Lee, Stephanie J.

Abstract Newer approaches to control alloreactivity may produce similar transplant outcomes using HLA-mismatched donors vs HLA-matched unrelated donors (MUD). However, prospective comparisons are lacking. BMT CTN 1702 used a donor search prognosis score to assign patients to an 8/8 HLA MUD or the center’s preference of haploidentical related donors (HAPLO), mismatched unrelated donors (MMUD), or umbilical cord blood (UCB). Outcomes of MUD were compared to HAPLO, MMUD, and UCB transplantation after adjusting for covariates. Patients (n = 1179 [93% adults]) underwent transplantation with MUD (n = 772), HAPLO (n = 254), MMUD (n = 112), and UCB (n = 41) at a median of 3.7, 3.4, 3.9, and 3.8 months from enrollment. Posttransplant cyclophosphamide (PTCy) was used in 23.9%, 83.9%, 65.2%, and 0% of MUD, HAPLO, MMUD, and UCB. In multivariate analyses, compared to MUD, survival was lower for UCB (hazard ratio [HR], 2.65; P< .001) but not statistically different for HAPLO and MMUD (HRs, 1.08 and 1.18, respectively). Relapse risk was not significantly different by donor, but treatment-related mortality (HR, 3.31; P< .001) and disease-free survival (HR, 1.99; P = .002) were inferior for UCB but not different for HAPLO and MMUD than MUD. In patients who received PTCy, HAPLO and MMUD were associated with increased grade 3 or 4 acute graft-versus-host disease (GVHD; HR, 2.39 [P = .017] and 2.53 [P = .038], respectively) and chronic GVHD (HR, 1.71 for both; P = .028 and .080) than MUD, but other outcomes were not different. HAPLO or MMUD may be used to expedite transplantation when finding MUD is unlikely. This study was registered at www.clinicaltrials.gov as NCT03904134.

PURPOSE The likelihood of finding a human leukocyte antigen (HLA)–matched unrelated donor (MUD) for hematopoietic cell transplantation can be predicted using a donor search prognosis score. Patients without a MUD may use alternative donors (haploidentical related, mismatched unrelated, or umbilical cord blood). METHODS This multicenter biological assignment trial was conducted by the Blood and Marrow Transplant Clinical Trials Network (BMT CTN 1702). Eligibility criteria were broad to mirror clinical practice. The primary end point was 2-year survival from evaluability and compared between those Very Likely (>90%) and Very Unlikely (<10%) to find a MUD. All other patients, Less Likely to find a MUD, were enrolled in an observational arm. Transplant outcomes were compared for all three groups. RESULTS A total of 1,751 evaluable patients at 47 centers were Very Likely (54.7%), Less Likely (29.5%), and Very Unlikely (15.8%) to identify a MUD. Survival did not differ in univariate (hazard ratio [HR], 1.00 [95% CI, 0.82 to 1.21]; P = .98) or multivariate (HR, 1.07 [95% CI, 0.86 to 1.33]; P = .56) analyses between the Very Unlikely and Very Likely groups, measured through 2 years from the beginning of a search for a MUD or alternative donor. Of the transplanted patients (n = 1,179), MUD was used for 94% of the Very Likely, 38% of Less Likely, and 9% of Very Unlikely patients. Multivariate analyses showed no differences in relapse, treatment-related mortality, disease-free survival, and acute and chronic graft-versus-host diseases for the three search prognosis groups after transplantation. CONCLUSION Using a donor search prognosis strategy to prioritize an alternative donor for patients Very Unlikely to find a MUD resulted in survival and transplant outcomes that were not statistically different compared with those Very Likely to find a MUD.

