ICI Max Efficacy Min Risk
Maximizing Efficacy, Minimizing Risk: New Biomarkers and Ongoing Challenges With ICIs

Released: November 30, 2020

Expiration: November 29, 2021

Marianne Davies
Marianne Davies, DNP, CNS, ACNP-BC, AOCNP
John L. Marshall
John L. Marshall, MD

Activity

Progress
1
Course Completed

Immune checkpoint inhibitors (ICIs) continue to revolutionize management of numerous advanced solid tumors. In this commentary, John L. Marshall, MD, and Marianne Davies, DNP, CNS, ACNP-BC, AOCNP, discuss the latest developments in leveraging tumor mutational burden (TMB) as a biomarker to select patients most likely to benefit from ICI-based therapy. Then, they address ongoing challenges to optimal management of immune-related adverse events (irAEs) associated with ICIs.

Using TMB to Identify Patients Most Likely to Benefit From ICIs

John L. Marshall, MD
The most important question when considering ICI-based therapy is: Who should receive them and who should not? There is an ongoing search for biomarkers that can identify which patients are most likely to benefit. We want to avoid exposing patients unlikely to benefit to the risks of irAEs, as my colleague will discuss shortly.

Because many ICIs target components of the PD-1/PD-L1 pathway, clinicians originally thought that PD-L1 must be expressed for a PD-1/PD-L1–targeted ICI to work. However, ICIs have demonstrated activity in many cancers regardless of PD-L1 expression, although patients with high PD-L1 expression as measured by IHC generally benefit more.

The next clinically important biomarker found to predict for ICI benefit is microsatellite instability (MSI). Cancers with defective mismatch repair (dMMR) accumulate mutations in short, repetitive DNA sequences known as microsatellites—which in turn leads to a higher number of neoantigens that “tickle” the immune system into responding against the tumor. In 2017, the FDA issued a landmark tissue/site-agnostic approval for pembrolizumab for treatment of unresectable/metastatic, MSI-high/dMMR solid tumors that had progressed after prior therapy and lacked alternative treatment options. MSI status can be assessed by performing IHC for inactivation of at least 1 of 4 key dMMR proteins (MLH1, MSH2, MSH6, and PMS2). It can also be assessed using next-generation sequencing (NGS), which can also evaluate TMB.

TMB, which is a measure of the total number of mutations per coding area in a tumor’s genome, has emerged as an important biomarker for ICI efficacy. Patients with MSI-high tumors—which also have a high TMB—are more likely to benefit from ICIs. Indeed, a preplanned retrospective analysis of the nonrandomized phase II KEYNOTE-158 trial found that in patients with advanced solid tumors with a TMB ≥ 10 mutations/Mb, the ORR was 29% with pembrolizumab. Half of patients responded for at least 2 years. These results led to a second landmark decision in 2020, when the FDA approved pembrolizumab for treatment of unresectable/metastatic, TMB-high (≥ 10 mutations/Mb) solid tumors that progressed after prior therapy and lack alternative treatment options.

Biomarker Testing: Who, When, and How

John L. Marshall, MD
We are now faced with a clinical conundrum. Because most patients with advanced solid tumors need molecular testing to determine their eligibility for targeted therapies, should we perform limited testing against known actionable mutations (eg, ROS1) and perform NGS later to assess for ICI eligibility? Or should we use NGS to obtain a broad initial molecular profile for all patients with metastatic solid tumors? In our clinic, we choose the latter approach. Although NGS often finds nothing currently actionable, we have found unexpected alterations with treatment implications (eg, BRCA mutations in colon cancer or NTRK fusions). 

The challenges to implementing NGS as the standard initial testing include the need for pathology support, familiarity with the NGS platform, cost, and limited tissue availability. Fortunately, NGS results are generally consistent between primary and metastatic sites. Despite these challenges, using NGS is critical to identify patients most likely to benefit from ICIs, given the expense and safety risks inherent with immunotherapy. Furthermore, these biomarkers can help clinicians manage patients’ expectations for ICI-based therapy, which has generated a great deal of excitement and coverage of high-profile successes like Jimmy Carter’s pembrolizumab treatment for melanoma.

Marianne Davies, DNP, CNS, ACNP-BC, AOCNP
I have also cared for patients with unrealistic expectations based on exciting media reports. We must be clear that not all patients are candidates for ICIs. For example, a patient with a very large tumor burden and rapidly progressing disease cannot wait for ICIs to achieve an antitumor response.

John L. Marshall, MD
Yes. We should also bear in mind that patients can be selected for ICIs using older molecular profiling results that were obtained before newer targeted agents and ICI-based therapies were approved. For example, older test results indicating a PTEN alteration may not have been actionable at the time, but are useful now. Not all genetic findings need to be included in the patient’s history of present illness in their EMR, but it is helpful to log when tests were conducted and key findings.

Marianne Davies, DNP, CNS, ACNP-BC, AOCNP
In my practice, I use the history of present illness field to track tests sent to external vendors, including when they were sent and whether results have returned. This helps us know when test results will be available.

I would caution that it is important to know which platform was used for the sequencing, as these platforms continue to evolve. If a patient received “full sequencing” years ago, it would make sense to resubmit tissue or test additional tissue with current test panels.

John L. Marshall, MD
In some centers, pathologists do not actively stay abreast of testing standards in specific malignancies, and instead rely on the oncology team to order the molecular tests. Oncology clinicians should review the pathology report carefully, because even findings with treatment implications can be consigned to an addendum, and many clinics do not have the infrastructure to identify unusual results and notify physicians.

How Can We Better Manage irAEs?

