Introduction: Ibrutinib is a first-in-class small molecule inhibitor that binds irreversibly to Bruton's Tyrosine Kinase (BTK) and has shown remarkable efficacy in the treatment of CLL. Current guidelines recommend life-long therapy with administration at a fixed daily dose of 420 mg. However, lower doses of ibrutinib have been demonstrated to adequately abrogate kinase function. Limited clinical data suggests that dose reductions are not associated with inferior outcomes. Hypothetically, judicious dose reductions to manage toxicities could result in improved tolerance and decreased discontinuation rates. Our objective was to study the impact of dose reductions on outcomes in CLL patients treated with ibrutinib in a real-world setting. Methods: We conducted a retrospective chart review of all CLL patients treated with ibrutinib at Roswell Park Comprehensive Cancer Center between January 2014 and June 2017. Patients who underwent ibrutinib dose reduction were identified. Baseline characteristics and outcomes were compared between patients who underwent dose reductions vs those who did not. Reason and timing of dose reduction was also elucidated. Mann-Whitney U and Fisher's Chi Square test were used to compare groups, Kaplan Meier methods were used for time to event analysis, and Cox regression was used to obtain hazard ratios (HR). Overall survival (OS) was calculated as time from start of ibrutinib until death or last follow-up, freedom from progression (PFS) was taken as time from start of ibrutinib until progression or last follow-up. All analyses were performed in SAS v9.4 (Cary, NC). Results: A total of 70 CLL patients treated with ibrutinib were followed for a median period of 21.9 months. Most patients had RR disease (n = 63) and 7 received ibrutinib as frontline therapy. All patients received a fixed dose regimen with the standard dose of 420 mg once daily. Twenty-three (31.3%) patients required dose reductions and received ibrutinib at a median dose of 140 mg . Eleven (47.8%) of these had dose reductions within 3 months of treatment initiation. There was no statistically significant difference in baseline characteristics including age, number of prior lines of treatment, WBC count, hemoglobin or LDH level in the dose reduced group (DRG) and standard dose group (SDG). Patients in the DRG had a lower median platelet count at initiation of ibrutinib than patients in the SDG (86 x 103/mm3 vs 145 x 103/mm3, p<0.05). The most frequent reasons for dose reductions included cytopenia (n=8), fatigue (n=5), infections (n=4) and atrial fibrillation (n=3) with 2 patients requiring dose reduction due to simultaneous CYP inhibitor administration. There was no statistically significant difference in the overall response rate (ORR) (p=0.25) and clinical benefit rate (CBR) (p=1.00) between the DRG (ORR- 65.2%, CBR- 91.4%) and SDG (ORR 78.7%, CBR 91.5%). Median PFS and OS was not reached in both groups. PFS at 12 months was 86% (95% CI 63-95%) and 87% (95% CI 71- 94%) in the DRG and SDG respectively (p=0.92). OS at 12 months was 83% (95% CI 60-93%) and 84% (95% CI 69-92%) in the DRG and SDG respectively (p=0.53). We noticed no statistically significant difference in the median PFS and OS between the two groups (Figure 1). In a subgroup analysis, we compared outcomes in patients in the SDG with those who had dose modifications within 3 months of initiating therapy or later and found no statistically significant differences in OS and PFS between the groups (p=0.15 and p=0.46 respectively). Of the 23 patients requiring dose modifications, 11 (47.8%) had to eventually discontinue therapy as compared to 17 (36.2%) patients in the SDG and we noticed similar 1-year discontinuation rates in the DRG and SDG groups (22% vs 21% respectively, p=0.91). Conclusion: In CLL patients treated with ibrutinib, reductions in the dose of ibrutinib were not associated with worse outcomes. These results extended to all patients with dose modifications, irrespective of whether the modification was made within 3 months of treatment initiation, or later in the course of treatment. Disclosures No relevant conflicts of interest to declare.

