Treatment with RAF-1 inhibitorPatient characteristics

Between January 2010 and November 2022, 44 patients with BRAF aberrated advanced solid tumors who received the first RAFi (RAF-1i) as monotherapy or in combination and were rechallenged with a second RAFi (RAF-2i) with or without other therapies at the Clinical Center for Targeted Therapy, University of Texas MD Anderson Cancer were identified for study analysis. Patients received therapy either as standard of care or as part of a clinical trial. RAF-1i and RAF-2i included class I, pan-RAF or dimer-selective RAF inhibitors. The Institutional Review Board independently reviewed and approved each clinical trial in which patients presented within this analysis were enrolled. The patients provided written informed consent before treatment with investigational therapy. All procedures conformed with the ethical standards of the institutional research committee and with the Declaration of Helsinki. Patient and disease characteristics before rechallenging are shown in Table S1.

The median age of trial participants was 54.5 years (25 years – 76 years) while in gender distribution, male participation was predominant (n = 26; 59%) compared to female participation (n = 18; 41%). The majority of the patients were of Caucasian ethnicity (n = 36; 82%) followed by Hispanic (n = 6; 14%) and African-American (n = 1; 2%) and Asian ethnicities (n = 1; 2%) Tumor types and histologies included cutaneous melanoma (n = 16; 36%), colorectal carcinoma (CRC) (n = 10; 23%), thyroid cancers with papillary thyroid histology (n = 3; 7%), anaplastic histology (n = 1; 2%), Central Nervous System (CNS) tumors including glioblastoma (n = 2; 5%), pleomorphic xanthoastrocytoma (n = 1; 2%), anaplastic astrocytoma (n = 1; 2%). Other tumor types included cholangiocarcinoma (n = 3; 7%); pancreatic ductal adenocarcinoma (PDAC) (n = 2; 5%), ovarian serous carcinoma (n = 2; 5%), non-small lung cancer of adenocarcinoma histology (n = 1; 2%), triple negative breast cancer (n = 1; 2%) and neuroendocrine carcinoma (n = 1; 2%). All patients had locally advanced, recurrent, or metastatic disease before RAF-1i therapy. 22% of patients had 1–2 lines of prior therapies of which 21 patients (48%) had prior targeted therapy, 17 patients (39%) had prior immunotherapy (IO), 23 patients (52%) had prior chemotherapy and 16 patients (36%) had prior radiation therapy.

For RAF-1i therapy, 21 patients (48%) received dabrafenib, while 10 patients (23%) received vemurafenib and encorafenib, respectively and 3 patients (7%) received investigational therapy as part of clinical trial participation. The majority of patients had an Eastern Cooperative Oncology Group (ECOG) of 1 (n = 30) followed by ECOG of 0 (n = 13) and ECOG of 2 (n = 1). Most patients had no CNS disease at the time of RAF-1i therapy (n = 36; 82%). The burden of metastatic disease sites ranged from 0 to 7 sites with at least 15 patients (34%) with 1 metastatic site at therapy. 24 patients (55%) had RAF-1i as an investigational agent while 20 patients (45%) had RAF-1i as standard of care (SOC) therapy. 37 patients (84%) received RAF-1i as part of combination therapy with immune checkpoint inhibitors (ICI) (n = 9; 20%), therapy targeting MEK, Epidermal Growth Factor Receptor (EGFR), B-cell lymphoma 2 (BCL2), and multi-kinase pathways (n = 34; 77%) or chemotherapy (n = 6; 14%).

Safety and tolerability

Toxicities were evaluated based on the National Cancer Institute (NCI) Common Terminology Criteria for Adverse Events version 4 or 5 (CTCAE) [9]. The median treatment duration was 7 months for RAF-1i while the median number of cycles with RAF-1i was 7 (range: 1–70). Thirty-six patients came off therapy due to PD (82%) while 6 patients (14%) came off therapy due to toxicities as detailed below and 2 patients (5%) completed duration of intended therapy either as standard of care or as part of a clinical trial. Eleven patients (25%) developed grade 3 or 4 toxicities (G3 or G4) secondary to RAF-1i therapy. Of this, 6 patients developed treatment-related adverse events (TRAEs) secondary to investigational therapy such as G3 fatigue (n = 1), G3 neutropenia (n = 2), G3 transaminitis (n = 1) and G3 cutaneous rash (n = 2) where 4 patients received combination regimens with ICI, BCL-2i or MEKi. Five patients developed G3/G4 toxicities to combinatory standard of care therapies targeting BRAF and MEK pathway with G3 cutaneous rash (n = 3), G3 pyrexia (n = 1), and G3 acute kidney injury (n = 1). Hence, six patients discontinued therapy secondary to the above toxicities. Notably, four of these patients had the best responses with 3 partial responses (PR) and 1 complete response (CR) with a median duration of response (DOR) of 27.4 months which included cutaneous melanoma, cholangiocarcinoma, anaplastic thyroid carcinoma (ATC) and papillary thyroid carcinoma (PTC).

