Milademetan, an oral MDM2 inhibitor, in well-differentiated/ dedifferentiated liposarcoma: results from a phase 1 study in patients with solid tumors or lymphomas

Abstract

Results: 75 pts were treated with PLX8394 (450 mg BID, 450 mg TID, and 900 mg BID) with (N = 56) or without cobi (n = 19). Pts had BRAF V600 (n = 49), BRAF non-V600 (n = 17) or non-BRAF mutated (n = 9) tumors. The use of cobi resulted in 2-3-fold increase of PLX8394 exposure, demonstrating a non-saturating dose proportional increase in exposure. Grade 3 (G3) AST elevation and Grade 3 (G3) blood bilirubin increase were the only DLTs and PLX8394 900 mg BID with cobi was declared as RP2D. Other ≥G3 toxicities in ≥2 pts included increased ALT (4), increased AST (3), blood bilirubin increase (4), and diarrhea (2). Of 45 evaluable pts with BRAF-mutations, who received a median of 3 prior therapies, including 12 pts (27%) which received prior MAPK pathway targeted therapies, 10 pts (22%) achieved confirmed and mostly durable partial responses (gliomas (3), ovarian (2), and 1 each for papillary thyroid cancer, small bowel, colorectal carcinoma, anaplastic thyroid carcinoma, and melanoma). At time of analysis 10 pts remain on study for ≥24 months (range, 24-59 months) providing encouraging long-term safety data including lack of secondary skin lesions observed with Class 1 BRAF inhibitors. Conclusions: PLX8394 + cobi demonstrated favorable safety profile and encouraging activity in refractory solid tumors with BRAF mutations including BRAF fusion. Further studies to refine the RP2D with a modified tablet formulation to reduce dose burden and improve dose linearity are ongoing. Background: Detection of molecular alterations in circulating tumor DNA (ctDNA) has been recently established as a tool to detect molecular targets for cancer therapy. Preliminary data suggest that dynamic changes in ctDNA quantity can be associated with outcomes of cancer therapy. Material and Methods: We isolated ctDNA from serially collected blood samples from eligible patients with advanced cancers obtained during treatment on early-phase clinical trials with systemic therapies. Blood samples were collected at baseline (BL) (C1D1), mid-cycle (MC) (C1D21), at time of the first restaging (FR), and if feasible throughout the entire course of their therapy. We performed genomically-informed (by the results of molecular testing of tumor tissue) testing of ctDNA using digital droplet PCR (BioRad) and the quantity of ctDNA was measured as mutant variant allele frequency (VAF). Patients were classified based on results of their first restaging imaging as responders (complete [CR] or partial response [PR]) vs. non-responders (stable disease [SD] or progressive disease [PD]), or progressors (PD) vs. non-progressors (CR, PR, SD). Delta value as well as slope value (as a reflection of the trend of ctDNA dynamic) were calculated at MC and at FR and were correlated with clinical outcomes. Results: A total of 218 patients with advanced cancers (most frequent tumor types: gastrointestinal cancers, n = 101; melanoma, n = 35; and breast cancer, n = 24) treated in early-phase clinical trials (immunotherapy, n = 31; non-immunotherapy, n = 187) were enrolled. Progression in ctDNA quantity preceded or co-occurred with clinical or radiological PD in 65.6% of patients. Detectable ctDNA at BL, MC, or FR was associated with shorter time to treatment failure (10 wks vs. 18 wks; 10 wks vs. 20 wks; 10 wks vs. 24 wks; all P < 0.001). Quantity of ctDNA as aggregate VAF was higher in progressors compared to non-progressors (9.7% vs. 4.5%; 8.7% vs. 3.8%; 13.6% vs. 3.5%; all P < 0.001) and in non-responders compared to responders (7.4% vs. 3.3%, P = 0.036; 6.8% vs. 0.6%, P < 0.001; 7.9% vs. 2.7%, P = 0.002) at all timepoints. FR delta value was significantly higher in progressors compared to non-progressors (13.6% vs 3.5%, P = 0.004) and in non-responders compared to responders (7.9% vs 2.7%, P = 0.026). Delta value positivity and slope value positivity both correlated with higher likelihood of disease progression, lesser chance of response and shorter time to treatment failure (all P < 0.05). Conclusions: Early dynamic changes in the quantity of ctDNA within the first few weeks of therapy can predict clinical outcomes of systemic therapies in advanced cancers.