In REACH3 (ClinicalTrials.gov identifier: NCT03112603 ), ruxolitinib was investigated versus best available therapy (BAT) for 3 years in patients with steroid-refractory/dependent chronic graft-versus-host-disease (SR/D-cGVHD). Patients received ruxolitinib (10 mg twice daily) or BAT for 24 weeks; thereafter (weeks 24-156), patients continued randomized treatment, entered long-term survival follow-up, or crossed over from BAT to ruxolitinib. In 329 randomly assigned patients (ruxolitinib: 165; BAT: 164), the median failure-free survival (FFS) was 38.4 months for ruxolitinib versus 5.7 months for BAT (hazard ratio, 0.36 [95% CI, 0.27 to 0.49]). Median duration of response (DOR) was not reached for ruxolitinib versus 6.4 months for BAT. Ruxolitinib-treated patients had a higher probability of FFS (ruxolitinib: 56.5%; BAT: 18.2%) and maintaining a response (ruxolitinib: 59.6%; BAT: 26.7%) at 36 months. Median overall survival was not reached. Nonrelapse mortality and malignancy relapse/recurrence events were low. In 70 patients who crossed over to ruxolitinib, the overall response rate (50.0%) at week 24 and best overall response (81.4%) during the crossover period were consistent with the primary analysis of randomly assigned patients. No new safety signals were observed. Ruxolitinib provided longer FFS and DOR than BAT, demonstrating sustained efficacy and manageable safety over 3 years of follow-up in patients with SR/D-cGVHD.

Objectives/Goals: Manual skin assessment in chronic graft-versus-host disease (cGVHD) can be time consuming and inconsistent (>20% affected area) even for experts. Building on previous work we explore methods to use unmarked photos to train artificial intelligence (AI) models, aiming to improve performance by expanding and diversifying the training data without additional burden on experts. Methods/Study Population: Common to many medical imaging projects, we have a small number of expert-marked patient photos (N = 36, n = 360), and many unmarked photos (N = 337, n = 25,842). Dark skin (Fitzpatrick type 4+) is underrepresented in both sets; 11% of patients in the marked set and 9% in the unmarked set. In addition, a set of 20 expert-marked photos from 20 patients were withheld from training to assess model performance, with 20% dark skin type. Our gold standard markings were manual contours around affected skin by a trained expert. Three AI training methods were tested. Our established baseline uses only the small number of marked photos (supervised method). The semi-supervised method uses a mix of marked and unmarked photos with human feedback. The self-supervised method uses only unmarked photos without any human feedback. Results/Anticipated Results: We evaluated performance by comparing predicted skin areas with expert markings. The error was given by the absolute difference between the percentage areas marked by the AI model and expert, where lower is better. Across all test patients, the median error was 19% (interquartile range 6 – 34) for the supervised method and 10% (5 – 23) for the semi-supervised method, which incorporated unmarked photos from 83 patients. On dark skin types, the median error was 36% (18 – 62) for supervised and 28% (14 – 52) for semi-supervised, compared to a median error on light skin of 18% (5 – 26) for supervised and 7% (4 – 17) for semi-supervised. Self-supervised, using all 337 unmarked patients, is expected to further improve performance and consistency due to increased data diversity. Full results will be presented at the meeting. Discussion/Significance of Impact: By automating skin assessment for cGVHD, AI could improve accuracy and consistency compared to manual methods. If translated to clinical use, this would ease clinical burden and scale to large patient cohorts. Future work will focus on ensuring equitable performance across all skin types, providing fair and accurate assessments for every patient.

