The spectrum of cutaneous adverse events during Encorafenib and Binimetinib treatment in BRAF mutant advanced melanoma
N.P. Grafa, P.Koelblingera,b, N. Gallikera, S. Conrada, M. Baryscha, J. Manganaa, R. Dummera, P.F. Chenga, and S.M. Goldingera
Abstract:
Background: BRAF inhibitor encorafenib alone and in combination with MEK inhibitor binimetinib improves survival in BRAF mutated melanoma patients. So far, the range of cutaneous adverse events has been characterized only for established BRAF inhibitors (vemurafenib, dabrafenib) and MEK inhibitors (trametinib, cobimetinib).
Objective: The aim of this study was to investigate cutaneous adverse events emerging in melanoma patients treated with encorafenib and binimetinib.
Methods: Patients treated with BRAF and MEK inhibitors in clinical trials at the University Hospital of Zurich were identified. Frequency and features of cutaneous adverse events as well as their management were assessed based on the prospectively collected clinical and histopathological data. The events emerging during encorafenib and/or binimetinib therapy were compared to other BRAF and MEK inhibitors at the institution and in the literature.
Results: The most frequent cutaneous adverse events observed in patients treated with encorafenib alone (n=24) were palmoplantar hyperkeratosis (54%), palmoplantar erythrodysesthesia (58%) and alopecia (46%). Drug-induced papulopustular eruptions prevailed in patients with binimetinib monotherapy (n=25). The most frequent cutaneous adverse events in patients treated with encorafenib/binimetinib (n=49) was palmoplantar hyperkeratosis (10%).
Conclusion: Compared to data published for established BRAFi, encorafenib monotherapy showed less hyperproliferative cutaneous adverse events. In contrast, palmoplantar hyperkeratosis and palmoplantar erythrodysesthesia seem to occur more often. The combination encorafenib with binimetinib is well tolerated and induces few cutaneous adverse events.
Introduction:
A mutation in the serine–threonine protein kinase B-rapidly accelerated fibrosarcoma (BRAF) gene can be detected in approximately 50% of melanoma patients.(1,2) Per current treatment guidelines, patients with unresectable or metastatic BRAF V600 mutated melanoma are treated either with a combination of a BRAF inhibitor (BRAFi) and a MEK inhibitor (MEKi) or with anti-PD1- antibody based immunotherapy.(3)
Both regulatory approved BRAFi vemurafenib and dabrafenib have improved progression-free survival (PFS) and overall survival (OS) compared to chemotherapy in BRAF mutated advanced melanoma.(4-6) Subsequently, dual inhibition of the mitogen-activated protein kinase (MAPK) pathway with BRAFi- and MEKi (cobimetinib or trametinib) was investigated to overcome development of treatment resistance. Combination of BRAFi and MEKi led to further improvement of both response rates and duration of response and has therefore evolved as standard of care in patients with BRAF mutated melanoma.(3, 7-9) Recently, the combination of encorafenib, a second generation BRAFi, with MEKi binimetinib, demonstrated improved progression-free survival (PFS) and overall survival (OS) in BRAF mutant melanoma in comparison to vemurafenib monotherapy.(10-12) This combination was FDA approved in June 2018.
Both BRAFi and MEKi can cause cutaneous adverse events (cuAEs). CuAEs occur more frequently with monotherapy than with combined BRAFi/MEKi treatment. Typically, they include both inflammatory (maculopapular exanthema, papulopustular exanthema, palmoplantar erythrodysesthesia) and hyperproliferative (keratosis pilaris, acanthopapilloma, squamous cell carcinoma etc) features.(13-16) The incidence of cutaneous squamous cell carcinomas (cuSCCs) and keratoacanthomas is particularly high with vemurafenib monotherapy.(17) The frequency and pathogenesis of cuAEs have been thoroughly investigated for the first generation BRAFi and MEKi.(8, 14, 15, 18-24) In contrast, except for reports as part of clinical trial publications, the cuAE profile of encorafenib and/or binimetinib has not been characterized in detail yet. The aim of the present study was to characterize cuAEs induced by encorafenib and/or binimetinib in advanced melanoma and to put them into context with the established BRAF and MEKi.
Patients and Methods:
Patients with BRAF V600 or NRAS mutated melanoma treated with BRAFi and/or MEKi (vemurafenib, dabrafenib, encorafenib, , trametinib, binimetinib) within approved clinical trials (www.clinicaltrials.gov: NCT01543698, NCT02159066, NCT01597908, NCT01909453, NCT01763164, NCT01436656, NCT01320085, NCT01820364, NCT01245062, NCT01682083) were included in this analysis.(9, 12, 25-32) Written informed consent to use health related data and biomaterial was obtained from all participants. Patients who withdrew consent and those whom unblinding was not possible were excluded. All patients’ data including cuAEs were prospectively and systematically assessed by experienced dermatologists during routine visits performed in the respective clinical trials. Clinical photographs and histological assessments of cuAEs were conducted based on the study physician’s discretion. Cutaneous adverse events, treatment related data and cuAE management were retrospectively collected from patient’s charts and characterized for this analysis. CuAEs emerging during encorafenib and binimetinib were compared to those arising from established BRAF and MEKi. This correlation was attempted by direct comparison with the available patients at the centre and by available data from the literature. Analyses on cutaneous adverse events were performed using descriptive statistics.
