists remain “dependent upon [these conventional] cardiovascular risk calculators” is a well-recognized problem that needs resolution.

Dr. Choi has financial relationships with AstraZeneca, GlaxoSmithKline, Mallinckrodt. MitogenDx, Organon, and Werfen International. Dr. Costenbader reports no potential conflicts of interest. Dr. McMahon has financial relationships with AstraZeneca, Aurinia Pharmaceuticals, Eli Lilly, and GlaxoSmithKline.

A new tool for predicting which patients with psoriasis will develop psoriatic arthritis (PsA) is showing promise for such clinical applications as early treatment in those at risk or trials to prevent PsA, according to a summary of progress at the annual meeting of the Canadian Rheumatology Association.

Based on current levels of sensitivity and specificity, psoriasis “can be predicted with reasonable accuracy,” reported Lihi Eder, MD, PhD, director of research in the rheumatology division at the University of Toronto.

The predictive method, called PRESTO (Prediction of Psoriatic Arthritis Tool), is based on variables readily available in clinical practice, according to Dr. Eder. Once values are assigned to the risk factors, the risk of PsA over a 1-year or 5-year time frame can be estimated with a calculator.

She called PRESTO the “first clinical tool for predicting PsA among psoriasis patients.”

The work on this tool began in 2006 when the International Psoriasis and Arthritis Research Team (IPART) initiated a prospectively collected cohort of psoriasis patients. To be enrolled, patients had to be free of signs and symptoms of arthritis upon examination by a rheumatologist. They were then invited to return annually for follow-up that again included screening for joint involvement by a rheumatologist.

At baseline and at follow-up evaluations, 13 predictors were evaluated. These involved psoriasis characteristics, such as nail pitting; symptoms, such as stiffness; comorbidities, such as additional inflammatory diseases; and laboratory values, such as upregulated markers of inflammation.

Symptoms and signs used to predict PsA

Dr. Eder and her colleagues applied regression models to select an optimal combination of variables weighted for predictive value. Variables offering predictive value included higher PASI (Psoriasis Area and Severity Index), greater fatigue score as measured by FACIT (Functional Assessment of Chronic Illness Therapy) score, greater morning stiffness, and greater pain.

When applied to 635 patients in the IPART cohort, in which there were 51 incident PsA cases over 1 year and 75 incident cases over 5 years, the area under the curve (AUC) for PRESTO at the cutoffs studied was 72% for the 1-year time window and 75% for the 5-year time window.

These levels are associated with adequate accuracy, according to Dr. Eder, who explained that “an AUC greater than 70% is considered reasonable” for clinical applicability.

Moreover, the cutoffs can be adjusted for the specific purpose of the predictive tool. For example, to screen patients for risk, lower cutoffs could be employed to increase sensitivity. In order to select patients for a clinical trial to prevent PsA, higher cutoffs could be employed to increase specificity.

But sensitivities and specificities move in opposite directions when cutoffs are adjusted. Showing data from the 5-year prediction model, Dr. Eder reported that specificities climbed from about 58% to 97% as cutoffs were increased. The sensitivities with these adjustments fell from 79% to 14%.

In general, Dr. Eder said there was “excellent calibration” for the cutoffs employed when they compared the predicted and observed rates of PsA according to quintile of predictive probability. The differences were particularly minor over a 1-year time period. Over the 5-year period, observed rates were somewhat higher than predicted in the fourth and fifth quintile, but, again, this discrepancy could be modified for specific applications with cutoff adjustments.

Validation studies are planned

Even though psoriasis patients in IPART represents one of the largest cohorts of prospectively collected psoriasis patients, Dr. Eder acknowledged that the sample size would be considered “moderate” for developing a predictive model. However, the fact that the data were collected prospectively using standardized methodology strengthens the findings and provides the basis for the next step.

“Validation studies are planned with external cohorts,” said Dr. Eder, who indicated that a viable tool for identifying psoriasis patients at risk for PsA is likely. Even if it is not employed routinely in its current form at the level of individual patient care, she predicted that it will have value at a research level for understanding the relationship of psoriasis to PsA.

Christopher T. Ritchlin, MD, a professor and researcher at the University of Rochester (N.Y.), agreed that PRESTO has important potential as a clinical tool. Dr. Ritchlin has been involved in the development of PRESTO but was not involved in the presentation made at the CRA annual meeting.

“The PRESTO tool has the ability to predict the 2- and 5-year risk of developing psoriatic arthritis, which is an important advance if confirmed,” he said in an interview. He pointed out that approximately 25%-30% who develop psoriasis will go on to develop PsA but until now there has been no way to identify them.

“This tool may provide a pathway to early intervention,” he said.

Dr. Eder has financial relationships with AbbVie, Eli Lilly, Fresenius Kabi, Janssen, Novartis, Pfizer, Sandoz, and UCB. Dr. Ritchlin has financial relationships with many of the same companies.

A new tool for predicting which patients with psoriasis will develop psoriatic arthritis (PsA) is showing promise for such clinical applications as early treatment in those at risk or trials to prevent PsA, according to a summary of progress at the annual meeting of the Canadian Rheumatology Association.

Based on current levels of sensitivity and specificity, psoriasis “can be predicted with reasonable accuracy,” reported Lihi Eder, MD, PhD, director of research in the rheumatology division at the University of Toronto.

The predictive method, called PRESTO (Prediction of Psoriatic Arthritis Tool), is based on variables readily available in clinical practice, according to Dr. Eder. Once values are assigned to the risk factors, the risk of PsA over a 1-year or 5-year time frame can be estimated with a calculator.

