Selumetinib Response Not Driven by MAPK Expression in Pediatric and AYA Refractory Solid Tumors

Olive S. Eckstein, MD

Selumetinib did not elicit clinical activity in pediatric and young adult patients with refractory cancers who had actionable mutations in the RAS/RAF/MAPK1/2-ERK pathway, according to cohort findings from the phase 2, single-arm National Cancer Institute (NCI)-Children’s Oncology Group (COG) Pediatric MATCH trial (NCT03155620) that were presented during the 2021 ASCO Annual Meeting.¹

Although the agent was well tolerated, no objective responses were observed with selumetinib. Three patients achieved stable disease with the agent, while 15 patients had progressive disease. One patient was not RECIST classifiable, and 1 patient was not evaluable. Additionally, the 6-month progression-free survival (PFS) rate was 15% (95% CI, 4%-34%).

“Selumetinib was generally well tolerated in this cohort of children, who had diverse treatment-refractory cancers. Unfortunately, no objective responses were observed,” Olive S. Eckstein, MD, of Texas Children’s Hospital/Baylor College of Medicine and an investigator on the trial, said in a presentation during the meeting. “The big point to discuss and to think about in the future, when we design these trials, is histology and mutation status. We learned here that MAPK pathway mutation status alone is insufficient to predict response to selumetinib monotherapy.”

Although pediatric cancers harbor targetable molecular alterations that cross multiple disease histologies, similar to adult malignancies, the low frequency of individual gene alterations makes evaluating targeted agents in pediatric tumors challenging, Eckstein explained.

Several malignancies have been known to have a high frequency of MAPK pathway mutations, such as RAS (hairy cell leukemia, melanoma, and papillary thyroid cancer), BRAF (hairy cell leukemia, melanoma, Langerhans Cell Histiocytosis, and papillary thyroid cancer), and MEK1/2 (colorectal cancer, glioma, Langerhans Cell Histiocytosis, and ovarian cancer). Of these pathway mutations, RAS is the most frequently observed, at 22%, followed by BRAF (7%), MEK1/2 (<1%), and ERK (rare).

Selumetinib is a potent, orally bioavailable selective inhibitor of MEK1/MEK2 that blunts the RAS-RAF-MEK1/2-ERK cascade, reduces cell proliferation, and promotes pro-apoptotic signal transduction. The TKI has previously shown efficacy in the phase 1 Pediatric Brain Tumor Consortium trial in low-grade gliomas;² because of these data, investigators chose to not include this patient subgroup in this screening protocol.

Selumetinib is also approved by the FDA for the treatment of patients with plexiform neurofibroma who are aged 2 years or older.

The data presented at during 2021 ASCO Annual Meeting were part of the NCI-COG Pediatric MATCH trial (APEC1621SC), which facilitates molecular profiling of treatment-refractory cancers in US pediatric and young adult patients. Patients between the ages of 1 year and 21 years with refractory solid tumors, lymphomas, and histiocytosis had to have measurable disease and adequate performance status.

Patients had their tumors biopsied before moving onto the screening protocol for tumor sequencing. If actionable mutations in the RAS/RAF/MAPK1/2-ERK pathway were detected—NF1, NRAS, KRAS, HRAS, ARAF, MAP2K1, GNA11, GNAQ mutations or BRAF mutations or fusion—patients were enrolled on the selumetinib subprotocol of pediatric MATCH (APEC1621E), arm E of the trial, which is designed to test the tolerability and efficacy of selumetinib in children with relapsed tumors, which contain molecular MAPK pathway alterations.

Also, on this subprotocol, patients needed to meet standard phase 2 protocol criteria, consisting of organ function and washout periods. FFPE tumor samples were centrally processed, and Targeted Oncomine was harmonized and performed in 3 laboratories.

Selumetinib monotherapy was administered orally at 25 mg/m² twice daily, with a maximum dose of 75 mg/m² for up to 2 years in 28-day cycles. Responses were evaluated every other cycle for 3 occurrences, and then every 3 cycles.

Eckstein noted that the trial would have potential for histology-specific expansion cohorts if objective responses were observed in at least 3 patients with the same histology.

The primary end point of the study was ORR; secondary end points included PFS and tolerability.

Twenty-one patients were enrolled on the trial, and 20 were eligible for selumetinib treatment. One patient was deemed ineligible because they had low-grade glioma. The median age was 14 years (range, 5-21 years), and 50% of patients were male. Most patients (n = 15) were Caucasian, and 1 patient each was African American, Asian, and Native American. Two patients’ ethnicities were not reported.

Eleven hotspot mutations were found in RAS genes: KRAS (n = 8), NRAS (n = 3), and HRAS (n = 1); 7 patients had inactivating NF1 mutations, and 2 patients had activating missense mutations in BRAF V600E.

The most common tumor histology subtypes were central nervous system astrocytoma (n = 7), of which there were BRAF mutations (n = 2), NF1 (n = 2), KRAS (n = 1), NF1 plus PTEN (n = 1), and KRAS plus NF1 (n = 1); and rhabdomyosarcoma, of which were had NRAS mutations (n = 2), KRAS (n = 4), and HRAS mutations (n = 1).

Additional tumor histologies were carcinoma (n = 2), neuroblastoma, plexiform neurofibroma, Yolk sac tumor, and other sarcoma (n = 1 each). Carcinoma harbored 1 mutation each of KRAS and NF1, neuroblastoma of 1 NRAS mutation, plexiform neurofibroma of NF1 plus BRCA2, Yolk sac tumor of 1 KRAS, and other sarcoma of 1 NF1 mutation.

The median number of completed cycles was 2 (range, 1-13), and the data cutoff date was December 31, 2020.

Of the 3 patients with SD, 1 had high-grade NF1-PTEN–mutant glioma and had received 6 cycles of treatment until disease progression, the second had high-grade KRAS-mutant glioma and received 12 cycles until experiencing disease progression, and the third had NF1-BRCA2–mutant plexiform neurofibroma and was treated for 13 cycles until experiencing grade 4 elevated creatine kinase (CPK) elevation.

Adverse effects that were potentially linked with selumetinib therapy included lymphopenia, uveitis, and thromboembolic event (n = 1 each; all grade 3), elevated CPK (n = 1; grade 4), and thromboembolic event through pulmonary embolism that led to death (n = 1; grade 5). The patient who died had adenocarcinoma and was on treatment for 11 days.

Future directions should focus on identifying histologic, molecular, and clinical features of responder and non-responders with gliomas, as well to determine biological features for predictive response to MEK inhibitors as single agents and exploring the synergy with selumetinib or other MEK inhibitors in combination with targeted or cytotoxic agents, Eckstein concluded.

References

1. Allen CE, Eckstein O, Williams PM, et al. Selumetinib in patients with tumors with MAPK pathway alterations: Results from Arm E of the NCI-COG pediatric MATCH trial. J Clin Oncol. 2021;39(suppl 15):10008. doi:10.1200/JCO.2021.39.15_suppl.10008
2. Banerjee A, Jakacki RI, Onar-Thomas A, et al. A phase I trial of the MEK inhibitor selumetinib (AZD6244) in pediatric patients with recurrent or refractory low-grade glioma: a Pediatric Brain Tumor Consortium (PBTC) study. Neuro Oncol. 2017;19(8):1135-1144. doi:10.1093/neuonc/now282