Feature

Studies show oncologists can adapt quickly to new evidence

Ronac Mamtani, MD 
Ronac Mamtani
Ravi B. Parikh, MD 
Ravi B. Parikh

Oncologists acted promptly in response to an FDA restriction last year on the label indication for two immunotherapies, effectively reducing their use by about 50% within 6 months, according to results of a study published in JAMA.

FDA approvals of anti-PD-1 treatments resulted in similarly swift action, with the agents quickly reaching eligible patients in clinical practice, according to results of a separate study published in JAMA Oncology.

As emerging cancer treatments continue to be approved at a rapid pace, these findings suggest that oncologists, who are reputed to be slow to change, can quickly adjust practice based on post-approval safety data or when novel cancer therapies enter the market.

“This happened at a very interesting time, when immunotherapy use was flourishing,” Ronac Mamtani, MD, MSCE, an author of the first study and assistant professor of hematology-oncology at Perelman School of Medicine at University of Pennsylvania, said in an interview with HemOnc Today. “What was unique about our study was the finding that clinicians change their practices not only based on published evidence, but also emerging data.”

Cary Gross, MD
Cary P. Gross

The rapid adoption of new therapies has been driven in part by the FDA’s accelerated approval program, which has faced criticism in recent months.

“The accelerated approval process, which approves drugs based on phase 2 evidence — noncomparative studies — has come under a lot of scrutiny,” Ravi B. Parikh, MD, co-lead author of the first study and instructor in medical ethics and health policy at University of Pennsylvania, told HemOnc Today. “Drugs that have a lot of potential, particularly in the cancer space, are getting out to the market without the same level of evidence that prior drugs have had.”

Rapid reversals

In the first study, Mamtani and colleagues evaluated usage rates of two immunotherapies approved in 2017 for first-line treatment of patients with advanced bladder cancer who were ineligible for standard cisplatin-based chemotherapy. The FDA based the accelerated approvals of both drugs, the PD-1 inhibitor pembrolizumab (Keytruda, Merck) and the PD-L1 inhibitor atezolizumab (Tecentriq, Genentech), on the strength of single-group phase 2 studies.

However, subsequent phase 3 study data showed reduced survival with these drugs compared with first-line, platinum-based chemotherapy for patients with PD-L1-negative tumors. This prompted the FDA in June 2018 to limit the label indications of the immunotherapies to cisplatin-ineligible patients who have PD-L1-positive tumors.

“To try to isolate the effect of this FDA label change, we used an interrupted time series analysis,” Mamtani told HemOnc Today. “This analysis uses regression modeling to determine differences in the utilization of chemotherapy, immunotherapy and PD-L1 testing, both after the policy and if the policy never occurred.”

The researchers assessed de-identified data on 1,965 patients (median age, 73 years; 26% women) in the Flatiron Health database who were diagnosed with advanced bladder cancer between Jan. 1, 2016, and Jan. 31, 2019, and treated with one or more lines of systemic therapy. The database includes health records from more than 280 oncology clinics across the U.S.

The time series analysis showed that between May 2018 and January 2019, the rate of immunotherapy use declined from 51.9 to 30.3 per 100 patients. Meanwhile, the rate of chemotherapy use increased from 37 to 60.6 per 100 patients. Researchers also observed an increase in the rate of PD-L1 testing, from 9.3 to 21.2 per 100 patients.

“Essentially the FDA label change was associated with rapid reversals for trends in immunotherapy and chemotherapy use,” Parikh told HemOnc Today. “The important point is that these changes happened in about 6 to 8 months. The rapidity of that practice change was what was most surprising to us, both as practicing oncologists and as researchers with expertise in this space.”

Real-world relevance

Results of the separate study by Cary P. Gross, MD, professor of medicine and director of the National Clinician Scholars Program at Yale University, and colleagues suggested that oncologists are more adept than previously believed at changing practice swiftly and substantially in response to new evidence.

In the retrospective cohort study, conducted between Jan. 1, 2011, and Aug. 31, 2016, the researchers evaluated the speed of anti-PD-1 agent uptake after FDA approval among patients in clinical practice. They study included 3,089 patients from the Flatiron Health database who were candidates for anti-PD-1 treatment of melanoma (n = 555; median age, 66 years), non-small cell lung cancer (n = 2,159; median age, 67 years) and renal cell carcinoma (n = 375; median age, 66 years).

Researchers also compared the ages of patients treated in clinical practice with the ages of those treated in pivotal clinical trials.

Gross and colleagues found that within 4 months of FDA approval, more than 60% of eligible patients in both the pivotal trial cohort and the clinical practice cohort had received anti-PD-1 treatment.

Similar percentages of older and younger patients received anti-PD-1 treatment in the first 9 months after FDA approval. However, researchers observed significant age differences between clinical trial participants and those exposed to anti-PD-L1 treatment in a clinical practice setting, with more patients aged older than 65 years seen in clinical practice (range, 60.6% to 63.9%) than in pivotal trials (range, 31.7% to 41%; P < .001 for all).

Gross said this finding raises concerns that some rapid FDA approvals are based on pivotal trials that may not be representative of a “real-world” population.