No abstract

8030 Background: Multiple myeloma (MM) usually responds to induction therapy but relapses with therapy-resistant disease. CD28 expressed on MM cells is correlated with worse outcomes. Bone marrow stromal cells expressing CD28 ligands CD80 and/or CD86 are cellular partners in the MM niche transducing a pro-survival signal to MM cells contributing to therapy resistance in relapsed disease. We have previously shown in in vitro and in vivo preclinical studies that blocking the pro-survival CD28 activation on MM cells with abatacept (CTLA4 IgG binding to CD80/CD86 and blocking engagement to CD28) reverses chemotherapy resistance and re-sensitizes MM cells to drugs they previously were resistant to. Methods: We tested efficacy and safety of combining Abatacept with the proteasome inhibitor (PI) Ixazomib (Ixa) and Dexamethasone (Dex) in patients with MM who had relapsed (or primary refractory) disease following treatment with first line PI Bortezomib based regimen. Previous studies found that Ixa/Dex alone only had an 11% overall response rate (ORR) and 11% Clinical Benefit Rate (CBR) in patients with prior Bortezomib exposure. In our trial, patients with MM cells positive for CD28 or CD86 by flow cytometry or immunohistochemistry in any proportion were eligible. From September 11, 2018 to August 5, 2021, 15 patients received Abatacept loading dose cycle 1 day 1 followed by 125 mg subcutaneously on day 2 and then weekly. Patients received Ixa 4 mg on days 1, 8, and 15 and Dexamethasone 40 mg weekly of a 28-day cycle. The primary endpoint was ORR (partial response (PR) or better) according to International Myeloma Working Group criteria. Secondary end points were toxicity profile, progression-free (PFS), and overall survival (OS). Results: Median age was 62.8 years (range 50-83.9). Ten (66.6%) patients received prior autologous stem cell transplant. The ORR was 33.33% (90% CI 16.58%-54.46%) (one-sided Binomial exact test 0.4845). Complete remission (CR) was achieved in 6.7%, and 26.7% achieved PR with 53.3% of patients having stable disease (SD). Median time to best response among patients with CR and PR was 8.9 weeks (95% CI 4–21). Median time on treatment was 23 weeks (95% CI 12.143-48.143) for all patients and 52.5 weeks (95% CI 39.286-68) for patients with response. One-year PFS was 45% (90% CI 21%-66%), median PFS 12 months (90% CI 5.7-25.9) and 1-year OS 100%, median OS not reached. Two grade 3 treatment emergent adverse events (TEAE) (diarrhea and low platelets) occurred requiring hospitalization. Grade 1/2 GI TEAEs were most common. No grade 3/4 treatment related infections were observed and 2 grade 3 infections occurred. Conclusions: Despite prior exposure to Bortezomib, one third of patients had response to Ixa with Abatacept and Dex. Patients with response stayed on treatment for almost 1 year on average with limited AEs. Most patients on trial gained benefit with a clinical benefit rate (CR+PR+SD) of 86.66%. Clinical trial information: NCT03457142 .

The CORAL study highlighted the need to develop novel salvage regimens in relapsed/refractory (R/R) diffuse large B-cell lymphoma (DLBCL) previously treated with R-CHOP. Carfilzomib (CFZ) can overcome rituximab-chemotherapy resistance in lymphoma pre-clinical models by targeting the ubiquitin-proteasome system (UPS). We conducted an investigator initiated, single-center, open-label, prospective phase 1 study evaluating the safety and efficacy of CFZ in combination with R-ICE in high-dose chemotherapy with autologous stem cell transplant (HDC-ASCT) eligible R/R DLBCL pts (NCT01959698). In the dose-escalation phase, 18 pts were enrolled at six dose levels with no dose-limiting toxicities noted. CFZ 45 mg/m2 was selected as the recommended dose for expansion. Eleven additional pts were enrolled in the dose-expansion phase. Overall response rate (ORR) was 66% (48% CR, 17% PR), 52% pts underwent HDC-ASCT. An ORR of 85% was observed in pts with non-germinal center B-cell-like (non-GCB) DLBCL compared to only 13% in GCB-DLBCL. Median PFS was 15.2 months (5.1 mo- not reached), and median OS was 22.6 months (6.8 mo- NR). Pts with non-GCB subtype had a significantly longer PFS (NR vs. 6.6 mo, p= 0.0001) and OS (NR vs. 6.6 mo, p= 0.001) than those with GCB-subtype. C-R-ICE is well tolerated in pts with R/R DLBCL with toxicities comparable to R-ICE therapy. Our data show that pts with non-GCB DLBCL benefit significantly from incorporating CFZ into second-line therapy and HDC-ASCT.

No abstract