9563 Background: DMFS is an important endpoint for patients with stage III cutaneous melanoma, as delaying or preventing systemic disease is associated with improved clinical and patient-reported outcomes. Prior results from the phase 3 COMBI-AD trial (NCT01682083) showed 5-year DMFS rates of 65% with adjuvant D + T vs 54% with placebo (PBO; hazard ratio [HR] = 0.55; 95% CI: 0.44-0.70). An analysis of DMFS by AJCC-7 stages IIIA-C suggested a similar benefit of D + T vs PBO regardless of stage (Dummer R et al. N Engl J Med. 2020). Here, we report 5-year DMFS rates by AJCC-8 stages IIIA-D, other prognostic subgroups, and results of a regression tree analysis with DMFS. Methods: Patients with resected AJCC-7 stage III BRAFV600E/K-mutant melanoma were randomized to either D (150 mg twice daily) + T (2 mg once daily) or 2 matched PBOs for 12 months. Primary endpoint was relapse-free survival (RFS); DMFS was a secondary endpoint. Kaplan-Meier survival analyses were performed to assess the long-term benefits for DMFS rates with D + T vs PBO. The regression tree analysis (data cutoff: 5 years) for all patients (N = 870) evaluated potential prognostic/predictive factors of long-term DMFS including baseline age, sex, region, BRAF mutation type, body mass index, lactate dehydrogenase levels, ECOG, T and N categories, histology, primary tumor ulceration, treatment type, number of lymph nodes with metastases, tumor mutational burden, and interferon-gamma gene expression signature (IFN-γ GES). Results: At 5 years, DMFS rates were higher for patients with AJCC-8 stages IIIB-D disease receiving adjuvant D + T vs PBO (table). Five-year DMFS rates also favored D + T vs PBO in subgroups of patients with microscopic or macroscopic lymph node involvement (table) and those with or without primary tumor ulceration and/or in-transit metastases. A regression tree revealed T and N stage, treatment type, and IFN-γ GES as important variables defining 5-year DMFS subgroups. Conclusions: In this retrospective analysis, adjuvant D + T provided long-term DMFS benefit vs PBO in stage IIIB-D patients with resected BRAFV600E/K-mutant melanoma. Key clinical and patient factors impacting DMFS were similar to prior RFS findings (ESMO 2021; Robert C et al. Ann Oncol. 2021) and included T and N stage, treatment type, and IFN-γ GES. These results further validate the robust long-term clinical benefit of adjuvant D + T for patients with melanoma. Clinical trial information: NCT01682083. [Table: see text]

No abstract

Background:Patients (pts) diagnosed with HR MDS are predominantly older (median age ≈70 years), with severe comorbidities and poor performance status, and often experience suboptimal outcomes with currently available therapies. In recent years, approved treatment options have remained limited; therefore, more effective therapies with tolerable safety profiles are needed to improve prognosis. This lack of evidence has prompted the evaluation of real-world evidence to understand and establish the level of baseline care and current unmet needs in pts with HR MDS.