She called PRESTO the “first clinical tool for predicting PsA among psoriasis patients.”

The work on this tool began in 2006 when the International Psoriasis and Arthritis Research Team (IPART) initiated a prospectively collected cohort of psoriasis patients. To be enrolled, patients had to be free of signs and symptoms of arthritis upon examination by a rheumatologist. They were then invited to return annually for follow-up that again included screening for joint involvement by a rheumatologist.

At baseline and at follow-up evaluations, 13 predictors were evaluated. These involved psoriasis characteristics, such as nail pitting; symptoms, such as stiffness; comorbidities, such as additional inflammatory diseases; and laboratory values, such as upregulated markers of inflammation.

Symptoms and signs used to predict PsA

Dr. Eder and her colleagues applied regression models to select an optimal combination of variables weighted for predictive value. Variables offering predictive value included higher PASI (Psoriasis Area and Severity Index), greater fatigue score as measured by FACIT (Functional Assessment of Chronic Illness Therapy) score, greater morning stiffness, and greater pain.

When applied to 635 patients in the IPART cohort, in which there were 51 incident PsA cases over 1 year and 75 incident cases over 5 years, the area under the curve (AUC) for PRESTO at the cutoffs studied was 72% for the 1-year time window and 75% for the 5-year time window.

These levels are associated with adequate accuracy, according to Dr. Eder, who explained that “an AUC greater than 70% is considered reasonable” for clinical applicability.

Moreover, the cutoffs can be adjusted for the specific purpose of the predictive tool. For example, to screen patients for risk, lower cutoffs could be employed to increase sensitivity. In order to select patients for a clinical trial to prevent PsA, higher cutoffs could be employed to increase specificity.

But sensitivities and specificities move in opposite directions when cutoffs are adjusted. Showing data from the 5-year prediction model, Dr. Eder reported that specificities climbed from about 58% to 97% as cutoffs were increased. The sensitivities with these adjustments fell from 79% to 14%.

In general, Dr. Eder said there was “excellent calibration” for the cutoffs employed when they compared the predicted and observed rates of PsA according to quintile of predictive probability. The differences were particularly minor over a 1-year time period. Over the 5-year period, observed rates were somewhat higher than predicted in the fourth and fifth quintile, but, again, this discrepancy could be modified for specific applications with cutoff adjustments.

Validation studies are planned

Even though psoriasis patients in IPART represents one of the largest cohorts of prospectively collected psoriasis patients, Dr. Eder acknowledged that the sample size would be considered “moderate” for developing a predictive model. However, the fact that the data were collected prospectively using standardized methodology strengthens the findings and provides the basis for the next step.

“Validation studies are planned with external cohorts,” said Dr. Eder, who indicated that a viable tool for identifying psoriasis patients at risk for PsA is likely. Even if it is not employed routinely in its current form at the level of individual patient care, she predicted that it will have value at a research level for understanding the relationship of psoriasis to PsA.

Christopher T. Ritchlin, MD, a professor and researcher at the University of Rochester (N.Y.), agreed that PRESTO has important potential as a clinical tool. Dr. Ritchlin has been involved in the development of PRESTO but was not involved in the presentation made at the CRA annual meeting.

“The PRESTO tool has the ability to predict the 2- and 5-year risk of developing psoriatic arthritis, which is an important advance if confirmed,” he said in an interview. He pointed out that approximately 25%-30% who develop psoriasis will go on to develop PsA but until now there has been no way to identify them.

“This tool may provide a pathway to early intervention,” he said.

Dr. Eder has financial relationships with AbbVie, Eli Lilly, Fresenius Kabi, Janssen, Novartis, Pfizer, Sandoz, and UCB. Dr. Ritchlin has financial relationships with many of the same companies.

A new tool for predicting which patients with psoriasis will develop psoriatic arthritis (PsA) is showing promise for such clinical applications as early treatment in those at risk or trials to prevent PsA, according to a summary of progress at the annual meeting of the Canadian Rheumatology Association.

Based on current levels of sensitivity and specificity, psoriasis “can be predicted with reasonable accuracy,” reported Lihi Eder, MD, PhD, director of research in the rheumatology division at the University of Toronto.

The predictive method, called PRESTO (Prediction of Psoriatic Arthritis Tool), is based on variables readily available in clinical practice, according to Dr. Eder. Once values are assigned to the risk factors, the risk of PsA over a 1-year or 5-year time frame can be estimated with a calculator.

She called PRESTO the “first clinical tool for predicting PsA among psoriasis patients.”

The work on this tool began in 2006 when the International Psoriasis and Arthritis Research Team (IPART) initiated a prospectively collected cohort of psoriasis patients. To be enrolled, patients had to be free of signs and symptoms of arthritis upon examination by a rheumatologist. They were then invited to return annually for follow-up that again included screening for joint involvement by a rheumatologist.

At baseline and at follow-up evaluations, 13 predictors were evaluated. These involved psoriasis characteristics, such as nail pitting; symptoms, such as stiffness; comorbidities, such as additional inflammatory diseases; and laboratory values, such as upregulated markers of inflammation.

Symptoms and signs used to predict PsA

Dr. Eder and her colleagues applied regression models to select an optimal combination of variables weighted for predictive value. Variables offering predictive value included higher PASI (Psoriasis Area and Severity Index), greater fatigue score as measured by FACIT (Functional Assessment of Chronic Illness Therapy) score, greater morning stiffness, and greater pain.