Introduction: Corticosteroids (CS) are the standard first-line treatment for patients (pts) with chronic graft-versus-host disease (cGvHD), but pts can be unresponsive or become refractory to or dependent on CS (SR/D).Ruxolitinib (RUX), a JAK1/2 inhibitor, is approved for the treatment of pts ≥12 years of age with SR/D-cGVHD, based on the primary outcomes from the randomized, phase III REACH3 (NCT03112603) study, which demonstrated the superior efficacy of RUX vs best available treatment (BAT) in pts with SR/D-cGVHD. Here, we present the final, long-term efficacy and safety outcomes from REACH3. Methods: Pts ≥12 years of age with moderate or severe SR/D-cGVHD were randomized 1:1 to receive either RUX 10 mg twice daily (BID) or investigator-selected BAT and followed for 3 years, till discontinuation or death. The primary analysis was conducted at week 24 (Cycle 7 Day 1 [C7D1]) in randomized pts, then pts entered the extension period (C7-39) in which they continued treatment, switched from BAT to RUX (crossover cohort), or discontinued treatment and entered long-term survival follow-up (FU). Failure-free survival (FFS; key secondary endpoint), duration of response (DOR), overall survival (OS), non-relapse mortality (NRM), malignancy relapse (MR), and safety were analyzed at 3 years (week 156; final data cut off 15 Dec 2022). Overall response rate (ORR) and best overall response (BOR) during the crossover treatment period were analyzed for pts who switched from BAT to RUX on or after C7D1. Results: Of 329 pts randomized, 53 completed the treatment period and 276 discontinued, 115 entered survival FU (RUX: 73/165; BAT:42/164); 70 pts crossed over from BAT to RUX of which 16 completed the crossover treatment period and 24 entered survival FU. Median FFS was longer in the RUX vs BAT arms (38.4 vs 5.7 months; hazard ratio [HR]=0.361, 95% confidence interval [CI]: 0.268, 0.485) with 12-month FFS probabilities of 64.0% (95% CI: 56.1, 70.8) and 28.8% (95% CI: 21.8, 36.1), respectively. Median OS was not reached and there was no difference in risk of death between the arms (HR=0.851, 95% CI: 0.544, 1.331) [Table 1]. Median DOR was 6.4 months (95% CI: 4.9, 11.4) in the BAT arm but was not reached for the RUX arm; notably, the probability of maintaining DOR at 3 years was higher in RUX (59.6%; 95% CI: 50.4, 67.6) vs BAT (26.7%; 95% CI: 18.5, 35.5). NRM event rates were similar between the arms (RUX: 29/165; BAT 34/164), and MR events were similar and low in both arms (13/156 and 11/160, RUX and BAT respectively) up to 3 years. The ORR at week 24 after crossover from BAT to RUX was 50.0% (95% CI: 37.8, 62.2), including complete response (CR) in 4 (5.7%) pts and partial response (PR) in 31 (44.3%) pts. The BOR during crossover was 81.4% (95% CI: 70.3, 89.7), including CR of 7.1% and PR of 74.3%; disease progression only occurred in 1 pt. Almost all pts treated during the main treatment period (RUX: 100%; BAT: 93.7%) experienced ≥1 adverse event (AE) and, in general, AE rates were higher in the RUX vs BAT arms [Table 2], likely due to prolonged FU and exposure to RUX (median exposure to treatment of 52.9 weeks vs. 24.1 weeks for RUX and BAT, respectively).Anemia was the most common AE (RUX: 33.9%; BAT: 15.8%) and grade ≥3 AE for RUX (17.6% vs BAT 9.5%, respectively); grade ≥3 neutropenia, thrombocytopenia, alanine aminotransferase increase, and gamma-glutamyltransferase increase were also ≥5% higher for RUX than BAT. Anemia (24.2%; grade ≥3, 10.3%) and thrombocytopenia (3.8%) were the most common RUX and BAT treatment-related AEs, respectively. Infections, excluding tuberculosis, were the most common AE of special interest (77.6% vs. 68.4%; grade ≥3, 31.5% and 26.6%, for RUX and BAT, respectively). On-treatment deaths were mainly due to cGvHD (n/N, RUX: 10/18; BAT 6/12). Conclusions: After 3 years of treatment in REACH3, the longer FFS and higher DOR with RUX vs BAT indicated that cGVHD was more in control with RUX treatment. Notably, efficacy was demonstrated in pts who switched from BAT to RUX with an ORR similar to that in pts randomized to RUX. RUX was well tolerated with no unexpected toxicities and safety that was consistent with earlier studies. Overall, the long-term control of cGVHD and tolerability of RUX was demonstrated for pts ≥12 years of age with SR/D-cGvHD.

Abstract Standard initial therapy of chronic graft vs. host disease (cGVHD) with glucocorticoids results in suboptimal response. Safety and feasibility of therapy with ofatumumab (1000 mg IV on days 0 and 14) and prednisone (1 mg/kg/day) was previously established in our phase I trial (n = 12). We now report the mature results of the phase II expansion of the trial (n = 38). The overall NIH severity of cGVHD was moderate (63%) or severe (37%) with 74% of all patients affected by the overlap subtype of cGVHD and 82% by prior acute cGVHD. The observed 6 month clinician-reported and 2014 NIH-defined overall response rates (ORR = complete + partial response [CR/PR]) of 62.5% (1-sided lower 90% confidence interval=51.5%) were not superior to pre-specified historic benchmark of 60%. Post-hoc comparison of 6 month NIH response suggested benefit compared to more contemporaneous NIH-based benchmark of 48.6% with frontline sirolimus/prednisone (CTN 0801 trial). Baseline cGVHD features (organ involvement, severity, initial immune suppression agents) were not significantly associated with 6-month ORR. The median time to initiation of second-line therapy was 5.4 months (range 0.9-15.1 months). Failure-free survival (FFS) was 64.2% (95% CI 46.5-77.4%) at 6 months and 53.1% (95% CI 35.8-67.7%) at 12 months, whereas FFS with CR/PR at 12 months of 33.5% exceeded a benchmark of 15% in post-hoc analysis, and was associated with greater success in steroid discontinuation by 24 months (odds ratio 8 (95% CI 1.21-52.7). This single-arm phase II trial demonstrated acceptable safety and potential efficacy of the upfront use of ofatumumab in combination with prednisone in cGVHD. This trial was registered at www.clinicaltrials.gov as #NCT01680965.

