In our study, a trend was observed toward an increased risk of isolated brain relapse in BRAF V600 mutated patients in univariate analysis, none of the tested factor was identified to be an independent associated factor or isolated brain relapse in multivariable analysis.
We report a median OS time from diagnosis of brain metastatic disease of 14.1 months, comparable with Gaudy-Marqueste [14] who reported an OS of 11 months. The population in their study was comparable to our own, since it consisted of patients treated with stereotactic brain radiotherapy, combined with systemic treatment by immunotherapy or targeted therapy. Tetu et al [15] found a median overall survival of 16.8 months, and F.Martins et al. [16] 12 months. Lannier et al. found a median overall survival of 15.9% in patients treated with immunotherapy and stereotactic brain radiotherapy [17]
After an initial metastatic event in the brain treated with stereotactic radiotherapy, unsurprisingly, overall survival was better for non-metastatic patients, both cerebral and extra-cerebral.
Nevertheless, OS seems to be better for cerebral vs. cerebral and extra-cerebral progression, itself having a better OS than isolated extra-cerebral progression as shown on Fig3. This data is surprising, and suggests that after a first episode of brain metastasis, the prognosis would be more influenced by progression at non-cerebral sites. It is no longer linked to brain metastases but to extra cerebral disease control. This observation is interesting because comparable to the work of Milsh et al [18]. It is therefore necessary to find an effective systemic therapy to control the extra-cerebral stage, in order to highlight a clinical benefit to local brain control due to brain SRT. This could lead to a reconsideration of brain involvement as a specific disease, different from that of other noble organs, in terms of management, reduced response to immunotherapy and involvement in OS.
Patients with brain metastasis have an increasingly long life expectancy when using immunotherapy and few studies confirm this data. In fact, NIBIT-M2, a phase III study recruited patients with BRAF wild-type or mutant melanoma, and active, untreated, asymptomatic brain metastases randomized (1:1:1) to fotemustine (F), ipilimumab plus fotemustine (I-F), or ipilimumab plus nivolumab (I-N). Median OS was respectively 8.5 months [4.8–12.2] 8.2 months [2.2–14.3] and 29.2 months [0–65.1] in the F arm, I-F arm and I-N arm suggesting superiority of immunotherapy combination.
Four-year survival rate was significantly higher for I+N (41%) vs. F (10.9%) vs I+F (10.3 [19]. With a hindsight of several years, this study also suggests that the greatest impact on OS is borne by extra-cerebral rather than brain metastatic progression after IT-treated cerebral metastatic progression, as we did in our study.
To date, studies point to a better long-term efficacy of immunotherapy treatment in responder patients, and to investigate the best sequential treatment in BRAF V600 mutated melanoma, SECOMBIT study was carried out in patients with BRAF V600 metastatic melanoma with inactive brain metastases.JAK mutation was associated to a better overall survival [20]. Another area under study is the combination of targeted therapy and immunotherapy as TRICOTEL, a phase II trial which combines BRAFi + MEKi + anti PD-L1 (atezolizumab). Preliminary results show an intracranial progression-free survival of 7.2 months for symptomatic patients. Objective intracranial response was 42% for BRAF V600 mutated patients, and 27% for BRAF V600 wild-type patients, but 50 to 60% of patients in both cohorts experienced grade 3 side effects [21] . At ESMO 2023, Elizabeth Burton presented a phase II study challenging the safety and efficacy of atezolizumab, bevacizumab, and cobimetinib in 20 patients with treatment-refractory melanoma brain metastases (Abstract 1085O). This toxic combination in refractory patients highlights the need for therapies adapted to the complexity of the brain tumor environment.
Various trials are still in progress: SWOG S2000 [22] comparing the efficacy of encorafenib (BRAFi) + binimetinib (MEKi) + nivolumab (anti PD-1) with ipilimumab (anti CTLA-4) + nivolumab in patients with BRAF V600 mutated, brain metastatic melanoma that may be symptomatic.These different studies show that the survival of patients with brain metastasis is still challenging, with long term responders, and that it is essential to compare, on the one hand, drugs associations (targeted therapy, immunotherapy), and on the other hand, their sequence of use and the combination with local treatment to improve OS.
When metastatic melanoma is controlled extra cerebrally, we assume that secondary resistance mechanisms specific to the brain stage are responsible, as discussed above: cerebrospinal fluid proteins and efflux proteins (extrinsic resistance), appearance of secondary mutations (intrinsic resistance). It is needed to demonstrate the appearance of secondary mutations during the course of the disease. Ideally, biopsies should be taken from the organ undergoing progression, in this case the brain, and sequencing analyses performed using Next Generation Sequencing (NGS) [23].
Our study has a number of limitations, such as its retrospective monocentric nature and small sample size, which is a real limitation of our study. The symptomatic or asymptomatic nature of brain metastases, with or without corticosteroid use, was not collected, due to a large number of missing data on this subject. We also have to be cautious with regard to local brain progression since local escapes were judged on imaging, without biopsy, which may be confused with radionecrosis.
To the best of our knowledge, this is the first study to look for predictive factors for isolated cerebral progression after a first brain metastasis, which is a daily concern for oncologists and onco-dermatologists and impacts on the patient's care plan and life expectancy. It has recently been agreed that an asymptomatic cerebral metastatic patient has the same prognosis as a non-metastatic cerebral patient. Our study allows us to go further in the prognostic categorization of cerebral metastatic patients, since we argue that after a first metastatic episode in the cerebral stage, the prognosis and OS goes to the extra-cerebral evolution. Future prospective randomized studies should confirm our assertion.