When applied to 635 patients in the IPART cohort, in which there were 51 incident PsA cases over 1 year and 75 incident cases over 5 years, the area under the curve (AUC) for PRESTO at the cutoffs studied was 72% for the 1-year time window and 75% for the 5-year time window.

These levels are associated with adequate accuracy, according to Dr. Eder, who explained that “an AUC greater than 70% is considered reasonable” for clinical applicability.

Moreover, the cutoffs can be adjusted for the specific purpose of the predictive tool. For example, to screen patients for risk, lower cutoffs could be employed to increase sensitivity. In order to select patients for a clinical trial to prevent PsA, higher cutoffs could be employed to increase specificity.

But sensitivities and specificities move in opposite directions when cutoffs are adjusted. Showing data from the 5-year prediction model, Dr. Eder reported that specificities climbed from about 58% to 97% as cutoffs were increased. The sensitivities with these adjustments fell from 79% to 14%.

In general, Dr. Eder said there was “excellent calibration” for the cutoffs employed when they compared the predicted and observed rates of PsA according to quintile of predictive probability. The differences were particularly minor over a 1-year time period. Over the 5-year period, observed rates were somewhat higher than predicted in the fourth and fifth quintile, but, again, this discrepancy could be modified for specific applications with cutoff adjustments.

Validation studies are planned

Even though psoriasis patients in IPART represents one of the largest cohorts of prospectively collected psoriasis patients, Dr. Eder acknowledged that the sample size would be considered “moderate” for developing a predictive model. However, the fact that the data were collected prospectively using standardized methodology strengthens the findings and provides the basis for the next step.

“Validation studies are planned with external cohorts,” said Dr. Eder, who indicated that a viable tool for identifying psoriasis patients at risk for PsA is likely. Even if it is not employed routinely in its current form at the level of individual patient care, she predicted that it will have value at a research level for understanding the relationship of psoriasis to PsA.

Christopher T. Ritchlin, MD, a professor and researcher at the University of Rochester (N.Y.), agreed that PRESTO has important potential as a clinical tool. Dr. Ritchlin has been involved in the development of PRESTO but was not involved in the presentation made at the CRA annual meeting.

“The PRESTO tool has the ability to predict the 2- and 5-year risk of developing psoriatic arthritis, which is an important advance if confirmed,” he said in an interview. He pointed out that approximately 25%-30% who develop psoriasis will go on to develop PsA but until now there has been no way to identify them.

“This tool may provide a pathway to early intervention,” he said.

Dr. Eder has financial relationships with AbbVie, Eli Lilly, Fresenius Kabi, Janssen, Novartis, Pfizer, Sandoz, and UCB. Dr. Ritchlin has financial relationships with many of the same companies.

At the time of a nephritis flare in patients with systemic lupus erythematosus (SLE), elevated levels of two neutrophil extracellular trap (NET) protein complexes, elastase-DNA and HMGB1-DNA, predict declining renal function, poor response to therapy, and adverse renal outcomes, according to work presented at the annual meeting of the Canadian Rheumatology Association.

“These proteins are not only predominantly elevated in patients with proliferative lupus nephritis, but they correlate with adverse renal outcomes when patients are followed over 24 months,” reported Laura P. Whittall-Garcia, MD, a clinical fellow in rheumatology at the University of Toronto.

Lupus nephritis is common in SLE, developing in about 50% of patients, according to Dr. Whittall-Garcia. Of these, up to 20% will not respond to standard therapies, typically resulting in end-stage renal disease. Up until now, there has been no reliable method of predicting this adverse clinical course.
 

Proteins identified in NETs

The series of studies conducted by Dr. Whittall-Garcia and coinvestigators were focused on NETs, a network of strings of DNA that typically bind pathogenic microbes to prevent infection but can participate in the pathology of immune-mediated conditions. As Dr. Whittall-Garcia explained, DNA extruded from NETs has been a source of autoantigens.

Based on earlier work, they pursued the hypothesis that high mobility group box 1 (HMGB1) and elastase, which are both NET components, mediate NETosis, the immune response that protects against microbes in healthy individuals but contributes to tissue damage in patients with immune-related disorders. The first aim of this work was to confirm that elevations of elastase-DNA and HMGB1-DNA correlate with active lupus nephritis. The second aim was to determine if levels of these proteins at the time of lupus nephritis flare predicted renal outcomes at 12 and 24 months.

To pursue the first hypothesis, 49 patients with active SLE (18 of whom had active lupus nephritis) were evaluated along with 23 patients with inactive SLE and 20 healthy controls.

Highest levels seen in proliferative nephritis

Relative to healthy controls, patients with active SLE have highly significantly increased levels of both proteins (P < .0001). And relative to those with inactive SLE, the levels of active patients were higher but fell short of statistical significance. However, when the researchers compared those with active lupus nephritis with those who had active SLE but no nephritis, both proteins were significantly higher (P < .04), and the levels in patients with proliferative relative to nonproliferative lupus nephritis were higher still (P < .009).

To pursue the second aim of the study, the researchers retrospectively evaluated 109 patients with SLE. All had active lupus nephritis, a baseline estimated glomerular filtration rate (eGFR) greater than 30 mL/min prior to the flare, and at least 2 years of follow-up. They evaluated complete response at 12 and 24 months, percent decline in eGFR, and severe renal impairment (eGFR ≤ 30 mL/min) in the context of levels of elastase and HMGB1.