Antitumor activity

The best overall response was 36.3% seen in 16 patients (3 CRs and 13 PRs) with median DOR of 11.4 months. One patient with PTC and another patient with pleomorphic xanthoastrocytoma treated with investigational therapy experienced complete response while another patient with PTC treated with SOC therapy experienced complete response as well. Of the 13 PRs, nine partial responses were as part of investigational therapy which included TNBC (n = 1), NSCLC (n = 1), CRC (n = 1), cholangiocarcinoma (n = 2), cutaneous melanoma (n = 2) and ATC (n = 1) where 8 of the 9 patients received combinatory regimen with chemotherapy, ICI, MEKi or BCL-2i. Four patients experienced partial responses as part of SOC therapy including cutaneous melanoma (n = 2) and CRC (n = 1) which included a combination with MEKi, EGFRi, and ICI. 24 patients experienced stable disease (SD) with 11 patients experiencing durable responses greater than 6 months. The disease control rate (DCR) including CR, PR, and SD >/= 6 months was 61.3% seen in 27 patients. Four patients including 2 CRC, 1 PDAC, and 1 glioblastoma experienced progressive disease (PD) via SOC and investigational therapies after a median treatment duration of 3.2 months.

Survival outcomes

Of the 44 patients, six patients were censored due to discontinuation for toxicities. The median progression-free survival (PFS-1) with therapy with RAF-1i either as monotherapy or combination was 8.4 months. The median PFS-1 among responders was 11.4 months. The most common reason for discontinuation of therapy was progressive disease (n = 36) while two patients completed therapy (n = 2).

Treatment between RAF-1 and RAF-2 inhibitor therapies

Twenty-seven patients (61%) underwent intervening therapies before rechallenge with RAF-2i while seventeen patients (39%) did not have intervening therapies after RAF-1i. Investigational regimens included Phosphoinositide 3-kinase (PIK3)i, Janus kinase (JAK)1i, Porcupine homolog (PORCN)i, ICI, oncolytic viral therapy, cytokine therapy, extracellular signal-regulated kinase (ERK)i, immunomodulators targeting TLR7/8, antibody-drug conjugate therapy, and radiation. SOC therapies included chemotherapy, anti-VEGF (Vascular Endothelial Growth Factor) agents, ICI, BRAFi, and MEKi.

The median time to RAF-2i was 3.3 months (0.03-73.7 m) from the end of RAF-1i. In terms of survival outcomes, the median PFS with intervening therapies as stated above was 3.8 months (1-20.6 m).

Treatment with RAF-2 inhibitorPatient characteristics

Predominantly 93% (n = 41) of patients went on to receive RAF-2i as investigational therapy where 17 patients (41%) received in combination with an agent targeting BRAF (n = 2), MEK (n = 3), ERK (n = 1), BCL2 (n = 1), EGFR (n = 4), cytochrome P450 3 A (CYP3A)i (n = 1) and multi-kinases (n = 7). Twenty-four patients received investigational therapy as monotherapy. Three patients received SOC therapies with RAF-2i in combination with MEKi and/or ICIs. Most patients had an ECOG of 1 (n = 40; 90%) while 8 patients had CNS metastases with one patient who developed CNS involvement after RAF-1i therapy. The burden of metastatic disease was higher in this group with sites ranging from 0 to 6 sites with at least 12 patients (28%) with 3 metastatic sites at therapy.

