For the first time, the FDA has approved a drug for use in cancer—of any type—that harbors certain molecular features. Merck’s Keytruda, an immune oncology drug, may be prescribed for any resistant, metastatic tumor with microsatellite instability (MSI) or other evidence for defective DNA mismatch repair.
This is good news for patients. Previously, Keytruda (pembrolizumab) was approved by the FDA for use in some forms of lung cancer, melanoma, head and neck cancer, and Hodgkin’s lymphoma. Now, patients can try this medication if they have advanced cancer of any form with pathological MSI or DNA mismatch repair defects. Microsatellite instability most often appears in colon cancers, affecting around 15% of cases. Variants of DNA mismatch repair genes are implicated in heritable cancer dispositions such as Lynch syndrome.
The FDA’s accelerated approval of this drug might surprise traditional oncologists. It suggests the agency may be ditching an archaic system for classifying cancers based on body parts—like breast or liver or colon cancer—and instead will focus on molecular aspects of malignancies, qualities that render tumors vulnerable, or not, to targeted drugs.
At this point, Keytruda can hardly be called an experimental drug; it’s been prescribed to tens of thousands of cancer patients; oncologists are familiar with its side effects. It’s very reasonable for patients with refractory cancer and MSI or mismatch repair syndromes to try it, if they choose. The FDA cites data compiled from five non-randomized studies that support this drug’s effectiveness in various tumor types.
The foreseeable problems include costs. The ballpark price of Keytruda, an antibody given by intravenous infusion, is around $150,000 per year. And toxicity from this drug, an antibody to PD-1, is significant. Yet for some cancers, patients have tolerated this drug well and taken it for years, with prolonged responses and good quality of life; for a small fraction of all cancer patients, it’s a terrific drug.
The main issue with this FDA nod is physicians’ judgment. This is a good example of when an oncologist’s clinical experience will be essential, because Keytruda will only help a fraction of those patients who try it. Knowing which patients are good candidates for this medication, in the sense that it’s likely to help them, and once the drug is prescribed, being willing to “call it quits” if it’s not working in an individual patient’s case will be key—to curbing costs, minimizing toxicity and permitting those who benefit to continue receiving it.
A final concern I have is how cancer specimens were tested for MSI or DNA mismatch repair. The FDA notice refers to several methods of genetic analysis. Distinct ways of checking tumors for these abnormalities could yield variable results. This approval points to the need of accurate diagnostic tests for cancer samples, for precision oncology to move forward in a way that’s economical and most helpful to patients.