With elevations in either NET remnant, the odds ratio of failing to achieve a complete response at 24 months were approximately doubled for elastase-DNA (OR, 1.96; P = .01) and for HMGB1 (OR, 2.61; P = .02). For the endpoint of severe renal impairment 24 months after a lupus nephritis flare, there was also a positive association with both elastase-DNA (OR, 1.55; P = .005) and HMBG1-DNA (OR, 1.91; P = .01).

“For every 100-unit increase in elastase-DNA complexes, there is a 4.8% decrease in eGFR,” reported Dr. Whittall-Garcia, who noted this relationship was highly statistically significant (P < .0001). For HMGB1-DNA, each 100-unit increase was associated with a 5.3% decrease in eGFR (P = .0006).
 

No other biomarkers compare for prognosis

“After adjusting for multiple variables, these protein levels at the time of the flare outperformed all conventional biomarkers, including proteinuria and complement levels,” Dr. Whittall-Garcia said.

Larger validating studies are needed, but Dr. Whittall-Garcia is optimistic that measuring these NET remnant levels will prove useful for monitoring patients at the time of the lupus nephritis flare and over time for the purposes of predicting adverse outcomes and response to therapy.

Although more work is needed, Adegbenga A. Bankole, MD, associate professor of medicine at Virginia Tech University and chief of the rheumatology division at the Carilion Clinic, both in Roanoke, agreed that this is a promising research direction. He reported that NETs have been attracting interest at several research centers for their potential in helping to understand the pathogenesis of lupus nephritis.

“It is unlikely that any one test will ever be a panacea in the diagnosis or predictor of outcomes in lupus nephritis,” Dr. Bankole said in an interview, but “given the importance of NETosis in the development of this disease, studies like this will form the basis of the multistep process through which we will improve patient care.”

With further progress in this area, Dr. Bankole predicted that these studies will lead to clinical applications.

“Dr. Whittall-Garcia and her team will help in the development of diagnostic and/or predictive algorithms that may go on to help improve survival of future patients with SLE,” he said.Dr. Whittall-Garcia and Dr. Bankole report they have no relevant financial relationships.

At the time of a nephritis flare in patients with systemic lupus erythematosus (SLE), elevated levels of two neutrophil extracellular trap (NET) protein complexes, elastase-DNA and HMGB1-DNA, predict declining renal function, poor response to therapy, and adverse renal outcomes, according to work presented at the annual meeting of the Canadian Rheumatology Association.

“These proteins are not only predominantly elevated in patients with proliferative lupus nephritis, but they correlate with adverse renal outcomes when patients are followed over 24 months,” reported Laura P. Whittall-Garcia, MD, a clinical fellow in rheumatology at the University of Toronto.

Lupus nephritis is common in SLE, developing in about 50% of patients, according to Dr. Whittall-Garcia. Of these, up to 20% will not respond to standard therapies, typically resulting in end-stage renal disease. Up until now, there has been no reliable method of predicting this adverse clinical course.
 

Proteins identified in NETs

The series of studies conducted by Dr. Whittall-Garcia and coinvestigators were focused on NETs, a network of strings of DNA that typically bind pathogenic microbes to prevent infection but can participate in the pathology of immune-mediated conditions. As Dr. Whittall-Garcia explained, DNA extruded from NETs has been a source of autoantigens.

Based on earlier work, they pursued the hypothesis that high mobility group box 1 (HMGB1) and elastase, which are both NET components, mediate NETosis, the immune response that protects against microbes in healthy individuals but contributes to tissue damage in patients with immune-related disorders. The first aim of this work was to confirm that elevations of elastase-DNA and HMGB1-DNA correlate with active lupus nephritis. The second aim was to determine if levels of these proteins at the time of lupus nephritis flare predicted renal outcomes at 12 and 24 months.

To pursue the first hypothesis, 49 patients with active SLE (18 of whom had active lupus nephritis) were evaluated along with 23 patients with inactive SLE and 20 healthy controls.

Highest levels seen in proliferative nephritis

Relative to healthy controls, patients with active SLE have highly significantly increased levels of both proteins (P < .0001). And relative to those with inactive SLE, the levels of active patients were higher but fell short of statistical significance. However, when the researchers compared those with active lupus nephritis with those who had active SLE but no nephritis, both proteins were significantly higher (P < .04), and the levels in patients with proliferative relative to nonproliferative lupus nephritis were higher still (P < .009).

To pursue the second aim of the study, the researchers retrospectively evaluated 109 patients with SLE. All had active lupus nephritis, a baseline estimated glomerular filtration rate (eGFR) greater than 30 mL/min prior to the flare, and at least 2 years of follow-up. They evaluated complete response at 12 and 24 months, percent decline in eGFR, and severe renal impairment (eGFR ≤ 30 mL/min) in the context of levels of elastase and HMGB1.

With elevations in either NET remnant, the odds ratio of failing to achieve a complete response at 24 months were approximately doubled for elastase-DNA (OR, 1.96; P = .01) and for HMGB1 (OR, 2.61; P = .02). For the endpoint of severe renal impairment 24 months after a lupus nephritis flare, there was also a positive association with both elastase-DNA (OR, 1.55; P = .005) and HMBG1-DNA (OR, 1.91; P = .01).

“For every 100-unit increase in elastase-DNA complexes, there is a 4.8% decrease in eGFR,” reported Dr. Whittall-Garcia, who noted this relationship was highly statistically significant (P < .0001). For HMGB1-DNA, each 100-unit increase was associated with a 5.3% decrease in eGFR (P = .0006).
 