Safety and tolerability

The median treatment duration with RAF-2i was shorter with rechallenge at 2.6 months versus 7 months for RAF-1i while the median number of cycles with RAF-2i was also shorter at 3 (1–33) compared with RAF-1i at 7 (1–70). Nine patients developed G3/G4 TRAEs with rechallenge to RAF-2i, all related to investigational therapies. Of this, 3 patients received monotherapy and developed G3 cutaneous rash (n = 2) and G3thromboembolic episode (n = 1). Six patients received combinatory therapies targeting MEK, ERK, CYP3A, multi-kinase pathway, and EGFR and developed G3 hypertension and creatinine kinase elevation (n = 1), G3 cutaneous rash (n = 2), G3 lipase elevation (n = 1), G3 Alanine Aminotransferase (ALT) elevation (n = 1) and G3 vomiting (n = 1). Six patients underwent dose reduction. Interestingly, one patient with ovarian serous carcinoma currently off trial and another patient with papillary thyroid carcinoma who is still experiencing partial response had a median DOR of 15.4 months.

Antitumor activity

The best overall response was 18.1% in eight patients to rechallenge with RAF-2i with a partial response of 18% (7 cPR + 1 uPR). Sixteen patients experienced SD (36%) while 20 patients developed PD (45%) while the clinical benefit rate (PR + SD) was greater than 50% in the participants (54.5%). Of 24 patients with PR and SD, 8 (33%) patients had durable responses (3 PRs and 5 SD) lasting greater than 6 months (Table S2). PR’s were seen in thyroid cancer (1 anaplastic; 2 papillary), 1 ovarian serous histology, 2 cutaneous melanoma, 1 cholangiocarcinoma, and 1 anaplastic astrocytoma. However, the median DOR with RAF-2i was 2.5 months compared to 11.4 months with RAF-1i. Six patients received investigational therapy while 2 patients received SOC. Of the 36 patients who discontinued RAF-1i due to PD, 17% responded to rechallenge with 6 partial responses to RAF-2i. Among 16 patients who had CR or PR on RAF1i, 5 patients experienced partial responses again with RAF-2i while 3 patients who experienced SD with RAF-1i, had conversion to PR with RAF-2i. In the RAF-2i group, of all patients achieving PR (n = 8), 5 patients had other intervening therapies before RAF-2i (ICI; kinase inhibitors; chemotherapy; investigational therapies) while 3 had no interim therapies. 5 responders received a combination with MEKi, while 1 responder had MEKi and ICI and 2 responders had monotherapy with RAF-2i. Notably, all responders had a median time off therapy from RAF-1i at 5.6 months. The bar plot of responses between RAF-1i and RAF-2i therapies is depicted in Fig. 1 and stratified by histology and survival impact in Fig. 2.

Fig. 1

Anti-Tumor Activity of RAF2 inhibitor with rechallenge after RAF1 inhibitor. Illustration depicting the Anti-Tumor Activity of RAF2 inhibitor upon rechallenge subsequent to RAF1 inhibitor treatment. The figure visually captures the treatment response dynamics, showcasing the impact of RAF2 inhibitor rechallenge on tumor progression following initial RAF1 inhibitor therapy. CR = Complete Response; PR = Partial Response; ORR = Overall Response Rate; SD = Stable Disease; PD = Progressive Disease; DCR = Disease Control Rate; CBR = Clinical Benefit Rate; RAF1i = First RAF inhibitor; RAF2i = Second RAF inhibitor

Fig. 2

Overall Survival and Progression Free Survival Stratified by Histology in the Re-RAFFLE Study. The figure provides a comprehensive overview of treatment responses categorized by histological subtypes, highlighting the survival outcomes observed in the context of the Re-RAFFLE study. NSCLC = Non-Small Cell Lung Cancer; CRC = Colorectal Cancer; CUP = Carcinoma of Unknown Primary; OS1 = Overall Survival with first RAF inhibitor; OS2 = Overall Survival with second RAF inhibitor; PFS1 = Progression Free-Survival with first RAF inhibitor; PFS2 = Progression Free-Survival with second RAF inhibitor