“We found that patients treated in the real world were substantially older than patients who had enrolled in the pivotal clinical trials, and yet the oncologists were adopting the new therapies into their practice just as quickly for their older patients as for their younger patients,” he said in an interview with HemOnc Today. “This underscores the need to rigorously study the effectiveness of new drugs in real-world practice.”

Gross said the findings of his study reflect encouraging changes in how quickly oncologists react to new evidence.

“Conventional wisdom used to hold that doctors were very slow to change their practice in response to evidence,” Gross told HemOnc Today. “I remember being taught in medical school that it takes up to 17 years for evidence to change practice. Yet our study found that it took less than 6 months for a new cancer treatment to become standard of care.”

Gross also discussed Parikh and Mamtani’s study, noting that the evidence of “rapid reversals” was a novel and important finding.

“Their study built upon ours, in that they showed that oncologists can de-adopt, or abandon, a therapy if evidence suggests that it is not helpful,” he said. “A common denominator to both of our studies is that new evidence — either of benefit or of harm — can rapidly change practice.”

Gross said he hopes future research will focus on ensuring that pivotal clinical trials reflect the real-world patient population.

“The next step is to make sure that we are conducting studies using real-world patients and hard outcomes,” he said. “We want to be sure that oncologists are using evidence that is rigorous, generalizable and focuses on outcomes that actual patients will care about.” – by Jennifer Byrne

References:

Parikh R, et al. JAMA. 2019;doi:10.1001/jama.2019.10650.

O’Connor JM, et al. JAMA Oncol. 2019;doi:10.1001/jamaoncol.2018.0798.

For more information:

Cary Gross, MD, can be reached at 789 Howard Ave., New Haven, CT 06519; email: cary.gross@yale.edu.

Ron ac Mamtani, MD , can be reached at University of Pennsylvania, 16 Penn Tower, Philadelphia, PA 19104; email: ronac.mamtani@uphs.upenn.edu.

Ravi B. Parikh, MD, can be reached at Blockley Hall, 423 Guardian Ave., Philadelphia, PA 19104; email: ravi.parikh@pennmedicine.upenn.edu.

Disclosures: Gross reports research funding from NCCN/Pfizer, funding from Johnson & Johnson to design new approaches to sharing clinical trial data, and funding for traveling to and speaking at a scientific conference by Flatiron Inc. Mamtani reports personal fees from Roche. Parikh reports no relevant disclosures. Please see the studies for all other authors’ relevant financial disclosures.

Ronac Mamtani, MD 
Ronac Mamtani
Ravi B. Parikh, MD 
Ravi B. Parikh

Oncologists acted promptly in response to an FDA restriction last year on the label indication for two immunotherapies, effectively reducing their use by about 50% within 6 months, according to results of a study published in JAMA.

FDA approvals of anti-PD-1 treatments resulted in similarly swift action, with the agents quickly reaching eligible patients in clinical practice, according to results of a separate study published in JAMA Oncology.

As emerging cancer treatments continue to be approved at a rapid pace, these findings suggest that oncologists, who are reputed to be slow to change, can quickly adjust practice based on post-approval safety data or when novel cancer therapies enter the market.

“This happened at a very interesting time, when immunotherapy use was flourishing,” Ronac Mamtani, MD, MSCE, an author of the first study and assistant professor of hematology-oncology at Perelman School of Medicine at University of Pennsylvania, said in an interview with HemOnc Today. “What was unique about our study was the finding that clinicians change their practices not only based on published evidence, but also emerging data.”

Cary Gross, MD
Cary P. Gross

The rapid adoption of new therapies has been driven in part by the FDA’s accelerated approval program, which has faced criticism in recent months.

“The accelerated approval process, which approves drugs based on phase 2 evidence — noncomparative studies — has come under a lot of scrutiny,” Ravi B. Parikh, MD, co-lead author of the first study and instructor in medical ethics and health policy at University of Pennsylvania, told HemOnc Today. “Drugs that have a lot of potential, particularly in the cancer space, are getting out to the market without the same level of evidence that prior drugs have had.”

Rapid reversals

In the first study, Mamtani and colleagues evaluated usage rates of two immunotherapies approved in 2017 for first-line treatment of patients with advanced bladder cancer who were ineligible for standard cisplatin-based chemotherapy. The FDA based the accelerated approvals of both drugs, the PD-1 inhibitor pembrolizumab (Keytruda, Merck) and the PD-L1 inhibitor atezolizumab (Tecentriq, Genentech), on the strength of single-group phase 2 studies.

However, subsequent phase 3 study data showed reduced survival with these drugs compared with first-line, platinum-based chemotherapy for patients with PD-L1-negative tumors. This prompted the FDA in June 2018 to limit the label indications of the immunotherapies to cisplatin-ineligible patients who have PD-L1-positive tumors.

PAGE BREAK

“To try to isolate the effect of this FDA label change, we used an interrupted time series analysis,” Mamtani told HemOnc Today. “This analysis uses regression modeling to determine differences in the utilization of chemotherapy, immunotherapy and PD-L1 testing, both after the policy and if the policy never occurred.”

