Sunrise Over a New Horizon: A Coming Paradigm Shift in Adjuvant Melanoma

by Arnold DuBell | Sep 14, 2017

Co-authored by Stephanie Hawthorne, Ph.D.


Less than a decade ago, melanoma ranked near the top of the list in terms of cancers with high unmet need, driven by the poor survival outcomes [two-year survival of 25% for Stage IV disease (1)], very limited and ineffective treatment options (dacarbazine and temozolomide were the de facto standards of care, and yet never demonstrated any clinically meaningful activity in the disease), and numerous failed studies that had left the melanoma community with a paucity of new options on the horizon.  The tides began to change in 2011 when the CTLA-4 inhibitor Yervoy® (ipilimumab, Bristol-Myers Squibb / Ono Pharmaceuticals) was approved by the FDA and EMA for metastatic melanoma patients.  That first approval was followed quickly by several additional new targeted therapies that forever changed the landscape of metastatic melanoma and arguably moved melanoma several rungs down the proverbial unmet needs ladder.   Within five years there suddenly were eight targeted therapies (two BRAF inhibitors, two MEK inhibitors, one CTLA4 inhibitor, two PD-1 inhibitors, and one oncolytic virus) approved for this disease, and two-year overall survival in clinical trials is now exceeding 60% (2). 

Like many therapeutic advances for metastatic cancer, these agents have now turned their attention toward earlier stage disease, in the hopes of not just prolonging survival for patients but perhaps achieving the holy grail of actually curing patients of their cancer.  Although it may be too soon to truly use the word cure, the melanoma community may be getting closer day by day to achieving this goal.  The first step in this direction was achieved with the EORTC 18071 trial, which demonstrated a significant improvement in relapse-free survival (RFS) and ultimately overall survival (OS) favoring Yervoy over placebo in the adjuvant setting for resected Stage III melanoma patients (3).  In Monday’s Presidential Symposium III at the European Society of Medical Oncology (ESMO), new clinical data has staked even more footsteps in the sand – the first results for Phase III trials of Zelboraf® (vemurafenib, Roche/Daiichi Sankyo, Tafinlar® (dabrafenib, Novartis) plus Mekinist® (trametinib, Novartis), and Opdivo® (nivolumab, Bristol-Myers Squibb / Ono Pharmaceuticals).

The BRIM8 trial enrolled 498 melanoma patients into two cohorts - Cohort 1, 314 patients with Stage IIC, IIIA (with 1+ nodal lesions >1mm), or IIIB disease; Cohort 2, 184 patients with IIIC disease (4).  Patients were then randomized (2:1) to treatment with 1-year of oral daily Zelboraf or placebo.  In the pooled intent-to-treat population (e.g., representing both cohorts), Zelboraf improved disease-free survival (DFS) by 35%, with a two-year DFS of 62% in the Zelboraf arm and 53% in the placebo arm (HR 0.65, p=0.0013).  However, the trial was designed with hierarchical testing prespecifying DFS for Cohort 2 must reach significance (p≤0.05) in order for any of the other analyses to be considered for statistical significance.  This approach was chosen to address the researchers’ concern that the patients in this cohort would drive the analysis given these patients having faster events than those patients in cohort 1.  In Cohort 2, there was no significant benefit in DFS between the two arms (HR 0.80, p=0.2598).  Although there was an early and strong separation of the curves in the Stage IIIC Cohort 2, the curves began to converge after treatment was stopped at one-year (1-year DFS: 79% versus 58%) and by year two they were overlapping (2-year DFS: 46% versus 48%).  Due to this statistical design of the trial, the benefit observed in the ITT population and the benefit observed in the Stage IIC-IIIB Cohort C (HR 0.54, p=0.0010) were not deemed statistically significant.  These negative results were somewhat disappointing, and yet at the same time not surprising.  BRAF inhibitor monotherapy in the metastatic setting offers a short-term benefit to patients, but ultimately the disease nearly always recurs.  Based on the results of the BRIM8 trial, this appears to also be the case with BRAF inhibition in non-metastatic disease.

