Exceptional responders with invasive mucinous adenocarcinomas: a phase 2 trial of bortezomib in patients with KRAS G12D-mutant lung cancers.

KRAS G12D-mutant/p53-deficient non-small-cell lung cancer (NSCLC) models are dependent on the NF-κB pathway that can be down-regulated by the proteasome inhibitor bortezomib. Two exceptional responders were observed on prior clinical trials of bortezomib, both of whom had KRAS G12D-mutant NSCLC, prompting the initiation of this single-center phase 2 trial. Patients with advanced KRAS G12D-mutant NSCLC were eligible. Bortezomib was administered at 1.3 mg/m2 subcutaneously (days 1, 4, 8, 11; 21-d cycle) until progression or unacceptable toxicity. The primary objective was best objective response (RECIST v1.1). Sixteen patients with KRAS G12D-mutant lung adenocarcinomas were treated. Patients had a median pack year smoking history of 4 (range 0-45). A partial response (PR) was observed in one patient (-66% from baseline) and stable disease in five patients on the first stage of this study (overall response rate of 6%, 95% CI: 0.2-30.2), and further patients were not accrued. The median progression-free survival was 1 mo (95% CI: 1-6). The median overall survival was 13 mo (95% CI: 6-NA). The most common treatment-related adverse events were fatigue (38%) and diarrhea (26%). TP53 status did not predict response on exploratory testing. Of note, the patient with a PR had a unique subtype of lung adenocarcinoma-invasive mucinous adenocarcinomas (IMA)-and had rapid clinical improvement and substantial disease regression, which was also previously observed in two other patients with advanced KRAS G12D-mutant lung cancer with IMAs who received bortezomib on separate clinical trials. Exceptional responses to bortezomib can be achieved in KRAS G12D-mutant NSCLCs. KRAS G12D mutation alone, however, is not a robust predictor of response. Further evaluation should only be performed after further elucidation of other factors such as co-occurring alterations and histologic subtype such as IMA that may predict sensitivity to therapy.