Adjuvant nivolumab produced survival benefits in patients with resected stage IIIA melanoma, where a significant portion of patients were recurrence free at a median follow-up of over 20 months.
Adjuvant nivolumab (Opdivo) produced survival benefits in patients with resected stage IIIA melanoma, where a significant portion of patients were recurrence free at a median follow-up of over 20 months, according to data from a real-world analysis presented at the 19th International Congress of the Society for Melanoma Research.1
Patients treated with adjuvant nivolumab (n = 171) achieved a 12-month recurrence-free survival (RFS) rate of 97.1% (95% CI, 93.1%-98.8%) compared with 81.6% (95% CI, 65.2%-90.8%) for patients who underwent observation (n = 38). The RFS rates at 18 months were 91.4% (95% CI, 85.6%-94.9%) vs 78.6% (95% CI, 61.6%-88.7%), respectively. The median RFS was not reached in either arm.
Additionally, treatment with neoadjuvant nivolumab improved 12- and 18-month overall survival (OS) rates; 98.2% (95% CI, 94.6%-99.4%) and 94.2% (95% CI, 89.1%-96.9%) compared with 94.7% (95% CI, 80.6%-98.7%) and 91.8% (95% CI, 76.6%-97.3%) in the observation group, respectively. The median OS was 34.6 months and not reached, respectively.
“In 2 real-world studies that have been published, nivolumab was shown to provide benefit, however the numbers were small,” Anna C. Pavlick, BSN, MSc, DO, MBA, a medical oncologist at Weill Cornell Medicine, in New York, New York, said in a presentation of the data. “Therefore, we’re presenting the real-world Cardinal Health database [for patients] with resected stage IIIA melanoma treated with nivolumab or observation. This is the largest real-world database to be presented to date.”
The real-world study collected data from Cardinal Health’s propriety Oncology Provider Extended Network.2Data from the phase 3 CheckMate 238 trial (NCT02388906), led to the approval of nivolumab monotherapy for the adjuvant treatment of patients with resected stage III/IV melanoma in 2017.3 However, per AJCC-7 staging criteria, patients with stage IIIA melanoma were excluded and a minimal number of patients per AJCC-8 were enrolled.
Patients in the adjuvant nivolumab cohort (n = 171) and an observation cohort were aged 18 years and older with stage IIIA melanoma per AJCC-8 and had complete surgical resection between January 1, 2018, and December 31, 2019.2
The mean age of patients was 57.4 years in the adjuvant nivolumab cohort and 68.1 years in the observation group; both cohorts had a majority White male population. Year of resection was 30% in 2018 and 70% in 2019 vs 47% in 2018 and 53% in 2019, respectively.
Most patients in both cohorts had a lymph node involvement of 1, but 33% of patients in the nivolumab arm had an involvement of 2 or 3 nodes vs 26% in the observation arm. Only 1 patient had 4 or more and was in the nivolumab arm.1
In patients with known or available data, most in both arms had an sentinel lymph node tumor burden of 1 mm to 4 mm. Primary lesion sites were similar in the cohorts: trunk, (25% vs 24%), extremities (55% vs 53%), head/neck (20% vs 21%), and external genitalia (1% vs 3%), in the nivolumab and observational groups, respectively. Although ulcers were mostly absent, 22% of patients in the nivolumab arm and 18% patients in the observation arm had ulcers present. The median Breslow thickness of tumors at the time of resection was 1.80 vs 1.90, respectively. ECOG performance status ranged from 0 to 3 with most patients having a score of 0.
The median duration of treatment with nivolumab was 12.0 months (95% CI, 11.8-12.0). Three patients continued treatment and 91% of patients completed the scheduled duration of nivolumab therapy, with the most common reasons for discontinuation being patient choice, toxicity, and disease progression.
Additional data from the study showed that distant metastasis–free survival was examined as well and was 98.2% and 91.3% vs 94.7% and 88.8%, respectively.
The median time to first recurrence was 18.9 (range, 3.9-29.1) months in the nivolumab arm and 9.4 months (range, 2.6-30.6) in the observation arm with the most frequent sites being lungs and lymph nodes. Any first recurrence occurred in 11% vs 26% of patients, respectively. There were 11 deaths (6%) in the nivolumab group and 6 deaths (16%) in the observation group and the most common cause was disease progression. Cardiovascular disease, infectious disease, and unspecified were also cited as fatal events.1
Regarding safety, endocrine, dermatologic, gastrointestinal, and hepatic, treatment-related adverse events (TRAEs) of any grade occurred in the adjuvant nivolumab group at rates of 13%, 11%, 8%, and 5%, respectively. Grade 3 or 4 events were observed in the endocrine and gastrointestinal categories in 1% and 2% of patients, respectively, and 2% of patients discontinued treatment due to TRAEs. The data was collected from patient charts and may be underreported as the data is real-world, investigators noted.
Although the study used a nationwide databased to assess real-world data from this subgroup of patients that may be able to give more accurate date than randomized control trial population data, the analysis was retrospective. Further limitations included the shorter follow-up of the survival outcomes data, as stage IIIA melanoma patients have a relatively good prognosis, errors that could have occurred in data entry, and differences in the cohort’s baseline characteristics.
A lack of adjustment for heterogeneity between the 2 cohorts when reporting Kaplan–Meier curves was as a challenge, study authors wrote.
When asked if removing the IIIA less than 1 mm patients would result in the curves looking more like one would expect, Pavlick explained, “The number of measurements that were done there are very few patients that had a numerical measurement in their sentinel lymph node [and because] it was mostly reported as positive or negative that would be difficult to do.”