Outcomes of patients with melanoma who discontinue immunotherapy

9548

Background: The question of when to discontinue (d/c) anti-program death-1 (PD-1) monotherapy (mono) or nivolumab in combination with ipilimumab (combo) immunotherapy (IT) is unknown. Methods: After IRB approval, a single center (Memorial Sloan Kettering Cancer Center), retrospective study was performed of 162 pts with unresectable stage III or IV melanoma treated with either mono (n = 106) or combo (n = 56) IT. Objective response rate (ORR), progression free survival (PFS), and overall survival (OS) were calculated for all pts from the 1stdose of IT. For pts (n = 40; mono and n = 40; combo) who d/c IT due to reasons (Table) other than progression or death, starting from the last date of IT, we then reported PFS, time to treatment failure (TTF) defined as any subsequent surgery/radiation/systemic therapy, and OS. Results: For pts that were alive at time of analysis, the median follow up was 28 mos. For all 162 pts (demographics in Table), ORR was 38.7% (mono) and 60.7% (combo); median PFS and OS were 12 months (mos) and 25 mos for mono; 34 mos and not reached (NR) for combo, respectively. From the last dose of IT, the PFS, TTF, and OS for 40 mono pts and 40 combo pts who d/c IT for reasons other than progression/death are shown in Table. Reasons included CR, toxicity, or other (most commonly protocol completion or prolonged PR). Conclusions: Outcomes in this cohort of pts with long follow-up treated with mono or combo IT are similar to results from other clinical trials. Pts who d/c IT for reasons other than progression/death were a highly selected group. Nonetheless, favorable PFS, TTF, and OS were seen after IT d/c, even in pts who did not obtain a CR. [Table: see text]

Prognosis of Mucosal, Uveal, Acral, Nonacral Cutaneous, and Unknown Primary Melanoma From the Time of First Metastasis

BACKGROUND

Subtypes of melanoma, such as mucosal, uveal, and acral, are believed to result in worse prognoses than nonacral cutaneous melanoma. After a diagnosis of distant metastatic disease, however, the overall survival of patients with mucosal, uveal, acral, nonacral cutaneous, and unknown primary melanoma has not been directly compared.

MATERIALS AND METHODS

We conducted a single-center, retrospective analysis of 3,454 patients with melanoma diagnosed with distant metastases from 2000 to 2013, identified from a prospectively maintained database. We examined melanoma subtype, date of diagnosis of distant metastases, age at diagnosis of metastasis, gender, and site of melanoma metastases.

RESULTS

Of the 3,454 patients (237 with mucosal, 286 with uveal, 2,292 with nonacral cutaneous, 105 with acral cutaneous, and 534 with unknown primary melanoma), 2,594 died. The median follow-up was 46.1 months. The median overall survival for those with mucosal, uveal, acral, nonacral cutaneous, and unknown primary melanoma was 9.1, 13.4, 11.4, 11.7, and 10.4 months, respectively. Patients with uveal melanoma, cutaneous melanoma (acral and nonacral), and unknown primary melanoma had similar survival, but patients with mucosal melanoma had worse survival. Patients diagnosed with metastatic melanoma in 2006-2010 and 2011-2013 had better overall survival than patients diagnosed in 2000-2005. In a multivariate model, patients with mucosal melanoma had inferior overall survival compared with patients with the other four subtypes.

CONCLUSION

Additional research and advocacy are needed for patients with mucosal melanoma because of their shorter overall survival in the metastatic setting. Despite distinct tumor biology, the survival was similar for those with metastatic uveal melanoma, acral, nonacral cutaneous, and unknown primary melanoma.

IMPLICATIONS FOR PRACTICE

Uveal, acral, and mucosal melanoma are assumed to result in a worse prognosis than nonacral cutaneous melanoma or unknown primary melanoma. No studies, however, have been conducted assessing the overall survival of patients with these melanoma subtypes starting at the time of distant metastatic disease. The present study found that patients with uveal, acral, nonacral cutaneous, and unknown primary melanoma have similar overall survival after distant metastases have been diagnosed. These findings provide information for oncologists to reconsider previously held assumptions and appropriately counsel patients. Patients with mucosal melanoma have worse overall survival and are thus a group in need of specific research and advocacy.

Evaluation of the absolute lymphocyte count as a biomarker for melanoma patients treated with the commercially available dose of ipilimumab (3mg/kg)

8575

Background: Ipilimumab (ipi) has demonstrated an overall survival (OS) benefit in 2 phase III trials. Only ~30% of patients (pts) achieve clinical benefit, and factors that determine which pts benefit are unclear. For pts treated with 10mg/kg of ipi, we previously reported that an absolute lymphocyte count (ALC) ≥1000/μL prior to dose 3 [week (wk) 7] was associated with improved OS. Since the mean increase in ALC during ipi treatment correlates with dose, we investigated if ALC is also associated with improved OS at 3mg/kg, the currently FDA approved, commercially available dose. Methods: In an IRB-approved analysis, we evaluated landmark survival data from 137 pts treated with 3mg/kg of ipi at Memorial Sloan-Kettering Cancer Center. 67 pts were treated on an expanded access protocol (CA 184-045). 70 pts were treated per FDA approval (commercial ipi) with 4 standard induction doses. These 2 groups were analyzed separately because some pts in CA 184-045 received re-induction ipi. ALC was determined at first ipi dose (baseline, wk 1) and at subsequent doses (wks 4, 7, and 10). Results: Pts treated with 3mg/kg on CA 184-045 with a wk 7 (prior to dose 3) ALC ≥1000/µL had significantly improved OS compared to pts with an ALC at wk 7 <1000/μL (Median OS: not reached vs. 4.24 mos, p<0.001). This OS difference was also seen for pts treated with commercial ipi (Median OS: not reached vs. 4.44 mos, p<0.01). This difference remained significant in a multivariable model accounting for Karnofsky performance score, LDH, M-stage, and number of prior therapies for pts in the CA 184-045 group and commercial ipi group (p=0.01 and p=0.05, respectively). Baseline ALC ≥1000/µL was associated with improved OS (p=0.02) for pts in the commercial ipi group, though follow-up is limited. Conclusions: At the FDA approved dose of ipi, 3mg/kg, ALC at wk 7 remains significantly associated with improved OS. Our preliminary finding of improved OS for pts treated with commercial ipi whose pre-treatment baseline ALC ≥1000/µL deserves confirmation with longer follow-up and prospective validation. Baseline or on treatment ALC may be a marker of overall prognosis, regardless of therapy.