'Functional Cure' in Advanced Cancer Suggested by Long-term Data

Liam Davenport

August 06, 2019

Long-term survival data from some of the initial clinical trials with the immune checkpoint inhibitor nivolumab (Opdivo, Bristol-Myers Squibb) show that a proportion of patients with advanced cancer are living for years.

The results from an early phase 1 study, published online by JAMA Oncology July 25, show a "long tail" of overall survival, with a flattening off after 3 years. The 5-year overall survival reached 34% in patients with melanoma, 28% in those with renal cell carcinoma (RCC), and 16% in patients with non-small cell lung cancer (NSCLC).

We can think of that as a functional cure at least. Dr Suzanne Topalian

Lead author Suzanne L. Topalian, MD, Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins Bloomberg–Kimmel Institute for Cancer Immunotherapy, Baltimore, Maryland, discussed the results in an author interview on the JAMA Oncology website.

While talk of a cure may be premature, these results, along with others, show that "some patients can live for years and we can think of that as a functional cure at least," she commented.

Topalian added that other immune checkpoint inhibitors have slightly different mechanisms of actions — some are anti-programmed death ligand 1 (PD-L1) and some are anti-PD-1 drugs.

Nevertheless, she "would expect that the results we're discussing today may be generalizable across the whole drug class."

In an accompanying editorial, Stefan Zimmermann, MD, and Solange Peters, MD, PhD, from the Centre Hospitalier Universitaire Vaudois in Lausanne, Switzerland, say that placing the findings in context with outcomes achieved before the introduction of nivolumab underlines their importance.

"A comparison with 2008 historical controls only highlights the magnitude of the survival gain and the durability of treatment effect," they write.

"The availability of PD-1 blocking strategies will have inexorably modified survival curves in all clinical trials started after 2015 for multiple solid tumors, a fact that needs to be considered when analyzing survival."

One of the First Trials of Nivolumab

These latest results come from a phase 1 study in 270 patients with advanced or recurrent malignancies (known as MDX1106-03 or CA209-003), which was the first multidose trial of nivolumab.

Started in 2008, it initially included previously treated patients with advanced melanoma, RCC, NSCLC, colorectal cancer, and prostate cancer, but the latter two were discontinued as these cancers did not respond to immunotherapy. But the other three cancer types did respond, and all three are now approved indications for nivolumab and other checkpoint inhibitors. 

In their study, the authors present the results of a secondary analysis, based on 270 patients with melanoma (39.6%), RCC (12.6%) and NSCLC (47.8%).

All had received prior treatment: 40% had three or more prior systemic cancer therapies, "so we consider them to be heavily pretreated," Topalian commented.

The patients, from 13 centers in the US, were required to have an Eastern Cooperative of Oncology Group (ECOG) performance status of 0–2, and measurable disease on the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.0 criteria at study entry.

Topalian explained that all patients received nivolumab "every 2 weeks up to a maximum of 2 years if their tumors stabilized or regressed while on treatment."

"Because this was a phase 1 trial, patients were treated in different cohorts with different doses of nivolumab, so the dose range in this trial was between 0.1 mg/kg all the way up to 10 mg/kg."

The median number of doses received was 11 for patients with melanoma, 16 for patients with RCC, and six for those with NSCLC.

The objective response rate, including both partial and complete responses, was 31.8% for patients with melanoma, 29.4% for patients with RCC, and 17.1% for those with NSCLC, with responses lasting for a median of 22.9 months, 12.9 months, and 19.1 months, respectively.

The minimum follow-up was 58.3 months for patients with melanoma, 63.9 months in those with RCC, and 58.3 months in patients with NSCLC.

Median overall survival was 20.3 months for patients with melanoma, 22.4 months for patients with RCC, and 9.9 months for patients with NSCLC.

The team notes that the overall survival curves "showed an inflection toward flattening around 3 years" after starting treatment.