The researchers assessed de-identified data on 1,965 patients (median age, 73 years; 26% women) in the Flatiron Health database who were diagnosed with advanced bladder cancer between Jan. 1, 2016, and Jan. 31, 2019, and treated with one or more lines of systemic therapy. The database includes health records from more than 280 oncology clinics across the U.S.

The time series analysis showed that between May 2018 and January 2019, the rate of immunotherapy use declined from 51.9 to 30.3 per 100 patients. Meanwhile, the rate of chemotherapy use increased from 37 to 60.6 per 100 patients. Researchers also observed an increase in the rate of PD-L1 testing, from 9.3 to 21.2 per 100 patients.

“Essentially the FDA label change was associated with rapid reversals for trends in immunotherapy and chemotherapy use,” Parikh told HemOnc Today. “The important point is that these changes happened in about 6 to 8 months. The rapidity of that practice change was what was most surprising to us, both as practicing oncologists and as researchers with expertise in this space.”

Real-world relevance

Results of the separate study by Cary P. Gross, MD, professor of medicine and director of the National Clinician Scholars Program at Yale University, and colleagues suggested that oncologists are more adept than previously believed at changing practice swiftly and substantially in response to new evidence.

In the retrospective cohort study, conducted between Jan. 1, 2011, and Aug. 31, 2016, the researchers evaluated the speed of anti-PD-1 agent uptake after FDA approval among patients in clinical practice. They study included 3,089 patients from the Flatiron Health database who were candidates for anti-PD-1 treatment of melanoma (n = 555; median age, 66 years), non-small cell lung cancer (n = 2,159; median age, 67 years) and renal cell carcinoma (n = 375; median age, 66 years).

Researchers also compared the ages of patients treated in clinical practice with the ages of those treated in pivotal clinical trials.

Gross and colleagues found that within 4 months of FDA approval, more than 60% of eligible patients in both the pivotal trial cohort and the clinical practice cohort had received anti-PD-1 treatment.

PAGE BREAK

Similar percentages of older and younger patients received anti-PD-1 treatment in the first 9 months after FDA approval. However, researchers observed significant age differences between clinical trial participants and those exposed to anti-PD-L1 treatment in a clinical practice setting, with more patients aged older than 65 years seen in clinical practice (range, 60.6% to 63.9%) than in pivotal trials (range, 31.7% to 41%; P < .001 for all).

Gross said this finding raises concerns that some rapid FDA approvals are based on pivotal trials that may not be representative of a “real-world” population.

“We found that patients treated in the real world were substantially older than patients who had enrolled in the pivotal clinical trials, and yet the oncologists were adopting the new therapies into their practice just as quickly for their older patients as for their younger patients,” he said in an interview with HemOnc Today. “This underscores the need to rigorously study the effectiveness of new drugs in real-world practice.”

Gross said the findings of his study reflect encouraging changes in how quickly oncologists react to new evidence.

“Conventional wisdom used to hold that doctors were very slow to change their practice in response to evidence,” Gross told HemOnc Today. “I remember being taught in medical school that it takes up to 17 years for evidence to change practice. Yet our study found that it took less than 6 months for a new cancer treatment to become standard of care.”

Gross also discussed Parikh and Mamtani’s study, noting that the evidence of “rapid reversals” was a novel and important finding.

“Their study built upon ours, in that they showed that oncologists can de-adopt, or abandon, a therapy if evidence suggests that it is not helpful,” he said. “A common denominator to both of our studies is that new evidence — either of benefit or of harm — can rapidly change practice.”

Gross said he hopes future research will focus on ensuring that pivotal clinical trials reflect the real-world patient population.

“The next step is to make sure that we are conducting studies using real-world patients and hard outcomes,” he said. “We want to be sure that oncologists are using evidence that is rigorous, generalizable and focuses on outcomes that actual patients will care about.” – by Jennifer Byrne

References:

Parikh R, et al. JAMA. 2019;doi:10.1001/jama.2019.10650.

O’Connor JM, et al. JAMA Oncol. 2019;doi:10.1001/jamaoncol.2018.0798.

For more information:

PAGE BREAK

Cary Gross, MD, can be reached at 789 Howard Ave., New Haven, CT 06519; email: cary.gross@yale.edu.

Ron ac Mamtani, MD , can be reached at University of Pennsylvania, 16 Penn Tower, Philadelphia, PA 19104; email: ronac.mamtani@uphs.upenn.edu.

Ravi B. Parikh, MD, can be reached at Blockley Hall, 423 Guardian Ave., Philadelphia, PA 19104; email: ravi.parikh@pennmedicine.upenn.edu.

Disclosures: Gross reports research funding from NCCN/Pfizer, funding from Johnson & Johnson to design new approaches to sharing clinical trial data, and funding for traveling to and speaking at a scientific conference by Flatiron Inc. Mamtani reports personal fees from Roche. Parikh reports no relevant disclosures. Please see the studies for all other authors’ relevant financial disclosures.

    See more from HemOnc Today's PharmAnalysis