Intrahepatic cholangiocarcinoma is a rare malignancy, which is rich in actionable alterations. Genomic aberrations in the mitogen-activated protein kinase (MAPK) pathway are common, and BRAF exon 15 p.V600E mutations are present in 5–7% of biliary tract cancers (BTC). Dual inhibition of BRAF and MEK has been established for BRAF-mutated melanoma and lung cancer, and recent basket trials have shown efficacy of this combination in BRAF V600E-mutant BTCs. Here, we report on a patient with BRAF exon 15 p.V600E mutant metastatic intrahepatic cholangiocarcinoma who was started on BRAF and MEK inhibition with vemurafenib and combimetinib. Shortly thereafter, he developed debilitating myositis, which was refractory to corticosteroids, requiring therapeutic plasma exchange and intravenous immunoglobulin. We also review BRAF as a target in BTCs, relevant clinical trials, and adverse events associated with BRAF and MEK inhibition.
There has been growing interest in precision oncology, with the expectation that molecular profiling can help identify drivers of tumor growth, and thus, actionable alterations that can be targeted directly or indirectly with targeted therapy. It has been the hope that personalized cancer therapy with molecularly matched therapies will not only allow for greater antitumor efficacy but will also be less toxic. However, it should be noted that targeted therapies are associated with multiple adverse events, including those that are off-target, such as dermatologic reactions, hepatotoxicity, rhabdomyolysis, arthralgia, and ophthalmologic reactions as seen with BRAF and MEK inhibitors.[1–6]
Intrahepatic cholangiocarcinoma (ICC) is a rare malignancy that arises from the bile duct epithelial cells within the liver. Unfortunately, many patients present with metastatic disease, and the 5-year survival for patients with unresectable ICC is less than 5–10%.[8,9] Increased molecular profiling of cholangiocarcinoma (CCA) has prompted evaluation of targeted therapies for these malignancies, which are rich in actionable mutations. Large-scale sequencing studies have identified several targetable genetic alterations including FGFR2, ERBB2, IDH1, and BRAF, which have led to the development of various inhibitors currently under evaluation in multiple biomarker-selected clinical trials.[11–13]
BRAF, a serine/threonine protein kinase that activates the mitogen-activated protein kinase (MAPK) pathway, is an oncogenic driver in many human cancers. Activating mutations at codon 600 results in constitutively active BRAF and aberrant MAPK signaling. A recent multicohort basket study of the BRAF inhibitor vemurafenib in nonmelanoma BRAF V600E-mutant solid tumors in 172 patients with 26 unique cancer types, reported an overall response rate of 33% and median duration of response of 13 months. Responses were observed in 13 unique cancer types, highlighting the importance of BRAF V600E as a driver. BRAF mutations are present in 5–7% of biliary tract cancers (BTC), and BRAF V600E-mutant ICC has been associated with worse overall survival.[15,17] Combination therapy with dual BRAF and MEK inhibition (BRAFi + MEKi) has been used to provide greater inhibition of the MAPK pathway and overcome resistance to BRAFi monotherapy.[18,19] Several case reports, and a recent phase II basket trial (NCT02034110), have shown responses with BRAFi + MEKi in patients with BRAF V600-mutant BTC.[19–21]
Here, we present a case of a man with BRAF V600E-mutant ICC who progressed on chemotherapy. He was then enrolled in the American Society of Clinical Oncology (ASCO) Targeted Agent and Profiling Utilization Registry (TAPUR) trial (ClinicalTrials.gov identifier: NCT02693535) and treated with vemurafenib and cobimetinib. Shortly thereafter, he developed debilitating myositis requiring hospitalization.
A 77-year-old man presented with metastatic ICC. Six months prior, he was diagnosed with a 6-cm right liver mass and underwent surgical resection, with pathology revealing CCA with sarcomatoid features (stage T1bN0M0). Relevant past medical history included use of rosuvastatin at 40 mg by mouth (PO) once daily (QD). Genomic testing (Foundation One CDx) revealed a BRAF exon 15 p.V600E mutation along with CDKN2a loss, CDKN2b loss, PIK3CA H1047R, CDKN1A R48*, and PIM1 amplification. Subsequently, he received adjuvant capecitabine, which was discontinued after three cycles due to a drug-related rash. Restaging positron emission tomography–computed tomography revealed a 1.2-cm left upper lobe nodule and a biopsy specimen confirmed metastatic disease. He was transitioned to gemcitabine, cisplatin, and abraxane and received five cycles, with a best response of progressive disease. He was then enrolled in the TAPUR trial (ClinicalTrials.gov identifier: NCT02693535) based on his BRAF exon 15 p.V600E mutation and started on vemurafenib (960 mg PO twice daily [BID]) and cobimetinib (60 mg PO once daily [QD], 21 days on, 7 days off).
