Tamara Thomas

Urban Health Today

While PD-L1 inhibitors have changed the paradigm around treating many cancers, patients do not uniformly experience the same outcomes. Oncologist Dr. Andrew Hertler, the chief medical officer emeritus and a senior advisor at specialty and primary care enablement company <a href="https://www.evolenthealth.com/">Evolent</a>, has worked to stay on top of the evolving evidence around these drugs. He recently spoke with Urban Health Today about the need to understand which patients are most and least likely to benefit from them.
 
Urban Health Today (UHT): What has been the impact of PD-L1 inhibitors on the cancer treatment landscape? 
Dr. Hertler: There’s no doubt that PD-L1 inhibitors have significantly changed oncology. Years ago, metastasized malignant melanoma was essentially incurable, with most patients rarely living beyond six months. But now these patients are likely cured by these drugs – they are truly life-changing.
It’s important to note, however, that only about 30% of patients with cancer benefit from the PD-L1 drugs they are given. For the other 70%, they don’t work. They are also very expensive: Three months of therapy may cost about $50,000 just for the drug acquisition costs. So, we need ways to make sure these drugs get to the patients most likely to benefit from them, while helping patients who are highly unlikely to benefit from them avoid the costs and toxicities.
 
UHT: How can you predict which patients will benefit?
Dr. Hertler: One of our best tools is testing for biomarkers—molecules on the tumor that indicate the mechanisms driving a cancer’s growth. Some cancer cells produce the PD-L1 protein, which binds to receptors on the body’s T cells and dampens the immune system’s response to the disease.
Testing for this biomarker involves looking at tumor cells through a microscope and counting the ones that express the PD-L1 protein on special stains. It’s not a perfect test, but we can infer that the higher the expression, the more likely it is that patients will respond positively to PD-L1 inhibitors. If there’s no PD-L1 present, in theory, these drugs are going to be completely ineffective because there’s no protein for them to block.
 
UHT: Has the evidence supported that logic?
Dr. Hertler: Yes. For example, consider Keytruda and Opdivo, two PD-L1 inhibitors, in the treatment of gastric and esophageal cancer. Last fall, when an FDA advisory commission reviewed phase 3 clinical trials data for these indications, they found that patients who were PD-L1 negative — meaning they had zero expression of the protein — had virtually no benefit from the therapy when compared to standard chemotherapy. As a result, the committee recommended against continued approval of the drugs for these patients, and the FDA later adopted this recommendation. These results are meaningful to many patients, as about 25% of all patients with stomach cancer and about 50% of those with esophageal cancer are PD-L1 negative.
PD-L1 drugs are less toxic than chemotherapy. But they still do have side effects, such as skin rashes and dermatitis, colitis, hepatitis, hypothyroidism, adrenal insufficiency, diabetes, pneumonitis, and rarely, myocarditis and nephritis. So, giving these drugs to patients with zero PD-L1 expression risks creating more harm than benefit.
 
UHT: If the response was so poor, how did the drugs get approved in the first place? Was that a bad decision?
Hertler: Not in my opinion. The FDA moves fairly quickly through its accelerated approval pathway to get therapies to market and help patients with devastating diagnoses. The data they had at that time for these drugs showed that, across all patients, there was clearly a benefit — patients with stomach and esophageal cancers were living longer. But that accelerated approval comes with a stipulation for follow-up clinical trials. When the data came in and we could look more closely at how different patient subgroups fared, there just wasn’t a benefit there for PD-L1 negative patients. So I think the FDA also made the right decision to act again and withdraw approval for this group.
 
UHT: What do you think will be the impact of this decision to limit the use of these drugs?
Dr. Hertler: First and foremost, limiting the PD-L1 therapy protects patients from harm. It protects patients from side effects, and I don’t believe we are denying any significant number of patients benefits because we can’t show that there currently is any benefit. That’s not to say that there aren’t any isolated incidences of PD-L1 negative patients who have benefited from it.
But when that happens, we have to wonder if the testing was accurate. Second, limiting this therapy frees up more healthcare dollars to be used in other areas of healthcare and public health initiatives. Using expensive therapies that are unlikely to benefit a patient can limit the delivery of other healthcare and public health initiatives that actually have the potential to help a broader population.
 