Results:
Between January 2011 and February 2017, 111 patients were treated in at least one clinical trial at the University Hospital of Zurich (USZ). In total, 123 treatments were administered. The number of treatments is higher than the patient number as some patients were subsequently included in more than one trial. Table 1 summarizes the baseline characteristics of the study population. The treatment groups differed with regard to gender and age distribution, treatment duration, as well as previous treatments. Information on the divergent treatment duration is presented in Figure 1. Cutaneous adverse events were reported in 61% of all study treatments. In total, 259 cutaneous adverse events were documented. Of these, 211 were interpreted as related or possibly related to the study drug and are summarized in
The most frequently observed cuAEs in patients treated with encorafenib (n=24) palmoplantar erythrodysesthesia (PPD), palmoplantar hyperkeratosis (PPH), alopecia, macular and maculopapular exanthema, keratosis pilaris and xerosis cutis (Figs 2 and 3b). Differences in the cuAE profile were observed with encorafenib monotherapy compared to vemurafenib monotherapy in regard of keratinocytic proliferations, inflammatory disorders and hair changes. Related and possibly related cuAEs are summarized in Table 3.
Only a few patients treated with encorafenib presented with keratinocytic disorders, including acanthopapilloma (8%) and cutaneous squamous cell carcinoma (cuSCC, 4%). This is less frequent compared the small patient population treated with vemurafenib (n=6) at the USZ, where acanthopapilloma (83%) and cuSCC (33%) were reported (p=0.005). Palmoplantar erythrodysesthesia (PPD) and palmoplantar hyperkeratosis (PPH) were the most frequently noted inflammatory disorders in patients treated with single-agent encorafenib (n=24; 58 % and 54 % respectively). In patients treated with vemurafenib PPD and PPH was observed in half of the patients (50% each). Encorafenib-induced PPD and PPH were more severe (57.2% and 46.2% grade 2 or higher respectively) compared to vemurafenib (only grade 1 reported). Patients with PPD grade 2 or higher (n=8) were all treated with systemic pregabaline. In addition, half of the patients received topical corticosteroids. In one of the patients, treatment with encorafenib had to be interrupted and the dose had to be reduced. Most of the patients with PPH grade 2 (46%) were treated with either oral retinoids (83%) and/or with a retinoid-, urea- and pantothenol-based topical therapy (67%). Other inflammatory disorders included macular and maculopapular exanthema, which in this cohort occurred more frequently in patients treated with vemurafenib (67%), than in those treated with encorafenib (33%). Hair disorders included alopecia and wavy hair (cymotrichy). While alopecia was often reported in patients treated with single-agent encorafenib (46%) and vemurafenib (33%), chymotrichous hair was only noted in one patient treated with encorafenib.
Binimetinib Monotherapy
A total of 25 patients was treated with binimetinib monotherapy. Drug-induced papulopustular eruptions prevailed in both binimetinib and trametinib (n=8) groups (Fig. 3a) Fifty-nine percent of the binimetinib-induced papulopustular eruptions were grade 2 or higher. The majority of patients (>90%) with grade 2 papulopustular eruptions during MEKi treatment were treated with 40 mg doxycycline QD. In addition, most of those patients received local corticosteroids. Treatment interruption was necessary only in one patient treated with binimetinib. Other common cuAEs included alopecia, eczematous eruptions and macular/maculopapular exanthemas (Figure 3).
Encorafenib and Binimetinib Combination Therapy
BRAFi and MEKi combination led to fewer cuAEs compared to the monotherapy groups: 16 of 49 patients (33%) receiving encorafenib/binimetinib and five of 11 patients (45%) receiving dabrafenib/trametinib reported at least one cuAE. All reported cuAEs were grade 1 or 2. Grade 2 cuAEs were more frequent in patients treated with encorafenib plus binimetinib compared to dabrafenib plus trametinib (46% vs. 18%).
Ten percent of the patients treated with encorafenib/binimetinib developed PPH. PPH was not reported in patients treated with dabrafenib/trametinib and PPD was not observed in both combination therapies. Median time to onset of PPH in patients treated with encorafenib/binimetinib was 24 weeks (mean time to onset was 20.2 weeks). CuAE observed in patients with dabrafenib/trametinib combination therapy were maculopapular exanthemas (18%) and erythema-anulare-like eruptions (Fig. 3c; 18%). In general, cuAEs occurred later during the treatment course with combination therapy, than with monotherapy (Fig. 4). However, most cutaneous adverse events occurred within the first 26 weeks of treatment. Management was similar as for monotherapy.
