Eli Diamond, MD
The MEK inhibitor cobimetinib (Cotellic) led to objective responses in 14 of 16 patients with BRAF
-mutated and wild-type histiocytic disorders, according to results of a phase II trial presented at the 2017 ASH Annual Meeting.
The overall activity included 9 complete responses (CRs). Two additional patients had stable disease (SD), resulting in a clinical benefit rate (CBR) of 100%. Responding patients had a median time to best response of 2.6 months and a median duration of response (DoR) of 6.6 months.
“Single-agent cobimetinib has robust activity in Erdheim-Chester disease (ECD), Langerhans cell histiocytosis (LCH), and Rosai-Dorfman disease (RDD), regardless of BRAF mutational status,” said Eli Diamond, MD, a neuro-oncologist at Memorial Sloan Kettering Cancer Center. “We observed durable responses, up to 20.8 months of ongoing treatment, with no disease progression on treatment.”
BRAF inhibition has robust and durable efficacy against BRAF
-mutated histiocytosis. However, no comparably effective therapy exists for BRAF
wild-type disease, said Diamond. Recurrent MAP
-kinase pathway alterations have been observed in ECD, LCH, and RDD, providing a rationale for treatment with a MEK inhibitor. Early experience with cobimetinib and trametinib (Mekinist) in wild-type ECD has been promising, but MEK inhibition had not been evaluated in a prospective clinical trial.
The phase II results presented were from an ongoing evaluation of cobimetinib at 60 mg daily for 21 days of a 28-day cycle in patients with histiocytosis. The trial protocol mandated a biopsy for patients who had unknown mutational status. Response was assessed every 8 weeks by FDG-PET imaging and by organ-specific imaging.
Investigators enrolled patients aged ≥16 years with any histiocytic neoplasm. Eligible patients had disease that had proven refractory to therapy, that was newly diagnosed but had multisystem involvement, or that was newly diagnosed with single-system involvement and end-organ dysfunction, or deemed unlikely to respond to conventional treatment.
Patients with BRAF
-mutated or wild-type disease were eligible, provided they had demonstrated intolerance to a BRAF inhibition or did not have access to a BRAF inhibitor. The primary endpoint was best overall response (BOR), as determined by PET response criteria. Secondary endpoints included BOR by RECIST criteria, time to progression, progression-free survival, DoR, CBR, overall survival, and safety.
Response assessment included as many as 5 target lesions with maximal standard uptake value (SUVMAX
) twice the surrounding normal organ. Complete metabolic response was defined as normalization of all lesions to background SUV. Partial metabolic response was defined as ≥50% decrease from baseline in the sum of SUVMAX
for all target lesions relative to background SUV.
Progressive disease was defined as ≥50% increase in the sum of all SUVMAX
of target lesions, but a minimum absolute increase of 3 units SUV per target lesion, or presence of new evaluable lesions representing unequivocal disease progression. Stable disease referred to lesions that did not meet any of the other response/progression criteria.
The 16 patients evaluable for efficacy and safety included 4 with RDD and 1 with LCH. Multiple types of mutations were identified, including BRAF V600E, MAP2K1, MAP2K2, KRAS, CRAF, ARAF,
indeterminate. The patients had varied treatment histories that included interferon, anakinra (Kineret), steroids, immunosuppression, chemotherapy, radiation therapy, vemurafenib (Zelboraf), and no treatment.
Assessment of the primary endpoint showed that 9 of 16 patients met PET/organ-specific criteria for CR, and 5 others met criteria for partial response, resulting in an objective response rate of 87.5%. The remaining 2 patients had SD.
The time to best response ranged from 1.8 to 10.8 months, and DoR ranged from 1.0 to 20.8 months. No disease progression occurred during treatment. Four patients discontinued treatment: 1 death from pneumonia that was considered unrelated to treatment, 1 retinal vein occlusion considered treatment-related, and 2 patients withdrawing consistent (1 to pursue other treatment and 1 because of financial hardship). The remaining 12 patients continued on treatment.
Assessment of cytokine/chemokine levels showed downward trends in all evaluable patients (minimum of 5 patients). Analysis of patient-reported outcomes showed significant improvement from baseline in anxiety (P
=.007), depression (P
=.003), and symptom severity (P
=.02). Pain scores and overall quality of life did not change significantly.
Toxicity was consistent with previous experience with cobimetinib. No grade 3/4 adverse event occurred in more than 2 patients; these included 2 cases each of anemia, diarrhea, dyspnea, hyperglycemia, hypotension, increased lipase, decreased lymphocyte count, and respiratory failure.