Cancer immunotherapy, which uses the patient's immune system to eradicate disease, has been the subject of investigation for over a century. George Bernard Shaw, in his 1906 play, The Doctor's Dilemma, said "There is at bottom only one genuinely scientific treatment for all diseases and that is to stimulate the phagocytes" (white blood cells).
Despite its immense promise, immune therapy has suffered from an incomplete understanding of the intricacies of immune regulation that serve to both increase and decrease immunity in response to bacterial, viral, fungal or malignant insult.
The first breakthrough came with an agent known as Ipilumab.This antibody against a protein on lymphocytes turns “off” the “off switch” to allow lymphocytes to go after their cancerous prey. Ipilumab proved effective against metastatic malignant melanoma, an absolutely lethal disease, but responses came at the expense of toxicities, some quite severe.
The next breakthrough came with the discovery of a different immune regulator known as the PD-1/PDL-1 pathway. This was soon targeted by the antibodies Nivolumab and Pembroluzumab. By unleashing the immune system through this second path, responses were seen not only in melanoma, but also in lung, bladder, kidney, lymphoma and most recently head and neck cancers. Immune therapy appeared to be moving into high gear.
With lung cancer, the leading cause of cancer death in America and over 20% of patients with recurrent disease showing response, Nivolumab was moved to untreated patients in a clinical trial that pitted Nivolumab against chemotherapy. To participate, all patients had to express PD-L1, the target protein for Nivolumab, in at least 5% of their cells.
The results reported this week proved extremely disappointing as Nivolumab patients did not show improved progression-free survival over their chemotherapy counterparts. In the first line setting it appeared that “old fashioned” chemotherapy was equivalent to immunotherapy.
Should we be alarmed? Have we reached the zenith of this seemingly fantastic new form of therapy? Do we need to go back to the drawing board and look for the next breakthrough? I think not.
Like all discoveries in cancer, there is a flurry of interest and sense of exuberance as the first results arrive. Conversations turn to cure and the eradication of the disease. Yet, with each breakthrough comes the sobering reality of scientific investigation that things are rarely as good as they first appear, and over time even the most stunning findings regress to the mean.
This applied in the 1920’s (radiation), the 1950’s (chemotherapy), the 1980’s (monoclonal antibodies), the early 2000’s (targeted agents) as it does today in 2016 (immunotherapy). These breakthroughs are neither panaceas nor failures, just weigh stations on our slow progress toward a cure for cancer.
This clinical trial should not cause us to despair. Immune therapy approaches are brilliant. In those patients for whom they work, they are truly miraculous. I have patients alive after many years using these approaches. Drugs that target immune checkpoints remain outstanding examples of scientific discovery but they are not perfect. They do not work for everyone. They do not work forever. And there is much to learn.
What we need is the ability to better apply these important drugs. Today, there is no reliable method for the selection of treatment candidates. Yes, patients who express the highest levels of PDL-1 appear to do better, but some patients with very high levels of PDL-1 do not respond, while others with no detectable PDL-1 expression can still benefit.
Like radiation, which was the miracle of its time, immune therapy is just the newest addition to our armamentarium. We must now re-double our efforts to use laboratory platforms to better select responsive patients. This will save time, money and toxicity.
The recurring theme is that many cancer treatments work for some patients but no form of cancer treatment works for all patients. Rather than blindly administering expensive and potentially toxic agents, our job is to use validated laboratory models to identify responsive patients before they are treated.
As always, I appreciate your thoughts and comments.