No other biomarkers compare for prognosis

“After adjusting for multiple variables, these protein levels at the time of the flare outperformed all conventional biomarkers, including proteinuria and complement levels,” Dr. Whittall-Garcia said.

Larger validating studies are needed, but Dr. Whittall-Garcia is optimistic that measuring these NET remnant levels will prove useful for monitoring patients at the time of the lupus nephritis flare and over time for the purposes of predicting adverse outcomes and response to therapy.

Although more work is needed, Adegbenga A. Bankole, MD, associate professor of medicine at Virginia Tech University and chief of the rheumatology division at the Carilion Clinic, both in Roanoke, agreed that this is a promising research direction. He reported that NETs have been attracting interest at several research centers for their potential in helping to understand the pathogenesis of lupus nephritis.

“It is unlikely that any one test will ever be a panacea in the diagnosis or predictor of outcomes in lupus nephritis,” Dr. Bankole said in an interview, but “given the importance of NETosis in the development of this disease, studies like this will form the basis of the multistep process through which we will improve patient care.”

With further progress in this area, Dr. Bankole predicted that these studies will lead to clinical applications.

“Dr. Whittall-Garcia and her team will help in the development of diagnostic and/or predictive algorithms that may go on to help improve survival of future patients with SLE,” he said.Dr. Whittall-Garcia and Dr. Bankole report they have no relevant financial relationships.

At the time of a nephritis flare in patients with systemic lupus erythematosus (SLE), elevated levels of two neutrophil extracellular trap (NET) protein complexes, elastase-DNA and HMGB1-DNA, predict declining renal function, poor response to therapy, and adverse renal outcomes, according to work presented at the annual meeting of the Canadian Rheumatology Association.

“These proteins are not only predominantly elevated in patients with proliferative lupus nephritis, but they correlate with adverse renal outcomes when patients are followed over 24 months,” reported Laura P. Whittall-Garcia, MD, a clinical fellow in rheumatology at the University of Toronto.

Lupus nephritis is common in SLE, developing in about 50% of patients, according to Dr. Whittall-Garcia. Of these, up to 20% will not respond to standard therapies, typically resulting in end-stage renal disease. Up until now, there has been no reliable method of predicting this adverse clinical course.
 

Proteins identified in NETs

The series of studies conducted by Dr. Whittall-Garcia and coinvestigators were focused on NETs, a network of strings of DNA that typically bind pathogenic microbes to prevent infection but can participate in the pathology of immune-mediated conditions. As Dr. Whittall-Garcia explained, DNA extruded from NETs has been a source of autoantigens.

Based on earlier work, they pursued the hypothesis that high mobility group box 1 (HMGB1) and elastase, which are both NET components, mediate NETosis, the immune response that protects against microbes in healthy individuals but contributes to tissue damage in patients with immune-related disorders. The first aim of this work was to confirm that elevations of elastase-DNA and HMGB1-DNA correlate with active lupus nephritis. The second aim was to determine if levels of these proteins at the time of lupus nephritis flare predicted renal outcomes at 12 and 24 months.

To pursue the first hypothesis, 49 patients with active SLE (18 of whom had active lupus nephritis) were evaluated along with 23 patients with inactive SLE and 20 healthy controls.

Highest levels seen in proliferative nephritis

Relative to healthy controls, patients with active SLE have highly significantly increased levels of both proteins (P < .0001). And relative to those with inactive SLE, the levels of active patients were higher but fell short of statistical significance. However, when the researchers compared those with active lupus nephritis with those who had active SLE but no nephritis, both proteins were significantly higher (P < .04), and the levels in patients with proliferative relative to nonproliferative lupus nephritis were higher still (P < .009).

To pursue the second aim of the study, the researchers retrospectively evaluated 109 patients with SLE. All had active lupus nephritis, a baseline estimated glomerular filtration rate (eGFR) greater than 30 mL/min prior to the flare, and at least 2 years of follow-up. They evaluated complete response at 12 and 24 months, percent decline in eGFR, and severe renal impairment (eGFR ≤ 30 mL/min) in the context of levels of elastase and HMGB1.

With elevations in either NET remnant, the odds ratio of failing to achieve a complete response at 24 months were approximately doubled for elastase-DNA (OR, 1.96; P = .01) and for HMGB1 (OR, 2.61; P = .02). For the endpoint of severe renal impairment 24 months after a lupus nephritis flare, there was also a positive association with both elastase-DNA (OR, 1.55; P = .005) and HMBG1-DNA (OR, 1.91; P = .01).

“For every 100-unit increase in elastase-DNA complexes, there is a 4.8% decrease in eGFR,” reported Dr. Whittall-Garcia, who noted this relationship was highly statistically significant (P < .0001). For HMGB1-DNA, each 100-unit increase was associated with a 5.3% decrease in eGFR (P = .0006).
 

No other biomarkers compare for prognosis

“After adjusting for multiple variables, these protein levels at the time of the flare outperformed all conventional biomarkers, including proteinuria and complement levels,” Dr. Whittall-Garcia said.

Larger validating studies are needed, but Dr. Whittall-Garcia is optimistic that measuring these NET remnant levels will prove useful for monitoring patients at the time of the lupus nephritis flare and over time for the purposes of predicting adverse outcomes and response to therapy.

Although more work is needed, Adegbenga A. Bankole, MD, associate professor of medicine at Virginia Tech University and chief of the rheumatology division at the Carilion Clinic, both in Roanoke, agreed that this is a promising research direction. He reported that NETs have been attracting interest at several research centers for their potential in helping to understand the pathogenesis of lupus nephritis.