UHT: Are there other tools that can help to clarify which patients might benefit from PD-L1 inhibitors?  
Dr. Hertler: Machine learning and predictive models have shown promise. Based on a patient’s clinical parameters, including but not limited to biomarkers, AI tools can calculate the chance of a patient responding to a certain therapy. One of them, called LORIS, has proven to be remarkably accurate for predicting the likelihood of response to immune checkpoint inhibitors, which include PD-L1 inhibitors.
If it’s predicted that a patient only has a 5% chance of responding, they would most likely only experience side effects and get no benefit from the drug. Therefore, we would want to give them a different type of cancer treatment. When the chances of response are 10% or less, that would be no better than a placebo like a sugar pill.
 
UHT: Are there other therapies that might offer hope for patients with gastric or esophageal cancer who are PD-L1 negative?  
Dr. Hertler: There are already other targeted therapies, such as trastuzumab deruxtecan for HER2-positive tumors. I’ve also seen a few approaches that show promise.
CAR-T cell therapy, which targets specific antigens on cancer cells, is one of them. This involves modifying a patient’s T-cells to better recognize and attack cancer cells. Currently, CAR-T cell therapy is used in hematologic malignancies — leukemias, lymphomas, etc., but it is being explored for solid tumors like esophageal and gastric cancer. We are still waiting to see the results.
Another therapy we’re watching is bispecific antibodies. These therapies have “arms” and can bind to two different proteins. These may bind to a cancer cell but also bind to a white blood cell which could bring it next to the cancer cell and kill it. Research here is ongoing to develop bispecific antibodies for esophageal and gastric cancers.
Lastly, there are antibody-drug conjugates, which bind an antibody with a chemotherapy drug, allowing for targeted delivery of the drug to the cancer cells. This aims to minimize damage to healthy cells while effectively killing cancer cells.
 
UHT: Thank you for sharing your knowledge on this topic, Dr. Hertler.

