BACKGROUND:Human epidermal growth factor receptor 2 (HER2)-positive metastatic gastric and gastroesophageal adenocarcinoma (GEA) is globally treated with chemotherapy plus trastuzumab. Novel therapeutic strategies strive to not only optimize efficacy, but also limit toxicities. In MAHOGANY cohort A, margetuximab, an Fc-engineered, anti-HER2 monoclonal antibody (mAb) was combined with retifanlimab, an anti-programmed cell death protein 1 mAb, in the first-line HER2-positive/programmed death-ligand 1 (PD-L1)-positive GEA. PATIENTS AND METHODS:MAHOGANY cohort A part 1 is a single-arm trial to evaluate margetuximab plus retifanlimab in patients with HER2 immunohistochemistry 3+, PD-L1-positive (combined positive score ≥1%), and non-microsatellite instability-high tumors. Primary objectives for cohort A were safety/tolerability and the confirmed objective response rate (ORR). RESULTS:As of 3 August 2021, 43 patients were enrolled and received margetuximab/retifanlimab. Nine grade 3 treatment-related adverse events (TRAEs) were reported in eight (18.6%) patients and eight serious TRAEs in seven (16.3%) patients. There were no grade 4/5 TRAEs. Three patients discontinued margetuximab/retifanlimab because of immune-related adverse events. The ORR by independent assessment was 53% [21/40 (95% confidence interval (CI) 36.1-68.5)], with a median duration of response of 10.3 months (95% CI 4.6-not evaluable); disease control rate was 73% [29/40 (95% CI 56.1-85.4)]. The study sponsor discontinued the study in advance of the planned enrollment when it became apparent that the study design would no longer meet the requirements for drug approval because of recent advances in the treatment of GEA. CONCLUSIONS:The chemotherapy-free regimen of combined margetuximab/retifanlimab as first-line treatment in double biomarker-selected patients demonstrated a favorable toxicity profile compared with historical outcomes using chemotherapy plus trastuzumab. The ORR observed in this study compares favorably versus ORR observed with other chemotherapy-free approaches.

Platelet-derived growth factor receptor alpha (PDGFRα) is overexpressed in pancreatic cancer (PC) and contributes to activation of pancreatic stellate cells and formation of characteristic desmoplastic stroma. Despite availability of treatments, PC has a poor prognosis, indicating an unmet need for novel therapies. This study evaluated the efficacy and safety of olaratumab (OLA), an anti-PDGFRα recombinant human immunoglobulin G subclass 1–type monoclonal antibody with standard first-line therapy gemcitabine and nab-paclitaxel in treatment-naïve patients (pts) with unresectable metastatic PC.

The monoclonal antibody (mAb) trastuzumab plus chemotherapy has been 1st-line standard of care in HER2+ metastatic GEA for a decade. M is an investigational (in GEA) anti-HER2 mAb engineered to confer enhanced Fc dependent antitumor activity across all Fc region gamma receptor genotypes. M may potentiate innate immunity (i.e. antibody-dependent cellular cytotoxicity) and adaptive immunity (i.e. anti-HER2–directed T-cell responsiveness). R, an investigational humanized IgG4 mAb, binds to PD-1 and blocks its interaction with PD-L1/2. MAHOGANY is a 2-cohort study in unresectable metastatic/locally advanced GEA. Cohort A (100 pt single arm) will determine safety/efficacy of Q3W M 15 mg/kg plus R 375 mg in HER2 3+, PD-L1+ (CPS ≥1) and non–microsatellite instability high pts. Primary endpoint of Cohort A is centrally-reviewed overall response rate (ORR); Investigator-assessed ORR is supportive. As of 16 April 2021, 43 pts were enrolled, with median exposure of 4.2 months and median duration of follow-up of 4.6 months. Safety data are available on 43; efficacy data are available on 43 with baseline scan for response assessment by the investigators. Of these 43, 37 have reached first tumor reassessment, with 1 confirmed complete response, 11 confirmed partial responses, 1 unconfirmed complete response, 11 unconfirmed partial responses, 7 stable disease, and 6 progressive disease (5 PD at first scan, 1 clinical PD). Tumor shrinkage was seen in 30/35 pts (85.7%) with at least 1 post-baseline target lesion measurement. The most common treatment-related adverse events were infusion related reaction (8/43, 18.6%), diarrhea and fatigue (6/43, 14% each). Grade 3 treatment-related events were singularly reported with no ≥ Grade 3 infusion related reaction or diarrhea. Regarding immune-related events, 1 pt discontinued due to renal dysfunction possibly related to M+R, 4 had hypothyroidism and 1 had immune-related renal dysfunction. This first report of MAHOGANY Cohort A of M plus R suggests the combination is well tolerated. The updated efficacy data including centrally-reviewed scans will be available at presentation.

ERBB Receptor Feedback Inhibitor-1 (ERRFI-1) encodes MIG6, which is a negative regulator of EGFR and ERBB2 signaling. Loss of function alterations at ERRFI-1 would be expected to promote oncogenesis, but the role of ERRFI-1 alterations in conferring sensitivity to targeted therapies remains to be fully investigated. We reviewed 19,347 cases of NSCLC in the Foundation Medicine data base for ERRFI-1alterations that had been previously assayed by hybrid-capture based genomic DNA profiling of FFPE tissue specimens. Two patients, so identified, had been treated with EGFR pathway antagonist therapies and their outcomes are reported herein. ERRFI-1 truncating mutations were identified in 0.62 % (120/ 19,347) of all screened NSCLC specimens. ERRFI-1 alterations were seen in all NSCLC histologic subtypes examined at similar frequencies: adenocarcinoma (0.7%), squamous carcinoma (0.3%), large cell carcinomas (0.8%), adenosquamous (0.6%), sarcomatoid (0.6%), and not otherwise specified (0.6%). Co-existing alterations included: P53 (59%), KRAS (19%), EGFR exon 19 del (9.2%), EGFR L858R (3.3%), EGFR T790M (3.3%), EGFR amp (6.7%), ERBB2 mut (7.5%), and ERBB2 amp (3.3%). Two female patients with ERRFI-1 mutations who were wildtype for known NSCLC driver mutations and targeted therapy naive, achieved RECIST criteria partial responses after treatment with single agent EGFR TKI therapies. Following subsequent disease progression, one of these patients also achieved a secondary response to single agent EGFR directed monoclonal antibody therapy. To our knowledge, these are the first two reported patient outcomes for targeted therapies in ERRFI-1 altered NSCLC. The index cases presented here suggest that NSCLC patients with genetic lesions in ERRFI-1 may respond to both anti-EGFR TKIs and monoclonal antibodies. However, co-occurrence between ERFFI-1 mutations and alterations in known NSCLC drivers such as EGFR exon 19 del and L858R may also indicate that in some contexts, ERRFI-1 alterations may provide a mechanism for acquired resistance to targeted therapies as well. Further investigation including assessment of ERRFI-1 loss of heterozygosity, ERRFI-1 VUSs , and clinical evaluation of additional cases including response and resistance to targeted therapy will be performed to more fully delineate the role of ERRFI-1 in NSCLC.