Roy S. Herbst, MD, PhD
At the 2014 Multidisciplinary Symposium in Thoracic Oncology in Chicago, Roy S. Herbst, MD, PhD, professor of medical oncology and pharmacology at the Yale Cancer Center, led a general session on what the future holds for immunotherapy in lung cancer. “Oh, is that all?” he joked, noting that since symposium attendees had heard so many presentations on immunotherapy in non-small cell lung cancer (NSCLC) he had decided to present his thoughts in the form of a Top 10 list to avoid repeating prior presentations. He also noted that he would borrow from those other presentations to illustrate his points.
Here are Herbst’s top lessons learned, listed in reverse order:10. These therapies really work.
After noting that much of researchers’ recent work with immunotherapy in melanoma is guiding similar efforts in lung cancer, Herbst pointed out three specific advantages of immunotherapy: specificity, memory (durability) and adaptability. Among others, he singled out Checkmate-063 and Keynote-001 as examples of these qualities.9. It is still unclear what is the most appropriate endpoint.
Herbst noted that symposium attendees heard similar types of study results that had reported a variety of endpoints, then shared a slide from Yale colleague Mario Sznol, MD, that showed 12 endpoints for combinations with CTLA-4 or PD-1 pathway blockade. These ranged from objective response rate (ORR) to aggregate clinical activity to median progression free survival (PFS).8. All checkpoint antibodies are not the same? Or are they?
Herbst’s next slide asked physicians to consider five key differences in activity/toxicity among agents:
7. These agents, while different from chemotherapy, do have unique toxicities.
Antibody isotope (IgG4 versus IgG1versus engineered)
Anti-PD-1 versus anti-PD-L1
“As oncologists, we're all dealing with the ‘-itises’ again,” Herbst said. “Personally, I've had to become much more familiar with endocrine issues than I had been.” He listed a variety of immune-related adverse events from diverse systems including gastrointestinal, renal, pulmonary, dermatologic, hepatic, endocrine, neurologic and ocular.6. The PD-L1 biomarker(s) has some flaws.
“In fact, it’s really a mess,” Herbst said, drawing laughs from his colleagues. “The assay methodology has many, many issues to the point where it is a continuous variable, a constantly moving target for researchers,” he said. “I would go so far as to say that the assay will define the field.” He also noted that immune therapy has been placed in the “no marker” arm of the Lung-MAP trial: “That’s because we really don't know what the marker will be.”5. Science can help drive the show. Biopsies and immune monitoring should be done when possible.
“We can make the clinic our lab. That should be how we design our combinations,” Herbst said. “We can see dynamically when a treatment works by comparing patients’ pre- and post-treatment biopsies.” He also noted that a number of assays can be done to measure cytokines, to multiplex different agents and referred colleagues to the new field of cyTOF analysis, which provides deep profiling of single-cell data.4. Questions remain regarding dose and duration of therapy.
Herbst noted that practicing oncologists face these questions daily: How long should I treat? What is the ideal dose? These questions remain unsettled for immune therapies in particular.3. Combination therapy is a must: the search for other checkpoints should continue.
But what will they be? Herbst reiterated that the clinic must become the lab. Otherwise, without rigorous scientific analysis, he said it will be too easy for combinations to be tried that will lack scientific rigor, thereby having the potential to harm patients.2. Immunotherapy will be used in all lines of therapy.
After showing a list of ongoing phase II/III trials for advanced NSCLC, Herbst then superimposed a text block reading “maintenance, adjuvant, neoadjunvant and combination” as a reminder of how pervasive immunotherapy will become in treating lung cancer.1. It's a horse race!
Herbst admits he has no prediction as to which drug will win the race. “The winning group will be the one with the best drug, the best biomarker, the best strategy and a little bit of luck,” he said. “But really, there will be a clear winner and that’s our patients.”