While PD-L1 inhibitors have changed the paradigm around treating many cancers, patients do not uniformly experience the same outcomes. Oncologist Dr. Andrew Hertler, the chief medical officer emeritus and a senior advisor at specialty and primary care enablement company <a href="https://www.evolenthealth.com/">Evolent</a>, has worked to stay on top of the evolving evidence around these drugs. He recently spoke with Urban Health Today about the need to understand which patients are most and least likely to benefit from them. Urban Health Today (UHT): What has been the impact of PD-L1 inhibitors on the cancer treatment landscape? Dr. Hertler: There’s no doubt that PD-L1 inhibitors have significantly changed oncology. Years ago, metastasized malignant melanoma was essentially incurable, with most patients rarely living beyond six months. But now these patients are likely cured by these drugs – they are truly life-changing.It’s important to note, however, that only about 30% of patients with cancer benefit from the PD-L1 drugs they are given. For the other 70%, they don’t work. They are also very expensive: Three months of therapy may cost about $50,000 just for the drug acquisition costs. So, we need ways to make sure these drugs get to the patients most likely to benefit from them, while helping patients who are highly unlikely to benefit from them avoid the costs and toxicities. UHT: How can you predict which patients will benefit?Dr. Hertler: One of our best tools is testing for biomarkers—molecules on the tumor that indicate the mechanisms driving a cancer’s growth. Some cancer cells produce the PD-L1 protein, which binds to receptors on the body’s T cells and dampens the immune system’s response to the disease.Testing for this biomarker involves looking at tumor cells through a microscope and counting the ones that express the PD-L1 protein on special stains. It’s not a perfect test, but we can infer that the higher the expression, the more likely it is that patients will respond positively to PD-L1 inhibitors. If there’s no PD-L1 present, in theory, these drugs are going to be completely ineffective because there’s no protein for them to block. UHT: Has the evidence supported that logic?Dr. Hertler: Yes. For example, consider Keytruda and Opdivo, two PD-L1 inhibitors, in the treatment of gastric and esophageal cancer. Last fall, when an FDA advisory commission reviewed phase 3 clinical trials data for these indications, they found that patients who were PD-L1 negative — meaning they had zero expression of the protein — had virtually no benefit from the therapy when compared to standard chemotherapy. As a result, the committee recommended against continued approval of the drugs for these patients, and the FDA later adopted this recommendation. These results are meaningful to many patients, as about 25% of all patients with stomach cancer and about 50% of those with esophageal cancer are PD-L1 negative.PD-L1 drugs are less toxic than chemotherapy. But they still do have side effects, such as skin rashes and dermatitis, colitis, hepatitis, hypothyroidism, adrenal insufficiency, diabetes, pneumonitis, and rarely, myocarditis and nephritis. So, giving these drugs to patients with zero PD-L1 expression risks creating more harm than benefit. UHT: If the response was so poor, how did the drugs get approved in the first place? Was that a bad decision?Hertler: Not in my opinion. The FDA moves fairly quickly through its accelerated approval pathway to get therapies to market and help patients with devastating diagnoses. The data they had at that time for these drugs showed that, across all patients, there was clearly a benefit — patients with stomach and esophageal cancers were living longer. But that accelerated approval comes with a stipulation for follow-up clinical trials. When the data came in and we could look more closely at how different patient subgroups fared, there just wasn’t a benefit there for PD-L1 negative patients. So I think the FDA also made the right decision to act again and withdraw approval for this group. UHT: What do you think will be the impact of this decision to limit the use of these drugs?Dr. Hertler: First and foremost, limiting the PD-L1 therapy protects patients from harm. It protects patients from side effects, and I don’t believe we are denying any significant number of patients benefits because we can’t show that there currently is any benefit. That’s not to say that there aren’t any isolated incidences of PD-L1 negative patients who have benefited from it.But when that happens, we have to wonder if the testing was accurate. Second, limiting this therapy frees up more healthcare dollars to be used in other areas of healthcare and public health initiatives. Using expensive therapies that are unlikely to benefit a patient can limit the delivery of other healthcare and public health initiatives that actually have the potential to help a broader population. UHT: Are there other tools that can help to clarify which patients might benefit from PD-L1 inhibitors?  Dr. Hertler: Machine learning and predictive models have shown promise. Based on a patient’s clinical parameters, including but not limited to biomarkers, AI tools can calculate the chance of a patient responding to a certain therapy. One of them, called LORIS, has proven to be remarkably accurate for predicting the likelihood of response to immune checkpoint inhibitors, which include PD-L1 inhibitors.If it’s predicted that a patient only has a 5% chance of responding, they would most likely only experience side effects and get no benefit from the drug. Therefore, we would want to give them a different type of cancer treatment. When the chances of response are 10% or less, that would be no better than a placebo like a sugar pill. UHT: Are there other therapies that might offer hope for patients with gastric or esophageal cancer who are PD-L1 negative?  Dr. Hertler: There are already other targeted therapies, such as trastuzumab deruxtecan for HER2-positive tumors. I’ve also seen a few approaches that show promise.CAR-T cell therapy, which targets specific antigens on cancer cells, is one of them. This involves modifying a patient’s T-cells to better recognize and attack cancer cells. Currently, CAR-T cell therapy is used in hematologic malignancies — leukemias, lymphomas, etc., but it is being explored for solid tumors like esophageal and gastric cancer. We are still waiting to see the results.Another therapy we’re watching is bispecific antibodies. These therapies have “arms” and can bind to two different proteins. These may bind to a cancer cell but also bind to a white blood cell which could bring it next to the cancer cell and kill it. Research here is ongoing to develop bispecific antibodies for esophageal and gastric cancers.Lastly, there are antibody-drug conjugates, which bind an antibody with a chemotherapy drug, allowing for targeted delivery of the drug to the cancer cells. This aims to minimize damage to healthy cells while effectively killing cancer cells. UHT: Thank you for sharing your knowledge on this topic, Dr. Hertler.

Lou Portero

Cannabis Use Disorder Linked to Higher Mortality in Colorectal Cancer Patients

Colorectal Cancer

Barry Greenberg, PhD

Monique Nugent, MD, MPH

Van Ta Park, PhD, MPH

Julie Zissimopoulos, PhD

A Personalized Approach Can Transform Alzheimer's Disease Care

Alzheimer's & Dementia

Keightley Amen, BA, ELS

Diabetes Self-Management Can Improve Mental and Physical Quality of Life

Diabetes

Latest News

Dr. Andrew Hertler Discusses Patient Selection for Popular Immunotherapy Drugs in Cancer Treatment

Esophageal Cancer

Dr. Andrew Hertler spoke with Urban Health Today about the need to understand which patients are most and least likely to benefit from PD-L1 inhibitors.