In the metastatic setting, the faults of BRAF inhibitor monotherapy have been somewhat overcome through dual blockade of the RAS-RAF-MEK pathway, by using tyrosine kinase inhibitor (TKI) combinations of BRAF inhibitors and MEK inhibitors; indeed, two such combinations are FDA approved for use in first-line BRAF mutant melanoma.  The Phase III COMBI-AD trial (5)sought to evaluate the efficacy of such a combination approach in non-metastatic disease, by enrolling 870 patients with completely resected Stages IIIA (with 1+ nodal lesions >1mm), IIIB, or IIIC cutaneous melanoma and randomizing them 1:1 to one-year of treatment with either Tafinlar + Mekinist or matched placebos.  The results of this study showed a profound clinical benefit favoring the TKI doublet, with two-year RFS of 67% in the TKI arm and 44% in the placebo arm (HR 0.47, p=0.0000000000000153).  The lead author and presenter, Dr. Axel Hauschild commented that “this is the best Hazard Ratio I have ever seen in an adjuvant melanoma trial.”  Unlike the BRIM8 trial, the two curves in this study separated early and remained separated (3-year RFS: 58% versus 39%), and all patient subsets enjoyed the same magnitude of clinical benefit as in the ITT population.  Secondary endpoints of freedom from relapse (FFR, 2-year: 67% versus 44%, HR 0.47, p<0.001) and DSFS (2-year: 77% versus 60%, HR 0.51, p<0.001) also significantly favored the TKI arm.  Overall survival was immature, but trended strongly in favor of benefiting the TKI combination (2-year: 91% versus 83%, HR 0.57, p=0.006; not significant according to O’Brien-Fleming criteria for this trial).  Toxicities were manageable, but there was an increase in grade 3-4 toxicities in the TKI doublet arm (41% versus 14%).  Common toxicities of all grades included pyrexia (63% versus 11%), fatigue (47% versus 28%), nausea (40% versus 20%), headache (39% versus 24%) and chills (37% versus 4%).

Finally, the session ended with the results of the CheckMate-238 trial, which compared Opdivo (3 mg/kg q2w) versus Yervoy (10 mg/kg q3w for 4 doses, then q12w) in 906 melanoma patients with fully resected Stage IIIB, IIIC, or IV melanoma (6).  This trial also met its primary endpoint of improving RFS, with 18-month RFS of 66% in the Opdivo arm and 53% in the Yervoy arm (HR 0.65, p<0.0001).  This benefit accrued in most patient subtypes, including PD-L1-positive (expression level ≥5%; HR 0.50) and –negative patients (HR 0.71), BRAF mutant (HR 0.72) and wildtype patients (HR 0.58), and Stage III and IV disease (although the sample size for the latter group was quite small, and the IV M1c group appeared not to derive any benefit).   The secondary endpoint of DMFS was also improved (HR 0.73, p=0.0204).   Overall survival was immature and not presented.  The incidence of treatment-related grade 3-4 toxicities was improved with Opdivo (14% versus 46%), as well as the adverse events of any grade leading to discontinuation (8% versus 42%). Select adverse events of any grade included skin toxicities (46% versus 60%), gastrointestinal (25% versus 48%), and hepatic (9% versus 21%).

Taken together, these trials represent a significant advance forward in the melanoma field.  Yervoy paved the way for this advance with its 2015 FDA approval in the adjuvant setting, but the positive results in the COMBI-AD and CheckMate-238 trials now surpass what Yervoy initially offered – from the perspective of both efficacy and safety.  Multiple discussants for this session commented on the paradigm-shifting significance of these results, and their attention was primarily drawn to the PD-1 inhibitor results.  It is nearly impossible to compare the results of these two trials with one another due to the inherent differences in patient demographics - COMBI-AD did not include any Stage IV patients but CheckMate-238 did; COMBI-AD enrolled only BRAF mutant patients but CheckMate-238 enrolled BRAF mutant and wildtype; COMBI-AD used a placebo control but CheckMate-238 used an active control, Yervoy. 