9531 Background: Immunotherapy (IO) is commonly used to treat BRAF+ metastatic melanoma (MM) patients in the first line (1L) setting and has demonstrated durable outcomes in clinical trials. However, most patients discontinue IO therapy for toxicity, disease progression, or other reasons. To date, no multi-center or nationwide US-based study has examined treatment patterns and outcomes for patients with BRAF+ MM post discontinuation of 1L IO. The aim of this study is to describe real-world treatment patterns and outcomes among patients who discontinued 1L IO. Methods: This retrospective cohort study used the Novartis BRAF+ meLanoma patients ObsErvational (NOBLE) database, the harmonized customized data from Flatiron and ConcertAI. Patients were ≥18 years old, had a diagnosis of BRAF+ MM, were treated with pembrolizumab, nivolumab, or ipilimumab + nivolumab on or after 9/1/2014, and then discontinued 1L therapy. Reason for discontinuation was extracted from medical records. Descriptive statistics were used to describe baseline characteristics and treatment patterns. Kaplan–Meier curves were used to analyze time to progression or death (TTPD) and time to death (TTD). Results: Of the 898 included patients (mean age: 61 years; male: 65%); 46% initiated ipilimumab + nivolumab, 24% nivolumab, and 30% pembrolizumab. The most common reasons for 1L discontinuation were toxicity (26%) and progression (25%). Medical records noted 5.3% completed therapy on discontinuation and 34.5% provided no reason of discontinuation with median duration of therapy (MDOT) of 379 and 138 days respectively. At 6 months, 34 % (n = 303) remain on IL IO. MDOT for patients who discontinued IO due to toxicity and those who discontinued due to progression were 54 and 63.5 days respectively. Patients who discontinued due to toxicity had a median time of 142 days to next treatment. TTPD was best for patients who discontinued therapy due to completion or toxicity (6-month progression rate: 13% and 20%). Patients who discontinued due to progression did especially poorly (6-month progression rate: 59%). About 33% of patients (n = 296) needed second line (2L) therapy, and the majority (80.7%) received combination BRAF+ targeted therapy. Overall, 38% and 50% of patients died post IO discontinuation for any reason at 1 and 2 years, respectively. 31% of patients had brain metastases and a greater proportion (56%) died within a 2-year period compared with those without brain metastases (56% vs 49%, p = 0.0036). Conclusions: IO, although effective, is not curative for all patients. A significant number of patients discontinue therapy due to progression. Patients with BRAF+ MM who progress early on 1L IO therapy have a high risk of death and should be considered for other therapy options, including targeted therapy.

9532 Background: While immunotherapy (IO) and BRAF-targeted therapy (TT) have benefit in BRAFV600 mutant (BRAF+) metastatic melanoma (MM), there is a paucity of real-world data on the impact of systemic therapy choice on outcomes based on key characteristics such as site and number of baseline metastases. Patients with >1 baseline site and certain sites of metastases are also underrepresented or excluded in clinical trials. The aim of this study was to evaluate the association between these characteristics and survival among BRAF+ MM patients. Methods: This was a retrospective cohort study using the Novartis BRAF+ meLanoma patients ObsErvational (NOBLE) dataset – harmonized customized data from Flatiron and ConcertAI. It included patients ≥18 years, who received treatment with a first-line (1L) IO (anti-PD-1 mono or combination therapy ipilimumab + nivolumab) or TT (any BRAF/MEK-inhibitors) after 1/1/2014. Progression free survival (PFS) and overall survival (OS) for IO and TT were analyzed according to number (1, 2, 3+ sites) and location (brain, lung, liver, bone) of baseline metastasis. Treatment sequence from 1L to 2L (i.e. IO/TT vs TT/IO) were also compared for PFS and OS outcomes. Results: A total of 1,961 patients were included, with 620 patients (32%) on IO monotherapy, 501 patients (26%) on IO combo therapy, and 840 patients (43%) on TT in the 1L. When adjusted for sex, age, ECOG, and Charlson Comorbidity Index, there was no difference in PFS or OS between 1L IO mono, IO combo and TT therapies in patients who had 1, 2, or 3+ baseline metastases. For patients who had either baseline brain, liver, lung, or bone metastasis, there was no difference in PFS and OS between IO mono, IO combo, and TT combo therapies. Of the 521 patients included in the sequencing analysis (only patients who received 2L therapy), 239 patients (46%) had 1L IO/2L TT. There was no difference in PFS or OS between treatment sequences for patients with any number or location of baseline metastasis. Conclusions: In this real-world retrospective cohort study, there is no difference in survival between 1L TT and IO for BRAF+ MM patients. Outcomes are comparable regardless of number and location of metastases, including brain metastasis. Whether switching from 1L TT to IO before progression may account for differences compared to trial data will be explored further.[Table: see text]