CR 302: Nivolumab doubles the odds of long-term survival

This week at the American Association of Cancer Researchers (AACR) annual meeting, researchers released data regarding the long-term survival of one of the earliest human clinical trials of nivolumab (Opdivo). The study, titled "Durable, long-term survival in previously treated patients with advanced melanoma (MEL) who received nivolumab (NIVO) monotherapy in a phase I trial," shows that patients with metastatic melanoma who took Opdivo more than doubled their chances of surviving five years, compared to all previous therapies. The five-year survival rate for patients with advanced melanoma who got non-immunotherapy  treatments was 16.6% between 2005 and 2011, according to the National Cancer Institute.

This new study is the first to look at long-term survival for an anti-PD-1 immunotherapy. In 2008, Bristol-Myers Squibb began recruiting patients into one of the first Phase I human trials for nivolumab. Over the next four years, 107 patients were enrolled. All had been heavily treated with other therapies, such as chemotherapy, and all had failed. All were facing likely imminent death. The main focus of the study was determining appropriate doses of nivolumab. Different doses were given to different patients. Duration of dosages also varied, and some received nivolumab for as long as four years. Earlier analyses from this study showed little relationship between dose and response -- in other words, if the nivolumab worked, it worked -- and eventually BMS settled on a standardized dose of 3mg/kg. In 2015, the researchers went back and looked at who was still alive. By that time, all of the patients had started nivolumab two years or more.

Following is a chart from the study. OS means Overall Survival. The chart breaks out the subset of patients who received 3 mg/kg of nivolumab, compared to everyone in the study.

OS rates
	NIVO 3 mg/kg (n=17)	All Patients (N=107)
OS rate, % (95% CI)*
12-month	64.7 (37.7-82.3)	62.7 (52.6-71.2)
24-month	47.1 (23.0-68.0)	48.0 (38.1-57.2)
36-month	41.2 (18.6-62.6)	42.1 (32.4-51.4)
48-month	35.3 (14.5-57.0)	34.8 (25.7-44.1)
60-month	35.3 (14.5-57.0)	33.6 (24.6-42.9)

These data show that not everyone responds to nivolumab, most likely because not all tumors have a PD-1 and PD-L1 interaction. The PD-1/PD-L1 blockade is illustrated in a September 28, 2015 article in Drug Discovery and Development. (I've removed the image from this post because the formatting was off.)

According to the AACR study, for those patients who respond and live at least two years, there is a clear plateau in long term survival: 48% were still alive after two years, and it slowly declined to 34% after five years. This includes people who died of all causes, including non-cancer-related causes.

This study received widespread press coverage, including the Wall Street Journal and Washington Post. The Journal said that the results suggested that patients who survive for more than four years "are highly unlikely to relapse. That is essentially an unheard-of result in advanced cancer. The treatments appear to have enabled their immune systems to eradicate or take control of their tumors."

What does this mean for me and other patients with metastatic bladder cancer who are taking nivolumab? Of course, these other studies are of a different cancer, and each cancer has its own tendencies. But because the drug targets the same PD-1/PD-L1 mechanism, it is possible to extrapolate these data to my cancer. It suggests that we should see similar results for other cancers that have that same mechanism, including metastatic bladder cancer.

Patients with mets BC only first received nivolumab in 2014, so it will be six years before we see similar data for us. I'm expecting to see the first published report on nivolumab and mets BC next month, in conjunction with the 2016 AACO meeting. But this is very encouraging news. It suggests that patients like me who have responded to nivolumab have significantly better chances of long-term survival that any other type of non-immunotherapy. And perhaps the response truly will be durable. Dare I think that Opdivo may actually cure my cancer? The data isn't there yet, but there is reason to hope.