Abstract Introduction: Amphiregulin (AREG) is an epidermal growth factor receptor (EGFR) ligand that can restore integrity to damaged intestinal mucosa in murine models of acute graft-versus-host disease (aGVHD). We previously described AREG as a circulating biomarker of late-onset aGVHD, but its relevance combined with clinical risk factors has not yet been tested in a large cohort of patients with aGVHD occurring prior to day 100 post-transplant. We therefore tested samples from two aGVHD first-line treatment trials, Blood and Marrow Transplant Clinical Trials Network (BMT CTN) 0302 and 0802, and identified a clinically relevant threshold level of circulating AREG at aGVHD onset. We then investigated whether incorporating AREG into the refined Minnesota Risk Score could further risk-stratify patients. Patients and Methods: Blood samples were obtained at the onset of systemic aGVHD treatment within BMT CTN 0302 (serum) and BMT CTN 0802 (plasma). All patients with response data and samples for analysis from both trials were included (N=251). We determined the association of AREG with clinical outcomes, including risk stratification by the refined Minnesota criteria, day 28 complete/partial response (CR/PR) to first-line therapy, 2-year overall survival (OS) and 6-month and 2-year non-relapse mortality (NRM). We investigated the effect of AREG on clinical endpoints per a doubling in the value of AREG, defined a clinically relevant threshold level in the AREG values using two-fold cross-validation, and confirmed the clinical relevance of this cut-point in independent samples (N=92) from pooled from drawn from the Chronic GVHD Consortium and the Mount Sinai Acute GVHD International Consortium (MAGIC). Results: In patients enrolled in BMT CTN 0302/0802, AREG levels were 1.7-fold higher in patients with Minnesota high-risk (HR) compared to standard-risk (SR) aGVHD (HR median 53.4 vs SR 31 pg/ml, p&lt;0.01). Every 2-fold increase in AREG was associated with a 33% decrease in the likelihood of day 28 CR/PR (odds ratio [OR] 0.67, p&lt;0.01, table). Clinical factors alone, as determined by the refined Minnesota Risk Score, were associated with day 28 CR/PR (p=0.02, table). Adding AREG to the Minnesota Risk Score could further risk-stratify patients. Minnesota SR patients with elevated AREG ≥ 33 pg/mL showed a 59% lower odds of day 28 CR/PR than SR patients with low AREG (OR 0.41, p=0.02). Patients with Minnesota HR aGVHD with AREG ≥ 33 pg/mL had the worst outcomes, with an 82% lower odds of day 28 CR/PR in comparison to HR with low AREG (OR 0.18, p&lt;0.01). High AREG was associated with worse OS and NRM in both Minnesota SR (hazard ratios 2-year OS 2.3, p&lt;0.01 and 6 month NRM 2.95, p=0.01, respectively) and HR patients (hazard ratios 2-year OS 3.35, p&lt;0.01 and 6-month NRM 9.38, p&lt;0.01 respectively, figure). Finally, in independent samples from the Chronic GVHD Consortium/MAGIC we confirmed that high AREG was associated with worse day 28 CR/PR (55.2% vs. 79.4%, p=0.02) and significantly worse 6-month survival after the onset of aGVHD (57.1% vs. 82.5%, p=0.01). Conclusion: AREG is elevated in patients with poor aGVHD outcomes and adds to the accuracy of risk stratification when combined with the refined Minnesota Risk Score. AREG ≥33 pg/mL at aGVHD onset is associated with lower day 28 CR/PR and higher mortality in samples from 4 multicenter cohorts. The mechanism of elevated circulating AREG in severe aGVHD is not yet known, although we hypothesize the degree of AREG elevation reflects the intensity of immune-mediated tissue injury resulting in AREG release. With accumulating evidence of altered EGFR ligands in aGVHD, further investigation into epithelial repair pathways involving AREG may lead to new adjunctive therapies to overcome poor steroid response in high-risk aGVHD. Disclosures Holtan: Incyte: Other: One-time advisory board member. Khera: Novartis: Consultancy. Lee: Mallinckrodt: Honoraria; Amgen: Other: One-time advisory board member; Bristol-Myers-Squibb: Other: One-time advisory board member; Kadmon: Other: One-time advisory board member. Chen: Immudex: Research Funding. Arora: Takeda Oncology: Consultancy. Flowers: Pharmacyclics: Consultancy. Cutler: Pfizer: Consultancy; Kite: Consultancy; Incyte: Consultancy; Bristol-Myers Squibb: Consultancy; Pharmacyclics: Consultancy; Astellas: Consultancy. Jagasia: Janssen: Consultancy, Research Funding; Therakos: Consultancy, Research Funding; Mallinckrodt: Consultancy. Hexner: Novartis: Membership on an entity's Board of Directors or advisory committees, Research Funding. Levine: Novartis Pharmaceuticals Corporation: Membership on an entity's Board of Directors or advisory committees. MacMillan: Magenta Therapeutics: Research Funding.