“It is unlikely that any one test will ever be a panacea in the diagnosis or predictor of outcomes in lupus nephritis,” Dr. Bankole said in an interview, but “given the importance of NETosis in the development of this disease, studies like this will form the basis of the multistep process through which we will improve patient care.”

With further progress in this area, Dr. Bankole predicted that these studies will lead to clinical applications.

“Dr. Whittall-Garcia and her team will help in the development of diagnostic and/or predictive algorithms that may go on to help improve survival of future patients with SLE,” he said.Dr. Whittall-Garcia and Dr. Bankole report they have no relevant financial relationships.

Real-world, population-based data suggest that the discontinuation rates for biosimilars prescribed to treat inflammatory rheumatic diseases are similar to those for their corresponding originator biologics, according to a study of patients in British Columbia who were required to switch to biosimilars.

“The decision to mandate use of biosimilars provided an ideal context for a natural experiment,” Diane Lacaille, MD, chair in arthritis research at the University of British Columbia, Vancouver, explained in her presentation of the study at the annual meeting of the Canadian Rheumatology Association.

On the basis of the real-world data, which was collected before and after a province-wide requirement to use biosimilars in place of originator biologics, there was no major difference in discontinuation rates, an outcome that Dr. Lacaille characterized as “a surrogate for both efficacy and safety.”

In the 2019 rheumatoid arthritis treatment guidelines from the European Alliance of Associations for Rheumatology, biosimilars are advocated for addressing the high cost of biologics based on evidence of efficacy and safety comparable with originator biologics. According to one of the coauthors of those guidelines, Tom W. J. Huizinga, MD, PhD, head of rheumatology at Leiden (Netherlands) University Medical Center, there is reasonable confidence in biosimilars as an adequate substitute for originator drugs, but real-world data are welcome.

“Real-world data provide different information than controlled trials and long-term data as well, so these [Canadian findings] are useful to support the data from [randomized controlled trials],” Dr. Huizinga said in an interview. He was not involved in the Canadian study.
 

Survivorship evaluated after switch to biosimilars

In British Columbia, biosimilars were mandated province-wide for new prescriptions of infliximab and etanercept in June 2017. In 2019, the mandate was extended to patients already taking originator infliximab (Remicade) and originator etanercept (Enbrel). Since that time, the mandate for biosimilars has also been applied to adalimumab (Humira). For the comparison of infliximab and etanercept originators with their biosimilars, Dr. Lacaille and associates compared survivorship for the 3 years after the policy change, when patients were on biosimilars, with the 3 years prior to the change, when patients were on the originators. They compared survivorship with originator adalimumab with its biosimilars for prior to and after the switch.

”People were followed from anti-TNF [tumor necrosis factor] initiation until discontinuation for any reason,” reported Dr. Lacaille, who said data were censored for death and moving out of the province. In British Columbia, where there is universal health care, all dispensed medications can be tracked. The definition of anti-TNF discontinuation in this study was no prescription renewal for at least 6 months.

The follow-up was censored at March 2, 2020, to avoid the potential impact of COVID-19 on antirheumatic drug use. Discontinuation was standardized for the comparison of originator with biosimilar drugs as rates per 100 person-years. Statistical adjustments were made for potential confounders.

The researchers compared 1,312 patients on etanercept and 827 on a biosimilar of it, 230 patients on infliximab and 271 on a biosimilar of it, and 1,773 on adalimumab and 2,213 on a biosimilar of it. The indication was RA in approximately 60% of those on etanercept or a biosimilar and 50% of those on infliximab or adalimumab and their biosimilars. More than half of the remaining patients had indications for psoriatic arthritis, and the rest had ankylosing spondylitis.
 

No differences reach statistical significance

On the basis of discontinuation rates per 100 person-years, etanercept and its biosimilars performed almost identically (37.10 vs. 37.02, respectively). Although the discontinuation rate per 100 person-years was lower on infliximab than a biosimilar of it (29.97 vs. 37.96), the difference was not statistically significant (P = .076).

For adalimumab, the discontinuation rate was also lower on the originator drug than a biosimilar of it (32.92 vs. 36.36), but, again, this difference was also insignificant (P = .56).

When the discontinuation data were evaluated on the basis of a Cox model involving a propensity weight overlap, the univariate and the multivariable analyses found that the biosimilars had similar risks for discontinuation. Univariate analysis revealed hazard ratios for discontinuation of the biosimilar relative to the originator were 0.98 (P = .783) for etanercept, 1.17 (P = .242) for infliximab, and 1.08 (P = .09) for adalimumab. In the multivariable model, adjusted HRs for discontinuation were about the same for each of the biosimilars relative to the originator: 0.98 (P = .807) for etanercept, 1.19 (P = .183) for infliximab, and 1.08 (P = .089) for adalimumab.

Relative to previously published direct comparisons, this real-world analysis and its duration of follow-up address the limitations of formal trials. In a 2020 BMJ meta-analysis of published data from 45 trials comparing biosimilar with originator drugs in patients with RA who had failed methotrexate, the authors found only “minor differences in harms and benefits,” but they cautioned that the analysis was “hampered by a lack of long-term direct comparisons.”

In an interview, Dr. Huizinga noted that a systematic review of adalimumab biosimilars that he led 2 years ago showed that they perform comparably with the originator biologics. This and other published studies have consistently shown “that there is no difference between biologics and originators.”

