Insights

PD-L1 Biomarker Status: To Test or Not to Test?

Chase Doyle

The PD‐L1 biomarker has entered clinical oncology practice—along with the question of to test or not to test.

As the number of immunotherapies available for the treatment of cancers such as metastatic melanoma, non-small cell lung cancer, and bladder cancer continues to rise, so too does the debate regarding predictive biomarkers to guide therapy optimization. 

According to Sharon Karlsberg, Lead Principal, Oncology Marketing Solutions, ZS Associates—a global firm focused on improving business performance—there is still no consensus regarding the value of PD-L1 expression in tumors.

Positivity for PD-L1, while predictive of response to anti-PD1/anti-PDL1 therapy, does not always guarantee it. Many patients will benefit regardless of testing status.

To compound the problem, one therapeutic agent is approved for use with a companion diagnostic for PD-L1 expression, while other agents don’t require positive status.

In a survey conducted by ZS Associates and WebMD of 91 oncologists attending the 2016 American Society of Clinical Oncology (ASCO) annual meeting, more than half of physicians who do not use PD-L1 testing expressed uncertainty about the tests and are not sure if they are necessary.



The ZS survey also revealed that:

  • Three-fourths of oncologists have used PD-L1 testing, most frequently for NSCLC and metastatic melanoma (Figure 1)
  • PD-L1 testing rates range from ~40% for NSCLC patients to 14% for head and neck cancer patients (Figure 2)

Figure 1. Three-fourths of oncologists have used PD-L1 testing,
most frequently for NSCLC and metastatic melanoma.


Nivolumab vs. Pembrolizumab

While new agents in development are being combined with tests to evaluate PD-L1 expression, one agent that does not require a companion diagnostic for PD-L1 expression is nivolumab, the first PD-1 inhibitor approved for use in advanced, relapsed non-small cell lung cancer (NSCLC) in March 2015.

In clinical trials as a second-line treatment for squamous non-small cell lung cancer (NSCLC), nivolumab showed benefit over docetaxel chemotherapy regardless of patients’ PD-L1 status. However, the degree of expression of the PD-L1 biomarker was associated with an increased response rate in the NSCLC population.



Edward Garon, MD, Director of Thoracic Oncology at UCLA’s Jonsson Comprehensive Cancer Center, noted in an interview with Sharon Karlsberg that in NSCLC patients who had at least 10% staining for PD-L1, there was a “tremendous benefit” for nivolumab.

“When the PD-L1 expression was higher,” said Dr. Garon, “there was an increased response rate that actually translated into improvements in progression-free survival and overall survival, as well.”

Pembrolizumab, approved in September 2014for metastatic melanoma, showed similarly impressive response rates as nivolumab in patients with squamous or adenocarcinoma lung cancers expressing high levels of PD-L1.

For many oncologists, nivolumab and pembrolizumab are not differentiated in terms of clinical benefit; the only difference between the drugs is that pembrolizumab use requires patients to be PD-L1–positive.

That difference, said Ms. Karlsberg, may leave community providers feeling a bit bewildered when selecting a therapy for their patients.

“There's one drug that requires a diagnostic, where we can clearly predict a positive response for high expressers of PD-L1, and there's another drug that seems to have a benefit, regardless of PD-L1 expression,” she explained. “With limited tissue available and the complexity around which test to choose, some community oncologists have said that it's simply easier not to test.”

This may leave some oncologists asking themselves: If nivolumab is safe and effective for patients, outperforms chemotherapy, and can be given without a companion diagnostic, then why go through the trouble of testing for PD-L1 expression?

  

Figure 2. About 40% of NSCLC patients receive PD-L1 biomarker testing,
compared with ~14% of head and neck cancer patients.


Prognostic Value for Providers and Patients

One reason that physicians may feel comfortable recommending PD-L1 testing is that it provides better prognostic information for their patients.

“I think the reason to test is to help identify how likely a patient will benefit from the drug,” said Dr. Garon. “This can give both prognostic information when one starts the agent, and can allow you to prioritize your potential therapies in the most logical way for a given patient.”

Another reason might be patient convenience, as pembrolizumab is administered every 3 weeks compared with every 2 weeks for nivolumab.

“Sometimes it's a hardship for patients to get into the infusion suite,” said Roy S. Herbst, MD, PhD, Chief of Medical Oncology at Yale Cancer Center and Smilow Cancer Hospital. “In that case, it might make sense to do the test so you can give pembrolizumab.”

Although not required for nivolumab, in Dr. Herbst’s opinion, PD-L1 status has utility in predicting outcomes, which aids in conversations with patients.

“If my patient had PD-L1 expression that was high or at least in the positive range, I would say that I feel very confident that this is going to work,” said Dr. Herbst.

For the patient with low or no PD-L1 expression, however, Dr. Herbst would still use a checkpoint inhibitor, but he would underscore the risks.

“I see no reason not to test for PD-L1, unless there isn’t any tissue readily available,” said Dr. Herbst. “Why not have all the information? You'd rather have all the cards in anything in life, and certainly, if it involves your patient's life or your life, you want to be able to predict what's going to happen.”