Marianne Davies, DNP, CNS, ACNP-BC, AOCNP
As ICIs have continued to be more widely used, we have been challenged to better identify and manage irAEs, which are very different than toxicities observed with chemotherapy and targeted therapy. These irAEs result from nonspecific activation of the immune system by ICIs, and can affect all organ systems, with the most common sites being the skin, gastrointestinal tract, lungs, and endocrine system. Of particular clinical relevance are severe, potentially life-threatening immune reactions such as pneumonitis, myocarditis, colitis (with the potential for profuse diarrhea), and those affecting the nervous system. 

Current management strategies rely on immunosuppression, typically with corticosteroids; should patients become refractory to their initial immunosuppressant, clinicians should select subsequent immunosuppressants based on the organ affected and strategies for managing autoimmune diseases affecting that organ. For example, if a patient receiving ICIs develops colitis, clinical management can draw from experience with ulcerative colitis or Crohn’s disease.

Our initial strategy for irAEs is to keep patients on the corticosteroid until the irAE grade decreases, ideally to grade 1, after which we taper the steroids. Is this truly the best strategy? Can irAE management be improved by other means, perhaps with earlier use of alternative immunosuppressants? Trials are just beginning to evaluate irAE management questions to improve outcomes for patients receiving ICIs.

If trials are just now beginning, how have we managed irAEs to date? When ICIs were first approved, the clinicians with the most experience managing irAEs were at the large academic centers that had participated in the early trials. Our experience was largely confined to patients with melanoma and lung cancer—the first indications approved for ICIs. Over time, we have developed consensus guidelines for managing irAEs. However, these recommendations were not based on randomized clinical trials, but on real-world experience managing patients with other autoimmune syndromes.

Community Challenges to Managing irAEs
Thus, a widespread challenge to optimal management of irAEs is simply lack of experience. Now that multiple ICIs are approved and can be used in community care, often community practitioners have not yet gained the experience needed to manage irAEs. Notably, in a 2020 analysis of clinicians’ use of the irAE decision-support tool developed by Clinical Care Options in partnership with the NCCN®, 60% of clinicians reported seeing ≤ 50 patients receiving ICIs per year, with 27% seeing < 10 patients on ICIs. It is incumbent upon those of us with experience managing irAEs to help educate those in the community about these irAEs and the availability of clinical practice guidelines.

Another challenge in community practice can arise from differences in how patients are followed compared with clinical trials, where patients are usually monitored more closely. Patients may develop irAEs between visits, making it important that providers counsel patients on monitoring for and promptly reporting irAE symptoms.

Additional barriers include limitations and financial restrictions affecting access to certain immunosuppressants. Furthermore, community practitioners often do not have rapid, broad access to other specialists for consultation, such as cardiologists, pulmonologists, and dermatologists, who play important roles in helping manage irAEs affecting the relevant organ system.

Importance of Other Clinical Settings
Patients receiving ICIs will sometimes present to urgent care, the emergency department, and even their primary care provider with acute irAEs. Unfortunately, many clinicians working in these settings are underprepared to recognize and manage irAEs. I would recommend that all clinicians ask patients with cancer if they have ever received an ICI, because evaluation and management can be quite different from other similarly presenting conditions. Furthermore, all patients should be counseled to inform any healthcare provider they are seeing about their history of being treated with ICIs.

John L. Marshall, MD
I quite agree! For example, if I saw a patient yesterday and then today, I note that their transaminases increased from 30 to 60 IU/L. Normally, this would not be a big concern. But that same elevation in a patient on ICI-based therapy could be the beginning of an ultimately fatal immune hepatitis.

Marianne Davies, DNP, CNS, ACNP-BC, AOCNP
I agree. Furthermore, while oncology clinicians are trained to recognize symptoms of these toxicities, we are often not the first person to receive the patient’s call, email, or message. We now train our practice nurses and even our clinical secretaries in “telephone triage”—an algorithm that starts with asking if the patient is on ICI-based therapy.

Monitoring and Managing irAEs During COVID-19

John L. Marshall, MD
The standard of care for patients receiving ICIs is careful monitoring for potential irAEs, which includes a detailed history and physical examination prior to administration plus routine liver function tests and endocrinology assessments. However, due to the COVID-19 pandemic, clinicians are doing more remote screening and monitoring this year. This can involve reaching out to the patient or them contacting the clinic. Some patients need to be prompted for more frequent communication, whereas others are hypervigilant and call very frequently. 

Typically, we see patients in person for physical examination, then every 2-3 weeks for chemotherapy, including assessment for the next cycle. Those receiving ICIs can have a 6-week interval between courses, meaning we do not see them or conduct blood tests as often. Whereas AEs with chemotherapies tend to show up relatively early, irAEs usually display a delayed onset, with some patients not experiencing any toxicity for months. Patients need to be counseled on the need for ongoing monitoring and self-reporting.

Marianne Davies, DNP, CNS, ACNP-BC, AOCNP
Finally, I want to emphasize that clinicians and patients must be aware that irAEs can occur at any time. Across different malignancies, irAEs have been reported months after the discontinuation of ICI-based therapy.

Detailed recommendations regarding the management of a variety of organ-based irAEs associated with ICIs have recently been updated by the National Comprehensive Cancer Network® (NCCN®). To receive case-specific guidance from the NCCN Guidelines® on irAE management for unique patient scenarios, please visit the interactive algorithm tool developed by Clinical Care Options in partnership with the NCCN®.

Poll

1.
Which of the following biomarkers are you currently using in your practice to identify patients with solid tumors as candidates for ICI therapy?
Submit