It seems evident that Opdivo will become the new standard of care in adjuvant melanoma for BRAF wildtype patients, but what about in BRAF mutant patients, who will soon have these two options to choose from?  Favoring use of Opdivo in the adjuvant BRAF mutant setting is the biologic rationale that suggests use of immunotherapy may be most effective when used early, as well as the “hype” that PD-1 inhibitors have elicited in the metastatic setting.  Favoring use of Tafinlar + Mekinist in the adjuvant BRAF mutant setting is its oral nature (which may be considered more convenient for the patient).  Safety will be a concern and potentially decisive factor, although weighing the safety of these two treatment options may be an individualized decision based on the patients willingness and ability to tolerate specific adverse events.  Cost of therapy could also play a role, potentially favoring intravenous Opdivo if it means lower patient copays than the oral TKIs.  Ultimately, of course, physicians will scrutinize the trial demographics and subset outcomes from these two studies to look for any guidance on which regimen may be superior in a specific patient type (although that may be a fruitless task).  Additionally, thoughts regarding future sequencing of therapies (in the event that a patient did relapse after adjuvant therapy) may factor into the decision of whether to use PD-1 inhibition or BRAF/MEK inhibition in the adjuvant setting, and which mechanism of action to keep in the back pocket in case of relapse. 

Whichever regimen ultimately becomes standard of care, the adjuvant setting is about to expand considerably.  Currently, only one-third of U.S. Stage IIB-III melanoma patients receive adjuvant therapy (7). Historically, this was related to the lack of definitive proof of benefit with pre-existing adjuvant options (predominantly interferon), and more recently also limited by the tolerability of Yervoy in the adjuvant setting.  The better safety profile and stronger efficacy benefits demonstrated for Opdivo and for the Tafinlar-Mekinist combination could lead to an overall increase in the number of melanoma patients who receive adjuvant therapy.  In turn, an increase in use of adjuvant therapy with these effective agents could actually shrink the metastatic melanoma market as fewer patients recur with advanced disease.



  1. CancerMPact® Patient Metrics. Kantar Health. Available from www.cancermpact.com. Accessed 11 Sep 2017.
  2. Larkin J, Chiarion-Sileni V, Gonzalez, R, et al., "CT075- Overall survival (OS) results from a phase III trial of nivolumab (NIVO) combined with ipilimumab (IPI) in treatment-naïve patients with advanced melanoma (CheckMate 067)." Proc Amer Assoc Canc Res; Abstract CT075, 2017.
  3. Eggermont AMM, Chiarion-Sileni V, Grob J, et al., "Prolonged survival in Stage III melanoma with ipilimumab adjuvant therapy." N Eng J Med, 375: 1845-1855, 2016.
  4. Lewis K, Maio M, Demidov L, et al.; “BRIM8: a randomized, double-blind, placebo-controlled study of adjuvant vemurafenib in patients (pts) with completely resected, BRAFV600+ melanoma at high risk for recurrence;” Proc Eur Soc Med Oncol, Abstract LBA7_PR, 2017.
  5. Hauschild A, Santinami M, Long GV, et al.; “COMBI-AD: Adjuvant Dabrafenib (D) Plus Trametinib (T) for Resected Stage III BRAFV600E/K–Mutant Melanoma;” Proc Eur Soc Med Oncol, Abstract LBA6_PR, 2017.
  6. Weber J, Mandala M, Del Vecchio M, et al.; “Adjuvant therapy with nivolumab (NIVO) versus ipilimumab (IPI) after complete resection of stage III/IV melanoma: a randomized, double-blind, phase 3 trial (CheckMate 238);” Proc Eur Soc Med Oncol, Abstract LBA8_PR, 2017.
  7. Kantar Health, CancerMPact® Treatment Architecture U.S. Melanoma, accessed September 11, 2017.

Leave a comment

More than one Google Analytics scripts are registered. Please verify your pages and templates.