Dr. Lacaille disclosed financial relationships with Fresenius Kabi, Janssen, Organon, Pfizer, and Viatris. Dr. Huizinga disclosed financial relationships with Abbott, Ablynx, Biotest, Bioscience, Boehringer Ingelheim, Bristol-Myers Squibb, Crescendo Bioscience, Eli Lilly, Galapagos, Janssen, Merck, Novartis, MycoMed, Roche, Sanofi-Aventis, Takeda, and Zydus.

Real-world, population-based data suggest that the discontinuation rates for biosimilars prescribed to treat inflammatory rheumatic diseases are similar to those for their corresponding originator biologics, according to a study of patients in British Columbia who were required to switch to biosimilars.

“The decision to mandate use of biosimilars provided an ideal context for a natural experiment,” Diane Lacaille, MD, chair in arthritis research at the University of British Columbia, Vancouver, explained in her presentation of the study at the annual meeting of the Canadian Rheumatology Association.

On the basis of the real-world data, which was collected before and after a province-wide requirement to use biosimilars in place of originator biologics, there was no major difference in discontinuation rates, an outcome that Dr. Lacaille characterized as “a surrogate for both efficacy and safety.”

In the 2019 rheumatoid arthritis treatment guidelines from the European Alliance of Associations for Rheumatology, biosimilars are advocated for addressing the high cost of biologics based on evidence of efficacy and safety comparable with originator biologics. According to one of the coauthors of those guidelines, Tom W. J. Huizinga, MD, PhD, head of rheumatology at Leiden (Netherlands) University Medical Center, there is reasonable confidence in biosimilars as an adequate substitute for originator drugs, but real-world data are welcome.

“Real-world data provide different information than controlled trials and long-term data as well, so these [Canadian findings] are useful to support the data from [randomized controlled trials],” Dr. Huizinga said in an interview. He was not involved in the Canadian study.
 

Survivorship evaluated after switch to biosimilars

In British Columbia, biosimilars were mandated province-wide for new prescriptions of infliximab and etanercept in June 2017. In 2019, the mandate was extended to patients already taking originator infliximab (Remicade) and originator etanercept (Enbrel). Since that time, the mandate for biosimilars has also been applied to adalimumab (Humira). For the comparison of infliximab and etanercept originators with their biosimilars, Dr. Lacaille and associates compared survivorship for the 3 years after the policy change, when patients were on biosimilars, with the 3 years prior to the change, when patients were on the originators. They compared survivorship with originator adalimumab with its biosimilars for prior to and after the switch.

”People were followed from anti-TNF [tumor necrosis factor] initiation until discontinuation for any reason,” reported Dr. Lacaille, who said data were censored for death and moving out of the province. In British Columbia, where there is universal health care, all dispensed medications can be tracked. The definition of anti-TNF discontinuation in this study was no prescription renewal for at least 6 months.

The follow-up was censored at March 2, 2020, to avoid the potential impact of COVID-19 on antirheumatic drug use. Discontinuation was standardized for the comparison of originator with biosimilar drugs as rates per 100 person-years. Statistical adjustments were made for potential confounders.

The researchers compared 1,312 patients on etanercept and 827 on a biosimilar of it, 230 patients on infliximab and 271 on a biosimilar of it, and 1,773 on adalimumab and 2,213 on a biosimilar of it. The indication was RA in approximately 60% of those on etanercept or a biosimilar and 50% of those on infliximab or adalimumab and their biosimilars. More than half of the remaining patients had indications for psoriatic arthritis, and the rest had ankylosing spondylitis.
 

No differences reach statistical significance

On the basis of discontinuation rates per 100 person-years, etanercept and its biosimilars performed almost identically (37.10 vs. 37.02, respectively). Although the discontinuation rate per 100 person-years was lower on infliximab than a biosimilar of it (29.97 vs. 37.96), the difference was not statistically significant (P = .076).

For adalimumab, the discontinuation rate was also lower on the originator drug than a biosimilar of it (32.92 vs. 36.36), but, again, this difference was also insignificant (P = .56).

When the discontinuation data were evaluated on the basis of a Cox model involving a propensity weight overlap, the univariate and the multivariable analyses found that the biosimilars had similar risks for discontinuation. Univariate analysis revealed hazard ratios for discontinuation of the biosimilar relative to the originator were 0.98 (P = .783) for etanercept, 1.17 (P = .242) for infliximab, and 1.08 (P = .09) for adalimumab. In the multivariable model, adjusted HRs for discontinuation were about the same for each of the biosimilars relative to the originator: 0.98 (P = .807) for etanercept, 1.19 (P = .183) for infliximab, and 1.08 (P = .089) for adalimumab.

Relative to previously published direct comparisons, this real-world analysis and its duration of follow-up address the limitations of formal trials. In a 2020 BMJ meta-analysis of published data from 45 trials comparing biosimilar with originator drugs in patients with RA who had failed methotrexate, the authors found only “minor differences in harms and benefits,” but they cautioned that the analysis was “hampered by a lack of long-term direct comparisons.”

In an interview, Dr. Huizinga noted that a systematic review of adalimumab biosimilars that he led 2 years ago showed that they perform comparably with the originator biologics. This and other published studies have consistently shown “that there is no difference between biologics and originators.”