 

Figure 3.
75% of physicians use PD-L1 testing because it “helps identify how
well a patient is likely to respond to” a particular PD-1/PD-L1 immunotherapy.


Testing and Treatment Decisions

Although treatment decisions for relapsed lung cancer patients are unlikely to change based on PD-L1 status, it’s still possible, said Dr. Garon, who highlighted several scenarios where the likelihood of a patient responding would be important to know.

In a second-line setting for instance, there are likely patient populations where it would be helpful to know whether or not they have PD-L1 expression and at what level.

Dr. Garon noted that patients who have not been smokers—patients with EGFR mutations—for example, “clearly derive lesser benefit than the group of patients as a whole… If they did not have a very high expression of PD-L1, then I think it might be better to look for alternate approaches.”

According to Dr. Garon, there are also relative contraindications to PD-1 inhibition, like a history of preexisting autoimmune disease. A practitioner might feel differently about whether or not to start such a patient on a PD-1 inhibitor based on the likelihood of response.

In addition, PD-L1 status could influence the use of combination therapy in patients who may require more aggressive and intensive treatment regimens.

“We're still using PD-L1-inhibitor monotherapy on negative patients, but we might want to do a combination instead,” said Dr. Herbst. “If it’s been shown to work, then perhaps it’s worth… the toxicity of using two drugs versus one.”

Front Line Is the Future

The second-line debate may soon become a historical anecdote, as PD-L1 testing is now part of patient stratification for agents being studied in the front-line setting. 

Trials for both nivolumab and pembrolizumab in the front-line setting have completed in Rome, Italy, and are now awaiting maturation, said Dr. Herbst, who noted that both trials required biomarker positivity to be enrolled.



“PD-L1 testing is now part of all of the protocols for all of these drugs, and it will be part of the standard diagnostic workup that will be prepped for all of these lung cancer patients,” he reported.

Dr. Herbst said that's important, because it means that “we're going to start screening patients for PD-L1 in the front line, and depending on whether they are positive or not, many patients will be getting the drugs up front at this point.”

In fact, in PD-L1-positive patients with previously untreated advanced NSCLC, pembrolizumab just recently demonstrated superior progression-free and overall survival compared with chemotherapy in a clinical study. And, if these coveted datasets continue to show benefits from PD-1 inhibition in the front-line setting across tumor types, then PD-L1 testing will become standard in the front-line setting.

Dr. Herbst, for one, is confident in this vision of the future.

“In my opinion, these drugs are going to work in the front line,” he said. “I have no knowledge other than my own gut feeling, being someone who uses these agents.”

“And in the second-line setting,” he added, “perhaps other studies will show that pembrolizumab does have activity in the PD-L1-negative group, and it will broaden it's approval there, as well.”

It’s not just PD-L1 expression, however. Tumor-infiltrating lymphocytes, tumor mutational burden, immune gene signature, and several other factors are all being considered as candidate biomarkers, which make for an exciting but uncertain diagnostic outlook.

“We’re moving to a place where, at the outset of diagnosis, patients are going to get tested for different mutational status, PD-L1 biomarker, cellular infiltrate, etcetera, and all of these biomarkers are going to be part of the standard workup that helps oncologists determine what drugs to give and in what sequence,” said Ms. Karlsberg. “PD-L1 status will be essential so that we can plan for the entire course of treatment.”

  

Figure 4. ~60% of physicians who are not currently using PD-L1 testing
do not plan to start testing in the next 3 months.


Conclusions

Although PD-L1 biomarker status may not change treatment decisions today, its prognostic value can improve a clinician’s ability to counsel patients, and its influence on the treatment of tomorrow is just beginning to be realized.



“We see great opportunities to do a better job treating patients, selecting therapies, and sequencing treatment for lung cancer patients based on PD-L1 biomarker status and mutational testing,” said Ms. Karlsberg. “Regardless of front line or second line, having that information results in better conversations for our patients… I'm very optimistic that this novel class of agents and the explosion of available information are providing a lot hope for new patients.”

Or as Dr. Herbst put it: “Knowledge has become a standard therapy along with chemotherapy, radiation therapy, surgery, and targeted therapy. Now we have to figure out how to work them all together—and in what sequence.”


Meet the Expert





Sharon Karlsberg, a Principal with ZS Associates for 10+ years, works on a variety of marketing and commercial strategy issues with leading biopharmaceutical manufacturers. Her experience includes qualitative and quantitative market research, brand strategy, competitive readiness, scenario planning, and new product launch. Sharon’s therapeutic area expertise spans many specialty markets, with significant depth in oncology and hematological malignancies. Sharon is one of the founding leaders of ZS’ Oncology Vertical, a team dedicated to addressing sales, marketing and operations challenges for oncology drug and diagnostic manufacturers.

Sharon joined ZS in 2004. Previously, she worked in market research for the medical diagnostics industry, public and investor relations for the biotechnology sector, and a product marketing role at Boston Scientific, Neurovascular. Sharon holds an MBA and Certificate in Health Management from UC Berkeley's Haas School of Business, and a BA in Genetics and Cell Biology from Dartmouth College. 

Contact ZS