Supplementary Tables and Figures from A Phase Ib/II Study of Ivosidenib with Venetoclax ± Azacitidine in <i>IDH1</i>-Mutated Myeloid Malignancies

<p>Table S1: Pharmacokinetics of IVO+VEN and IVO+VEN+AZA Table S2: Treatment characteristics across the four dose levels in the phase 1b study population. Table S3: Study medication adherence: (A), Adherence to protocol administered medications during DLT evaluation period (cycles 1 and 2) of protocol directed therapy. (B), Adherence to protocol administered medications during the first 12 months of protocol directed therapy. Table S4: Dose modifications in patients experiencing hematologic adverse events during the entire study period until data cut-off on March 15th, 2022. Table S5: Regions covered by institutional next-generation sequencing (NGS) interrogating the entire exonic or hotspot regions of 81-genes frequently mutated in myeloid malignancies. Table S6: Variants identified using targeted 81-gene myeloid NGS panel Table S7: CyTOF antibodies utilized. Figure S1: Molecular, cytogenetic, and IDH1 variants across the P1b study population. (A), Oncoprint of molecular mutations identified at diagnosis in 31 patients enrolled demonstrated a diverse molecular landscape. (B-C), Neither median mutation burden nor IDH1 VAF was significantly different between included disease types. (D), IDH1 variants differed across disease types, with R132C mutations most frequent at diagnosis in ND and R/R-AML. IDH1 R132L and R132S variants were not identified in patients with MDS or MPN. (E), Clonal hierarchy of baseline mutations assessed using bulk myeloid NGS panel on baseline samples. Figure S2: Pharmacokinetics of IVO+VEN or IVO+VEN+AZA (continued): (A), AUC 24 and (B), Cmax of IVO+VEN or IVO+VEN+AZA within each respective dose level demonstrated a decrease in the presence of IVO when assessed on C2D14 compared to sampling on C1D14 in the absence of co- occurring IVO administration. Figure S3: CONSORT diagram of study participants Figure S4: Bone marrow response and recovery following IVO+VEN or IVO+VEN+AZA: (A), Bone marrow blast reduction following one cycle of therapy with IVO+AZA, or IVO+VEN+AZA among all patients with an adequate bone marrow samples (N=28). Three patients had inadequate/hypocellular EOC1 bone marrow aspirations and were excluded. One patient enrolled with MRD+ AML only. (B) Cycle lengths during the first four cycles of treatment. Patients treated with IVO+VEN had significantly shorter cycle lengths for cycle 1 compared to patients treated with IVO+VEN+AZA. Figure S5: Adverse events in patients treated with IVO+VEN or IVO+VEN+AZA: (A), Common adverse events occurring in four or more patients on study during the entire phase 1b study period demonstrated by grade, (B), by receipt of IVO+VEN vs. IVO+VEN+AZA, (C), and by dose level. Figure S6: Morphologic and MRD response to IVO+VEN in ND-AML. Morphologic response, MRD-MFC, and IDH1 ddPCR status by treatment cycle in patients with ND-AML treated with the doublet combination of IVO+VEN 400 or VEN 800 Figure S7: Morphologic and MRD response to IVO+VEN+AZA in ND-AML. Morphologic response, MRD-MFC, and IDH1 ddPCR status by treatment cycle in patients with ND-AML treated with the triplet combination of IVO+VEN 400 or VEN 800 +AZA Figure S8: Morphologic and MRD response to IVO+VEN or IVO+VEN+AZA in R/R-AML. Morphologic response, MRD-MFC, and IDH1 ddPCR status by treatment cycle in patients with R/R-AML treated with IVO+VEN with or without AZA Figure S9: Response and outcome of patients treated with IVO+VEN or IVO+VEN+AZA. (A), Swimmer's plot of patients on study with IVO+VEN +/- AZA by dose level. Figure S10: Landmark overall survival analyses based on IDH1 mutation detection in CRc using ddPCR as though performed on patients at baseline. (A), Overall survival from treatment start. (B), Including patients surviving at least 3-months (correlating with end of cycle 3). (C), Including patients surviving at least 5- months (correlating with end of cycle 5). (D), Including patients surviving at least 7-months (correlating with end of cycle 7). Figure S11: Influence of biological pathways on survival following IVO+VEN or IVO+VEN+AZA treatment. (A), Overall survival by methylation mutations. (B), Overall survival by signaling mutations. (C), Overall survival in all patients with signaling mutations based upon receipt of IVO+VEN or IVO+VEN+AZA. (D), Overall survival in patients with AML and signaling mutations based upon receipt of IVO+VEN or IVO+VEN+AZA. Figure S12: Persistent mutations in a long-term responder with ND-AML identifies mutations within a CD16+ monocytic population. (A and B), DAb-seq analysis at diagnosis and in remission in a patient with ND-AML (accession #18) and co-occurring RUNX1 p. L204Q and IDH1 p.R132H mutations treated with IVO+VEN+AZA demonstrated treatment eliminated the majority of IDH1 and RUNX1 co-mutated cells. (C and D), The predominant population of CD34+ myeloblasts was eliminated with therapy, however residual cells containing IDH1 and RUNX1 in remission were clustered with a monocytic cell population with increased CD16 expression (C and D). Figure S13: Immunophenotypic shift occurring under treatment selection with targeted therapy. Multiparameter flow cytometry in a patient with ND-AML (accession #20) demonstrating an alternative CD34+ population expanding at relapse compared to baseline, consistent with a phenotypic shift at disease. Figure S14: Increased alternative antiapoptotic protein expression levels correlate with resistance to IVO+VEN+AZA. CyTOF analysis in a patient with ND-AML (accession #11) treated with IVO+VEN+AZA who initially attained CRc followed by disease progression. Increased alternative anti-apoptotic protein levels (BCL-xL and MCL1) were observed, in addition to increased CD44 levels. Figure S15: Increased BCL2 levels relative to alternative anti-apoptotic proteins is associated with ongoing response to IVO+VEN+AZA. CyTOF analysis in a patient with ND-AML (accession #26) treated with IVO+VEN+AZA with a durable response to treatment following 18 cycles of therapy. The patient had multiple CD34+ cell populations at diagnosis, with higher BCL2 levels relative to BCL-xL or MCL-1. Following 3 cycles of therapy, marked reduction in these blast populations were observed. Figure S16: Increased alternative anti-apoptotic protein expression is observed in maturing myeloid populations with monocytic differentiation. CyTOF analysis in a patient with R/R-AML (accession #10) treated with IVO+VEN with a durable response to treatment following 41 cycles of therapy. Following cycle 3 of treatment, an expanding CD34+ cell populations with increased BCL-xL and MCL-1 levels was observed with an associated maturing CD14+ monocytic immunophenotype.</p>