Discussion:
Recently, the combination of encorafenib (BRAFi) and binimetinib (MEKi) has shown significant improvement of PFS and OS in BRAF mutant melanoma patients compared to encorafenib and vemurafenib monotherapy (PFS 14.9 vs. 9.6 vs. 7.3 months, respectively; and OS 33.6 vs. 23.5 vs. 16.9 months respectively). (10, 12, 33) Moreover, encorafenib monotherapy demonstrated superiority over vemurafenib. (10, 12, 33) CuAEs occurring with encorafenib and binimetinib have not been analysed in detail so far. Hence, the present study focused on the cuAE profile of encorafenib and binimetinib and attempted to put it into context with the established targeted therapies in advanced melanoma.
BRAF and MEKi frequently cause cuAEs, which can negatively influence the quality of life of affected patients. The character of these cuAEs has been well described for the first generation BRAFi, vemurafenib and dabrafenib and their corresponding MEKi, cobimetinib and trametinib. (13-15, 20, 34) One cuAE of special interest with targeted therapy is the development of cutaneous squamous cell carcinoma (cuSCC). The rate of cuSCCs vary during treatment with different BRAFi. SCCs occur most frequently with vemurafenib monotherapy. Their development is thought to be induced by a paradoxical activation of the MAPK- pathway in BRAF wild-type cells.(35, 36)
In an in vitro analysis with immortalized human keratinocytes, Adelmann et al. (17) have calculated a paradox index to determine the therapeutic window between clinical efficacy and the grade of paradoxical ERK activation of vemurafenib, dabrafenib, encorafenib and PLX8394. The higher the index the higher the therapeutic window of the tested BRAFi. Encorafenib led less frequently to paradoxic
activation of the MAPK pathway compared to vemurafenib and dabrafenib. Overall, vemurafenib had the strongest paradoxical activation and the smallest therapeutic window. Furthermore, next-generation BRAFi such as PLX8394, exhibit some small structural molecular changes that reduce the paradoxical pathway activation. This might reduce the formation of (cutaneous) adverse events and hence improve drug tolerability. (37, 38)
In the present study, cutaneous adverse reactions that occurred in patients with encorafenib monotherapy included, palmoplantar erythrodysesthesia, palmoplantar hyperkeratosis, alopecia, macular and maculopapular exanthemas, xerosis cutis and keratosis pilaris. In patients with binimetinib therapy, drug-induced papulopustular eruptions predominated. Finally, patients treated with combined encorafenib and binimetinib mainly experienced palmoplantar hyperkeratosis (5 of 49 patients). PPD and PPH were the most frequently reported cuAEs in our encorafenib treated patients. This slightly differs to another study, where alopecia was most frequent (56%) followed by PPD (51%) and hyperkeratosis (38%). Comparing encorafenib with our small vemurafenib cohort, we observed similar frequencies for PPD and PPH. So far with vemurafenib, PPD was reported only in 14%.(12) Despite the incidence for PPD and PPH was similar, the severity in our cohort seemed higher in the encorafenib group. The rate of PPD reported in our encorafenib population (58.3%) is congruent with the rates reported in the phase I (54%) and the phase III (52%).(12, 25)
In the pivotal Columbus trial, encorafenib induced PPD grade 1 and 2 in 38% and grade 3 in 14% respectively, whereas vemurafenib caused PPD grade 1 and 2 in 13% and grade 3 in 1% only.(12) PPD is one of the dose limiting effects of sunitinib and sorafenib, two multi-kinase inhibitors targeting the vascular endothelial growth factor (VEGFR) pathway.(39) Since VEGFR inhibits both tumour angiogenesis and the RAF pathway, one can assume that a vascular mechanism might contribute to the development of PPD and PPH in patients treated with encorafenib. However, the exact pathomechanism remains unclear. In our cohort, keratinocytic disorders, including acanthopapilloma and cuSCC seemed less frequent with encorafenib than with the established BRAFi. (13-16, 19, 21-23) These findings may be attributed to the pharmacological properties of encorafenib. In fact, this second generation BRAFi has a dissociation time which is more than 10 times longer than those of older BRAFi.(11, 25) In addition for patients with combination therapy, the balance between inhibition-potency of both BRAFi and MEKi may lead to the diversity of the observed cuAE profile.
Possible differences in the cuAE profile detected between the different single substances and combination therapies should be explored and discussed regarding their pathogenetic and pharmacologic background. Still, for the GDC-0973 majority of the described cuAEs, the underlying pathogenetic mechanism remains unclear, which mandates further pharmacological and molecular research in bigger cohorts in this field.