Abstract Attendance at professional conferences is an important component of career development, because conferences are a major forum for presenting new research, interacting with colleagues and networking. An extensive literature documents differences in the professional experiences of women and men, including experiences at professional conferences. We hypothesized that women are less likely than men to ask questions at conferences, thus forgoing opportunities for professional development. To address this issue, we analyzed the question-asking behavior of women and men at the 2019 Annual Meeting and Exposition of the American Society of Hematology. In all, 112 sessions (55% of those eligible) were randomly chosen for coding, yielding data on 577 presentations. Although approximately 50% of moderators and speakers were women, the proportion of questions asked by women was significantly lower compared with the estimated proportion of women attending the conference (23% vs 39%; P &lt; .0001). Women were more likely to ask questions if another woman asked the first question or if the session topic was red cells. These results suggest that although women are represented equally as moderators and speakers, they are less likely to engage in the postpresentation discourse by asking questions. Encouraging women to speak up in professional situations and providing training on question-asking skills can help address this gender gap that potentially contributes to disparities in professional visibility and career advancement for women in hematology.

BackgroundThe association of community factors and outcomes after hematopoietic cell transplantation (HCT) has not been comprehensively described. Using the County Health Rankings and Roadmaps (CHRR) and the Center for International Blood and Marrow Transplant Research (CIBMTR), this study evaluated the impact of community health status on allogeneic HCT outcomes.MethodsThis study included 18,544 adult allogeneic HCT recipients reported to the CIBMTR by 170 US centers in 2014‐2016. Sociodemographic, environmental, and community indicators were derived from the CHRR, an aggregate community risk score was created, and scores were assigned to each patient (patient community risk score [PCS]) and transplant center (center community risk score [CCS]). Higher scores indicated less healthy communities. The impact of PCS and CCS on patient outcomes after allogeneic HCT was studied.ResultsThe median age was 55 years (range, 18‐83 years). The median PCS was –0.21 (range, –1.37 to 2.10; standard deviation [SD], 0.42), and the median CCS was –0.13 (range, –1.04 to 0.96; SD, 0.40). In multivariable analyses, a higher PCS was associated with inferior survival (hazard ratio [HR] per 1 SD increase, 1.04; 99% CI, 1.00‐1.08; P = .0089). Among hematologic malignancies, a tendency toward inferior survival was observed with a higher PCS (HR, 1.04; 99% CI, 1.00‐1.08; P = .0102); a higher PCS was associated with higher nonrelapse mortality (NRM; HR, 1.08; 99% CI, 1.02‐1.15; P = .0004). CCS was not significantly associated with survival, relapse, or NRM.ConclusionsPatients residing in counties with a worse community health status have inferior survival as a result of an increased risk of NRM after allogeneic HCT. There was no association between the community health status of the transplant center location and allogeneic HCT outcomes.