ASCO 2018 Highlights: GI and Hepatic Toxicities of Immune Checkpoint Inhibitor Therapy

June 2, 2018

In this podcast, ASCO Educational Book contributor Dr. Shilpa Grover, from Brigham and Women’s Hospital, describes GI and hepatic AEs encountered in patients on ICI therapy.

As part of our coverage of the 2018 American Society of Clinical Oncology meeting, being held June 1–5 in Chicago, we spoke with Shilpa Grover, MD, MPH, about gastrointestinal and hepatic toxicities associated with immune checkpoint inhibitor antibody therapy. Dr. Grover is founder and director of the Onco-Gastroenterology Program at the Brigham and Women’s Hospital in Boston. She and her colleagues contributed a chapter to ASCO’s 2018 Educational Book that includes algorithms that help oncologists manage toxicities related to this type of cancer treatment.

-Interviewed by Anna Azvolinsky


Cancer Network: First, can you briefly talk about how immune checkpoint inhibitors, the anti-PD1 [programmed death 1] and anti-PD-L1 [programmed death ligand 1] antibodies, are currently used? There are quite a few tumor types for which these therapies are now approved.

Dr. Grover: Yes, that is correct. Immune checkpoint inhibitors have demonstrated efficacy in treating a number of different cancers. Since the initial approval of the anti-CTLA-4 [cytotoxic T-lymphocyte-associated antigen 4] antibody ipilimumab for the treatment of metastatic melanoma, the indications for immune checkpoint inhibitor treatments have been increasing. Immune checkpoint inhibitors are now also used to treat non–small-cell lung carcinoma, renal and bladder cancer, Hodgkin lymphoma, mismatch-repair–deficient solid tumors, hepatocellular carcinoma, and other types of malignancies.

What’s different about immune checkpoint inhibitors is that, unlike chemotherapy or targeted therapy, they don’t aim directly at cancer cells but rather they counteract immune inhibitory pathways. By doing so, they induce or enhance immune responses directed against specific antigens present on cancer cells. However, using the immune system in this way against cancer has its downsides as well. Activated T cells can cross-react with host antigens and cause a variety of immune-related adverse events.

Cancer Network: In general, the spectrum of toxicities that patients experience when on a checkpoint inhibitor can be quite different from toxicities experienced when a patient is receiving chemotherapy or targeted therapy. How are patients generally monitored?

Dr. Grover: Most immune-related adverse events usually occur in patients within several weeks to 3 months of starting treatment. Unlike toxicities from chemotherapy, they last for weeks to months. What’s also important to remember is that immune-related adverse events can occur even after patients complete treatment with an immune checkpoint inhibitor. These immune-related adverse events most commonly affect the gastrointestinal tract, lung, skin, and endocrine glands. In comparison, renal, cardiovascular, and neurologic immune-related adverse events are less common. In clinical trials of patients treated with ipilimumab, diarrhea is reported in up to 30% of patients and colitis occurs in approximately 12% of patients.

Patients treated with either a PD-1 or PD-L1 antibody have a lower incidence of colitis, about 4%, and symptoms are usually milder. The incidence is highest with the combination of an anti-CTLA-4 and anti-PD-1/PD-L1 antibodies. Approximately one-quarter of these patients develop colitis. While most immune-related adverse events are considered mild or moderate in severity, these can be serious or life-threatening. This is especially true when a combination of immune checkpoint inhibitors is used. So patients with immune-related adverse events require close monitoring and early evaluation to rule out alternative causes.

Cancer Network: What are the symptoms of gastrointestinal toxicities that patients can experience when being treated with these antibodies? Are certain patients more susceptible to these toxicities?

Dr. Grover: Patients with colitis usually present with frequent watery diarrhea. Bloody diarrhea is relatively rare. In patients with involvement of the small intestine, symptoms can include nausea, vomiting, early satiety, and bloating. These symptoms tend to occur early in the course of therapy, so approximately 6 to 8 weeks after initiation of therapy. In terms of who develops immune-related adverse events, first let me talk about immune-related adverse events in general and then I’ll specifically address GI-specific immune-related adverse events and what we know are the predictors of toxicity.

The correlation between dose and toxicity has not been clearly established for all checkpoint inhibitors, and there might be [toxicity] differences between different checkpoint inhibitors. For example, dose-related adverse effects induced by CTLA-4 inhibitors have been noted in a phase I study of ipilimumab and in a subsequent large meta-analysis. But a similar correlation has not been established with PD-1/PD-L1-directed therapies.

There are currently many ongoing studies to identify biomarkers related to patient characteristics, environmental factors, and tumor composition that could predict the development of immune-related adverse events. More recently, the gut microbiota has been identified as a potential predictor of response to PD-1 inhibitors, and the use of antibiotics has been found to negatively impact the efficacy of PD-1/PD-L1 inhibitors. Whether differences in gut microbiota are associated with immune-related adverse events is not known at the present time.

In terms of the rate of organ-specific immune related adverse events, there may be some variation depending on the primary tumor type or the site of the metastatic disease. For example, patients with non–small-cell lung cancer appear to have a higher risk of immune-related pneumonitis resulting from treatment with PD-1 inhibitors as compared to patients with melanoma. Whether this is true for patients who experience colitis and hepatitis is not known.

IL-17 levels at pretreatment and during treatment have been found to correlate with the incidence of grade 3 diarrhea and colitis in melanoma patients treated with ipilimumab. Conversely, patients with lower pretreatment IL-6 levels appear to be at higher risk to develop grade 3/4 toxicities related to ipilimumab. In addition, the use of nonsteroidal anti-inflammatory drugs may also increase the risk of colitis.

Cancer Network: And what about hepatic toxicities? What symptoms do patients experience?

Dr. Grover: Hepatotoxicity with immune checkpoint inhibitors usually results in transaminitis, which is an elevation of aspartate aminotransferase (AST) and alanine aminotransferase (ALT) [levels] and, less commonly, hyperbilirubinemia. Patients are usually asymptomatic. However, some patients may have fever, fatigue, or jaundice. In rare cases, fulminant hepatitis has been reported. The onset of hepatitis with immunotherapy is usually 8 to 12 weeks after initiation of treatment.

However, elevations have been reported as early as 8 days and up to 21 months after initiation of treatment. The incidence of hepatitis is lower than that of colitis. But again, similar to patients with colitis, the incidence is lower with monotherapy and highest with combination therapy.

It’s estimated that up to 30% of these patients develop immune-related hepatitis after treatment with combination therapy. So patients being treated with immune checkpoint inhibitors require periodic monitoring of liver tests. Similar to patients with other immune-related adverse events, diagnostic evaluation is necessary to rule out alternative etiologies.

Cancer Network: Are there any other gastrointestinal toxicities associated with immune checkpoint antibodies that we haven’t discussed yet?

Dr. Grover: Immune checkpoint inhibitors are associated with acute pancreatitis, but the incidence is very low. Most often, patients may develop pancreatic enzymes elevation without acute pancreatitis. The significance of [this] is unclear. Elevations in pancreatic enzymes may not necessarily be due to underlying pancreatic inflammation, but rather, due to T-cell–mediated inflammation of other organs that produce these enzymes.

It is also important to consider other non–immune-related causes, such as pancreatic duct obstruction from metastatic disease and renal failure, which can delay clearance of these enzymes. It is important that routine assessment of pancreatic enzymes be limited to patients with symptoms suggestive of acute pancreatitis. Patients with pancreatic enzyme elevations without abdominal pain or without evidence of acute pancreatitis on imaging should be monitored clinically and do not require immunosuppressive therapy.