Better Survival in Patients Who Had Adverse Events

Patients who experienced treatment-related adverse events had significantly better overall survival than those did not.

Patients who experienced adverse events with nivolumab had a median survival of 19.8 months, rising to 20.3 months among those with grade ≥3 adverse events. This is more than three times longer that the median survival of 5.8 months in patients who did not have adverse events (P < .001).

"We can reassure our patients, if they develop these side effects, that it may very well put them into a better response effect category," Topalian commented.

"We also know that patients who need to delay their nivolumab treatment or even stop treatment entirely due to side effects still have a very good chance of having durable anti-tumor responses," she added.

Topalian said that this observation is not likely to be the result of a treatment duration bias in which the longer patients stay on treatment, the more they are likely to experience adverse events.

She said: "We examined the rate of side effects that occurred in the first 6 months, or the next several months of treatment, or at the end of the 2 years."

"Surprisingly, we found that most of these side effects tended to occur early on, and that's why we don't think that a time bias is playing into this association of better anti-tumor response with side effects."

Topalian explained, however, that the anti-tumor efficacy of nivolumab cannot be dissociated from its immune-related adverse events "because the mechanism of the drug explains both."

"We're releasing the brakes on the immune response," she explained. "We would like that to be directed against the cancer cells, but sometimes these heightened immune responses can damage normal tissues as well."

"There's a lot of research going on now to try to understand the differences between those two processes, the anti-tumor response or normal tissue inflammation, so that we might be able to treat the side effects without any chance of interfering with the anti-tumor activity," she commented.

The authors also noted that patients with a large tumor burden at baseline had worse survival, and liver or bone metastases were associated with a 70% reduced likelihood of surviving for 5 years.

A good performance status when starting treatment made it 2.75 times more likely to survive out to 5 years compared with less fit patients.

The authors hope that this identification of factors associated with long-term survival "may inform treatment approaches for individual patients and strategies for future clinical trial development in immuno-oncology."

Not Efficient Yet at Predicting Who Will Benefit

Speaking to Medscape Medical News about the editorial, Peters said the survival curves with nivolumab and other immunotherapies have a "long tail" that can include, in the case of melanoma, up to 40% of patients, but physicians are currently "extremely inefficient" in predicting who will be in that tail.

Certain factors can, however, at least suggest who will not be a long-term survivor.

For example, individuals with a poor performance status who have been weakened by their disease and likely have a weakened immune system are less likely to survive long term.

Peters said that another factor is the "famous paradigm of liver metastases."

"Across all these entities, in particular solid tumors, it looks like the liver is an organ where the immune response probably takes place to a lesser degree or with more difficulty, leading to lower immunoresponsive phenotypes in patients suffering from liver metastases."

In lung cancer, she continued, patients with high PD-L1 expression are more likely to have a complete or partial response to immunotherapy, which "immediately makes the patient sit on a better curve, in terms of long-term outcome."

She cautioned: "However, in all these predictive models, you will find, at 5 years, patients who do not fulfill any of these conditions."

For example, the long tail will include 'never smokers,' epidermal growth factor receptor mutated patients, or patients with liver metastases.

Consequently, the negative predictive values indicating which patients will not be long-term survivors are currently "very poor."

"This means that we are still not able to predict which patients should not receive immunotherapy," Peters explained. The result is that, at present,  practically all eligible patients end up receiving the drugs.

In terms of when immunotherapy should be given, Peters noted that the available data show that the best response is seen in patients who are treatment-naïve.

She said that, even in stage IV metastatic disease, the capability of the immune system to respond to the drugs is higher in patients who have not previously been treated.

To understand that further, "we need to really have a deep look into the metastatic trials that had the opportunity to give immunotherapy frontline," as overall survival might increase from 15% to 25% or even 30%, "which is already good news."

The greatest promise of evoking a strong immune response and achieving a long-term benefit with immunotherapy is, however, in early disease, as "the way the immune system can control localized disease is completely different," she commented. This has been shown is several melanoma adjuvant trials.