Dr. Lacaille disclosed financial relationships with Fresenius Kabi, Janssen, Organon, Pfizer, and Viatris. Dr. Huizinga disclosed financial relationships with Abbott, Ablynx, Biotest, Bioscience, Boehringer Ingelheim, Bristol-Myers Squibb, Crescendo Bioscience, Eli Lilly, Galapagos, Janssen, Merck, Novartis, MycoMed, Roche, Sanofi-Aventis, Takeda, and Zydus.

Real-world, population-based data suggest that the discontinuation rates for biosimilars prescribed to treat inflammatory rheumatic diseases are similar to those for their corresponding originator biologics, according to a study of patients in British Columbia who were required to switch to biosimilars.

“The decision to mandate use of biosimilars provided an ideal context for a natural experiment,” Diane Lacaille, MD, chair in arthritis research at the University of British Columbia, Vancouver, explained in her presentation of the study at the annual meeting of the Canadian Rheumatology Association.

On the basis of the real-world data, which was collected before and after a province-wide requirement to use biosimilars in place of originator biologics, there was no major difference in discontinuation rates, an outcome that Dr. Lacaille characterized as “a surrogate for both efficacy and safety.”

In the 2019 rheumatoid arthritis treatment guidelines from the European Alliance of Associations for Rheumatology, biosimilars are advocated for addressing the high cost of biologics based on evidence of efficacy and safety comparable with originator biologics. According to one of the coauthors of those guidelines, Tom W. J. Huizinga, MD, PhD, head of rheumatology at Leiden (Netherlands) University Medical Center, there is reasonable confidence in biosimilars as an adequate substitute for originator drugs, but real-world data are welcome.

“Real-world data provide different information than controlled trials and long-term data as well, so these [Canadian findings] are useful to support the data from [randomized controlled trials],” Dr. Huizinga said in an interview. He was not involved in the Canadian study.
 

Survivorship evaluated after switch to biosimilars

In British Columbia, biosimilars were mandated province-wide for new prescriptions of infliximab and etanercept in June 2017. In 2019, the mandate was extended to patients already taking originator infliximab (Remicade) and originator etanercept (Enbrel). Since that time, the mandate for biosimilars has also been applied to adalimumab (Humira). For the comparison of infliximab and etanercept originators with their biosimilars, Dr. Lacaille and associates compared survivorship for the 3 years after the policy change, when patients were on biosimilars, with the 3 years prior to the change, when patients were on the originators. They compared survivorship with originator adalimumab with its biosimilars for prior to and after the switch.

”People were followed from anti-TNF [tumor necrosis factor] initiation until discontinuation for any reason,” reported Dr. Lacaille, who said data were censored for death and moving out of the province. In British Columbia, where there is universal health care, all dispensed medications can be tracked. The definition of anti-TNF discontinuation in this study was no prescription renewal for at least 6 months.

The follow-up was censored at March 2, 2020, to avoid the potential impact of COVID-19 on antirheumatic drug use. Discontinuation was standardized for the comparison of originator with biosimilar drugs as rates per 100 person-years. Statistical adjustments were made for potential confounders.

The researchers compared 1,312 patients on etanercept and 827 on a biosimilar of it, 230 patients on infliximab and 271 on a biosimilar of it, and 1,773 on adalimumab and 2,213 on a biosimilar of it. The indication was RA in approximately 60% of those on etanercept or a biosimilar and 50% of those on infliximab or adalimumab and their biosimilars. More than half of the remaining patients had indications for psoriatic arthritis, and the rest had ankylosing spondylitis.
 

No differences reach statistical significance

On the basis of discontinuation rates per 100 person-years, etanercept and its biosimilars performed almost identically (37.10 vs. 37.02, respectively). Although the discontinuation rate per 100 person-years was lower on infliximab than a biosimilar of it (29.97 vs. 37.96), the difference was not statistically significant (P = .076).

For adalimumab, the discontinuation rate was also lower on the originator drug than a biosimilar of it (32.92 vs. 36.36), but, again, this difference was also insignificant (P = .56).

When the discontinuation data were evaluated on the basis of a Cox model involving a propensity weight overlap, the univariate and the multivariable analyses found that the biosimilars had similar risks for discontinuation. Univariate analysis revealed hazard ratios for discontinuation of the biosimilar relative to the originator were 0.98 (P = .783) for etanercept, 1.17 (P = .242) for infliximab, and 1.08 (P = .09) for adalimumab. In the multivariable model, adjusted HRs for discontinuation were about the same for each of the biosimilars relative to the originator: 0.98 (P = .807) for etanercept, 1.19 (P = .183) for infliximab, and 1.08 (P = .089) for adalimumab.

Relative to previously published direct comparisons, this real-world analysis and its duration of follow-up address the limitations of formal trials. In a 2020 BMJ meta-analysis of published data from 45 trials comparing biosimilar with originator drugs in patients with RA who had failed methotrexate, the authors found only “minor differences in harms and benefits,” but they cautioned that the analysis was “hampered by a lack of long-term direct comparisons.”

In an interview, Dr. Huizinga noted that a systematic review of adalimumab biosimilars that he led 2 years ago showed that they perform comparably with the originator biologics. This and other published studies have consistently shown “that there is no difference between biologics and originators.”

Dr. Lacaille disclosed financial relationships with Fresenius Kabi, Janssen, Organon, Pfizer, and Viatris. Dr. Huizinga disclosed financial relationships with Abbott, Ablynx, Biotest, Bioscience, Boehringer Ingelheim, Bristol-Myers Squibb, Crescendo Bioscience, Eli Lilly, Galapagos, Janssen, Merck, Novartis, MycoMed, Roche, Sanofi-Aventis, Takeda, and Zydus.