Survival outcomes

Of the 44 patients, 8 patients were censored due to discontinuation for toxicities (n = 2), patient preference (n = 2), and on trial (n = 4). The median progression-free survival (PFS-2) with therapy with RAF-2i either as monotherapy or combination was shorter at 2.3 months (1.83-5.6 m) compared to 8.6 months (6.5-11.5 m) with RAF-1i. Of the eight rechallenge responders, four patients were censored as they remain on trial. However, the median PFS with RAF-2i responders (PFS-2) was improved at 12.8 months compared to 11.4 months with RAF-1i responders. At a median follow-up of 20 m, the median OS from retreatment with RAF-2i was 15.4 months (11.1-30.8 m). The median overall survival was 83.3 months for all patients who underwent rechallenge (39.1 m-N/A). RAF-2i therapy demonstrated a significantly inferior impact on PFS compared to RAF-1i therapy (Hazard Ratio [HR]: 2.2989, 95% confidence interval [CI]: 1.42–3.68, p = 0.000631). Concordantly, RAF-2i demonstrated a significant impact on OS with an HR of 2.62 (95% CI: 1.48–4.63, p = 0.000961). These survival graphs are depicted in Figure S1. However, one must interpret these findings with caution since this is a retrospective observational study with a small sample size.

Prognostic factors

Independent prognostic factors for significant inferior survival outcomes at rechallenge included male gender treated with RAF-2i, metastatic burden at the time of RAF-2i, ECOG 0 and 1 with RAF-2i while rechallenge with combination RAF-2i had a worse survival impact on PFS compared to RAF-2i monotherapy (PFS: 11.6 m vs. 2.7 m; HR: 3.6; p = 0.037). Further characteristics are provided in the univariate analysis provided in Table S3. Specific factors that impact survival outcomes are presented in Figure S2.

Receiving intervening therapies or not did not have a significant impact on survival for RAF2i rechallenge (HR:0.6143, 95% CI: 0.2932–1.287, p = 0.2 [PFS]; HR:0.9674, 95% CI: 0.4215-2.21, p = 0.94 [OS]) as in Figure S3. ORR did not significantly differ between the combination and monotherapy groups in RAF-2i (p = 0.056). However, the median PFS in the patients who received intervening therapies was 3.8 months (2.37-11 m).

In terms of combination therapies, RAF-2i regimens had a significantly worse impact on overall survival compared to combination RAF-1i regimens (HR = 3.5677, p = 0.0159). Similarly, The RAF-2i combination showed a significant association with a worse impact on PFS (HR = 6.2360, p = 0.00145) as depicted in Figure S4.

When examining the impact of SOC and investigational therapies on RAF-1i and RAF-2i, investigational RAF-2i demonstrated a significantly inferior outcome on PFS with an HR of 2.8753 (p < 0.001) and on OS with HR of 2.9158 (p = 0.0019) while no significant impact was noted with SOC therapies with RAF-1i or RAF-2i (Figure S5).

Genomic landscape

At baseline, 41 patients had BRAF V600E aberration, while aberrations in the RAF/RAS pathway were seen as BRAF V600K (n = 1); BRAF K601Q (n = 1) and KRAS G12S (n = 1). 13 patients had genomic testing post-RAF-1i or RAF-2i. Nine patients had a liquid biopsy and four patients had tissue based next-generation sequencing (NGS) performed on post-therapy tumor biopsies. Of the 9 patients with post-therapy liquid biopsy, 4 patients had prior tissue based somatic NGS testing. One patient with rectal adenocarcinoma had baseline BRAF V600E and PIK3CA I391M mutation at baseline tumor NGS while post-therapy liquid biopsy revealed persistence of BRAF V600E mutation, loss of PIK3CA aberration, the emergence of APC L629; APC T1556fs; EGFR 1134delins; GNAS R201C and TP53 V203fs aberrations with variant allele frequency (VAF) > 2% in each. Another patient with CRC had baseline tumor NGS with BRAF V600E; RNF43 V36fs and TP53 Q167 aberrations while post-therapy liquid biopsy revealed persistence of BRAF V600E and TP53 Q167 aberrations while emergent aberrations with CDK6 gain; EGFR gain and MYC gain was noted. A cholangiocarcinoma patient with baseline BRAF V600E mutation and BRAF amplification on tumor NGS revealed the persistence of BRAF V600E aberration and development of ESR1 L536F aberration on post-therapy liquid biopsy. The last patient with ovarian serous carcinoma had a baseline BRAF V600E mutation via tumor NGS with the persistence of BRAF V600E aberration and no newly acquired aberrations on post-therapy liquid biopsy.