In other cancer types, the current data come from only a small number of patients, but the magnitude of benefit seen in early disease is clearly much higher than that observed in stage IV patients.

"More than just treatment-naïve patients, I think early disease is the specific niche where immunotherapy might really double its efficacy in the future," Peters explained.

"We need to see more data, but there are many ongoing adjuvant and neoadjuvant trials in solid tumors, starting with melanoma, where the results were unexpectedly better than in advanced disease."

"I think it's going to enrich what we have observed in advanced cancer," she said.

How Long Should Treatment Continue?

Notably, however, the discussion over the optimal treatment duration persists.

Peters noted that a small note of caution was struck by data presented at the ESMO Immuno-Oncology Congress in 2018 by Herbst and colleagues on long-term data from the KEYNOTE-010 trial of pembrolizumab for advanced NSCLC.

They showed that, although most patients had a durable complete remission after 2 years of treatment, 30% of patients with controlled disease at 2 years progressed within a year of discontinuation.

We still don't know when to stop these checkpoint inhibitors. Dr Solange Peters

"Of course it's just observation," Peters said, "but for the time being, it makes me consider that we still don't know when to stop these checkpoint inhibitors, and I'm worried about stopping every time I have this discussion."

The study was sponsored by Bristol-Myers Squibb.

Topalian reported receiving grants from Bristol-Myers Squibb, Compugen, and Potenza Therapeutics during the conduct of the study; receiving travel reimbursements from Bristol-Myers Squibb, Dragonfly Therapeutics, Five Prime Therapeutics, and Merck; receiving consulting fees from AbbVie, Amgen, Avidity NanoMedicines, Bayer, Camden Nexus, DNAtrix, Dragonfly Therapeutics, Dynavax Technologies, Ervaxx, Five Prime Therapeutics, FLX Bio, ImaginAb, Immunomic Therapeutics, Janssen Pharmaceuticals, MedImmune/AstraZeneca, Merck, Pfizer, Rock Springs Capital, and WindMIL outside the submitted work; receiving stock or stock options from Aduro, Dragonfly Therapeutics, Ervaxx, Five Prime Therapeutics, FLX Bio, Jounce Therapeutics, Potenza Therapeutics, Tizona LLC, and WindMIL; having intellectual property licensed through her institution to Bristol-Myers Squibb, Aduro, Immunomic Therapeutics, NexImmune, and WindMIL; and having patents pending for cancer therapy via a combination of epigenetic modulation and immune modulation, checkpoint blockade and microsatellite instability, and biomarkers useful for determining response to PD-1 blockade therapy. Other authors' full disclosures can be found in the original article.

Zimmermann reports personal financial interests for consultancy (Roche), advisory roles (AstraZeneca, Bristol-Myers Squibb, Eli Lilly, MSD, and Novartis), and received travel grants (Amgen, AstraZeneca, Bayer, Roche, Bristol-Myers Squibb, Astellas, Merck KGaA, Vifor), institutional financial interests for contracted clinical research (Amgen, AstraZeneca, Bristol-Myers Squibb, Incyte, Iovance, Kite, MSD, Roche). Peters has received education grants, provided consultation, attended advisory boards, and/or provided lectures for AbbVie, Amgen, AstraZeneca, Bayer, Biocartis, Boehringer Ingelheim, Bristol-Myers Squibb, Clovis, Daiichi Sankyo, Debiopharm Group, Eli Lilly, F. Hoffmann-La Roche, Foundation Medicine, Illumina, Janssen, Merck Sharp & Dohme, Merck Serono, Merrimack, Novartis, Pharma Mar, Pfizer, Regeneron, Sanofi, Seattle Genetics, and Takeda, from whom she has received honoraria.

JAMA Oncol. Published online July 25, 2019. Full text, Editorial

For more from Medscape Oncology, follow us on Twitter:  @MedscapeOnc


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.