He first reported fatigue (grade 2) and anorexia (grade 1) 6 days after beginning treatment, and his cobimetinib dose was held the next week when his symptoms did not improve. On cycle 1 day 18, he presented with overwhelming fatigue (grade 2) and proximal upper or lower extremity weakness (grade 3), and was directed to the emergency department. Notably, no cutaneous lesions or rashes were observed. Admission labs were notable for elevated creatine kinase (CK) (3041 units/L), creatinine (1.66 mg/dL), transaminases (aspartate transaminase 137 units/L, alanine transaminase 68 units/L), alkaline phosphatase (362 units/L), aldolase (18.9 units/L), erythrocyte sedimentation rate (ESR), and C-reactive protein (CRP; 42, 53.6 mg/L), as well as several electrolyte derangements (potassium 2.7 mEq/L, sodium 135 mEq/L, calcium 8 mEq/L, phosphorous 2 mEq/L) and urine myoglobin more than 5000. Antinuclear antibody (ANA) titer was 1:160 with a speckled pattern. Autoantibody myositis panel and rheumatoid factor were within normal limits. Magnetic resonance imaging of the brain and spine were unremarkable. Cobimetinib, vemurafenib, and rosuvastatin therapies were discontinued and methylprednisolone 80 mg PO BID was started given concern for a drug-related inflammatory or immune adverse event.
While his CK and Cr initially improved with intravenous (IV) fluids, he developed worsening muscle weakness with rising CK levels, which prompted initiation of high-dose pulse corticosteroids with methylprednisolone 500 mg intravenously three times daily (TID). He then underwent 5 days of therapeutic plasma exchange (PLEX) given concern for immune-mediated myositis (Figure). A biopsy procedure was performed of his left quadriceps, revealing severely distorted skeletal muscle with rare degenerative and necrotic fibers, macrophages, rare B cells, and scattered T cells. His CK continued to rise, and he was started on a 4-day course of IV immunoglobulin (IVIG), which led to improvement of his CK levels and symptoms. Repeat ANA was 1:40 without ANA detected by immunofluorescence, CRP normalized, and ESR decreased to 25. Corticosteroids were tapered, and he was discharged home with improvement of his myositis (grade 2). Over the next several weeks, he reported worsening weakness requiring a wheelchair although his CK remained within normal limits. He unfortunately was unable to recover enough to receive additional anticancer therapy and died 1 month after discharge, 62 days after the last vemurafenib and cobimetinib treatment.
There is increasing awareness of BRAF V600E as a driver in several human cancers. A vemurafenib basket trial in BRAF V600E-mutant tumors reported an overall response rate of 33%, including two patients with CCA.BRAF mutations have been identified in 5% of BTCs and are more common in ICC than in extrahepatic cholangiocarcinoma (ECC).[19,22] A recent phase II basket trial (NCT02034110) evaluated BRAFi + MEKi in BRAF V600E-mutant BTCs; it reported an investigator-assessed overall response in 22 (51%; 95% CI, 36–67) of 43 patients, and an independent reviewer-assessed overall response achieved by 20 (47%; 95% CI, 31–62) of 43 patients. Further, the responses were durable; the median duration of response in 22 patients with an investigator-assessed objective response was 9 months (95% CI, 6–14 months). Taken together, these data suggest that BRAF V600E is an important target across histologies and represents a promising target for BRAF V600E-mutant CCA. However, even with targeted therapies that presumably hit the “Achilles heel” of the tumor, while relatively sparing the normal tissue, often there are adverse events.
Skeletal muscle–related adverse events have been reported with the use of MEKi monotherapy as well as the combination of BRAFi + MEKi (Table 1).23–25 Although myalgia is not uncommon with BRAFi + MEKi (14–19%), the diagnosis of myositis is much rarer, and was reported in only 1 of 247 patients enrolled in the coBRIM trial (ClinicalTrials.gov identifier: NCT01689519) evaluating the combination of vemurafenib and cobimetinib in patients with BRAF V600-mutant melanoma. Despite initial improvement of our patient's CK and Cr with fluid resuscitation and corticosteroids, his symptoms worsened and CK levels continued to rise, which ultimately reversed only after PLEX and IVIG administration (Fig. 1).
There are several possible etiologies for myositis in this case, including immune-mediated necrotizing myositis, direct inhibitor effects on skeletal muscle, paraneoplastic syndromes, and statin-induced autoimmune myopathy. Oncogenic BRAF mutations have been associated with T cell suppression, and induction of the tumor–host immune response has been proposed as an off-target mechanism of MAPK inhibition. In several cancers, antigens have been identified that are morphologically similar to myoblasts, resulting in cross-reactivity of autoantibodies and resultant tissue damage. Interestingly, modulation of the FoxP3 transcription factor by MEKi has been proposed as a mechanism to reduce the activity of regulatory T cells. An amplified immune response secondary to MAPK pathway inhibition, in combination with cross-reactivity of autoantibodies, may explain the presentation of myositis in our case. That being said, immune-mediated necrotizing myositis is often associated with antisignal recognition particle (SRP) antibodies and/or anti-3-hydroxy-3-methylglutaryl-coenzyme A reductase (HMGCR) antibodies, for which the former was normal, and the latter not evaluated. Myositis secondary to direct MEK inhibitor toxicity on skeletal muscle is also a consideration, as it has been seen with cobimetinib and selumetinib. While the effects of MEK inhibition on normal myocytes has not been elucidated, inhibition may have effects on cell survival and homeostasis. A case report from Harrison et al. described a patient with metastatic melanoma on dabrafenib and trametinib who developed dermatomyositis refractory to corticosteroids but responded to IVIG. The absence of skin changes and anti-transcription intermediary factor 1-γ, (TIF1) and/or anti-nuclear matrix protein (NXP) antibodies in our case make this diagnosis unlikely. Statin-induced autoimmune myopathy is also a consideration given our patient's history of rosuvastatin use and necrotic muscle fibers on histopathology, although it is not commonly associated with infiltrating lymphocytes.
To our knowledge, this is the first report of steroid-refractory myositis in a patient receiving the combination of vemurafenib and cobimetinib, requiring PLEX and IVIG. It is essential for clinicians to be aware of skeletal muscle–related adverse events in the setting of dual BRAF and MEK inhibition.
Sources of Support: This work was supported in part by the following: National Institute of Health (NIH) Training of Academic Surgical Oncologists (5T32CA009599-32), Sheikh Khalifa Bin Zayed Al Nahyan institute for Personalized Cancer Therapy, NIH/NCATS Center for Clinical and Translational Sciences (CCTS)/MD Anderson Translational Informatics (UL1 TR003167), and National Cancer Institute (NCI) Cancer Center Support (CORE) Grant (5P30CA016672-44). The Targeted Agent and Profiling Utilization Registry (TAPUR) trial (NCT02693535) was sponsored by the American Society of Clinical Oncology (ASCO).
Conflicts of Interest: Vivek Subbiah reports research funding/grant support for clinical trials from Roche/ Genentech, Novartis, Bayer, GlaxoSmithKline, Nanocarrier, Vegenics, Celgene, Northwest Biotherapeutics, Berghealth, Incyte, Fujifilm, Pharmamar, D3, Pfizer, Multivir, Amgen, Abbvie, Alfa-sigma, Agensys, Boston Biomedical, Idera Pharma, Inhibrx, Exelixis, Blueprint medicines, Loxo oncology, Medimmune, Altum, Dragonfly therapeutics, Takeda, National Comprehensive Cancer Network, NCI-CTEP, Univesity of Texas MD Anderson Cancer Center, Turning Point Therapeutics, Boston Pharmaceuticals; travel support from Novartis, Pharmamar, ASCO, ESMO, Helsinn, Incyte. Consultancy/ Advisory board: Helsinn, LOXO Oncology/Eli Lilly, R-Pharma US, INCYTE, QED pharma, Medimmune, Novartis, Relay Therapeutics, and Roche; and other support from Medscape. Funda Meric-Bernstam reports consulting fees from F. Hoffman-La Roche Ltd., Genentech Inc., Pfizer Inc. and sponsored research to the institution from Genentech Inc. The remaining authors have no relevant disclosures.