CAR-T Therapy can be a Useful Treatment Modality for more than Just Hematologic Malignancies.

Chimeric antigen receptor-T-cell (CAR-T) therapy for hematologic malignancies has made significant advancements over the years, and it is now incorporated as a treatment algorithm. Early phase clinical trials are underway for various solid tumors, and the effectiveness of CAR-T cell therapy has been demonstrated for specific types of glioma and several solid tumors. However, its efficacy does not match that observed in hematological malignancies. Recently, a case series reported CAR-T therapy targeting CD19 for autoimmune diseases such as systemic lupus erythematosus, leading to a dramatic improvement in the clinical symptoms and the possibility of discontinuing immunosuppressive agents. Furthermore, CAR-T cell therapy is expected to be effective against various viruses and Aspergillus spp. Finally, attempts have been made to introduce CAR constructs into regulatory T cells to target their immunosuppressive effects. This article introduces the current progress in CAR-T cell therapy beyond the treatment of only hematologic malignancies and discusses future directions, considering the current medical situation in Japan.

Donor-derived cytomegalovirus-specific CD8+ T cells restricted to shared, donor-specific, or host-specific HLA after HLA mismatched hematopoietic stem cell transplantation.

Immune reconstitution after human leukocyte antigen (HLA)-mismatched (haploidentical) hematopoietic stem cell transplantation (haplo-HCT) can significantly influence long-term outcomes. The three possible HLA haplotypes after transplantation are: one carried by both the patient and the donor (shared HLA), one by donor only (donor-specific HLA), and one by patient only (host-specific HLA), and the donor T cells remain restricted to one of these three haplotypes. Understanding the presence of donor T cells restricted to each haplotype may provide more detailed insights into post-transplant immune response and potentially provide valuable information for the development of chimeric antigen receptor T cell or T cell receptor T cell constructs. In this study, patients or donors with HLA-A24 or HLA-A2 were tested with HLA-A*24:02- and A*02:01-restricted cytomegalovirus (CMV)-specific tetramers for detecting the respective HLA-restricted T cells. Sixty-four samples from 40 patients were assayed. More than half of the patients at day 90 and all patients by day 900 had shared HLA-restricted T cells. After day 90, half of the patients had donor-specific HLA-restricted T cells, but no host-specific HLA-restricted T cells were found. In the comparative analysis of the transplant types, shared HLA-restricted T cells were positive in all three categories: haplo-HCT (50%), 2-haplo-mis-HCT (75%), and spousal HCT (67%). Furthermore, donor-specific HLA-restricted T cells demonstrated positivity in haplo-HCT at 57% and in 2-haplo-mis-HCT at 60%, with a threshold of 0.01%. Donor-specific HLA-restricted T cells for spousal HCT were not examined due to the lack of an appropriate HLA combination for the tetramers. The presence of shared HLA-restricted T cells explains the host defense after HLA-haploidentical transplantation, while the presence of donor-specific HLA-restricted T cells may account for host defense against hematotropic viruses, such as CMV. However, this study failed to detect host-specific HLA-restricted T cells, leaving the host defense against epitheliotropic viruses unresolved, thus requiring further investigation.

Trends in allogeneic hematopoietic cell transplantation survival using population-based descriptive epidemiology method: analysis of national transplant registry data.

Prognosis for patients undergoing hematopoietic cell transplantation (HCT) has been improving. Short-term survival information, such as crude survival rates that consider deaths immediately after the transplantation, may not be sufficiently useful for assessing long-term survival. Using the data of the Japanese HCT registry, the net survival rate of patients who survived for a given period was determined according to age, disease, and type of transplant. We included a total of 41,716 patients who received their first allogeneic hematopoietic cell transplantation between 1991 and 2015. For each disease, age group, graft source subcategory, net survival was calculated using the Pohar-Perme method, and 5-year conditional net survival (CS) was calculated. Ten-year net survivals of total patient cohort were 41.5% and 47.4% for males and females, respectively. Except for myelodysplastic syndrome, multiple myeloma, and adult T-cell leukemia/lymphoma, 5-year CS for 5-year transplant survivors exceeded 90%. CS was especially high for aplastic anemia, of which was over 100% for children and younger adults receiving cord blood, suggesting that these patients have similar longevity to an equivalent group from the general population. These findings provide useful information for long-term survival, and can serve as benchmark for comparisons among registries, including other cancers.

An open-label study of belumosudil, a selective ROCK2 inhibitor, as second or subsequent line of therapy for steroid-dependent/steroid-resistant chronic GVHD.

Belumosudil mesylate is a selective Rho-associated coiled-coil kinase 2 inhibitor with immunomodulatory and antifibrosis effects. This multicenter, open-label, single-arm study evaluated belumosudil 200 mg once daily as second or subsequent line of therapy (LOT) in 21 Japanese patients ≥12 years of age with steroid-dependent/steroid-resistant chronic graft-versus-host disease (cGVHD). The primary endpoint of best overall response rate (ORR) at 24 weeks after enrollment of the last patient was 85.7% (95% confidence interval [CI]: 63.7-97.0), and the lower limit of the 95% CI exceeded the pre-defined threshold of 25%. The Kaplan-Meier estimate of duration of response rate at 24 weeks was 75% (95% CI: 46-90); 13/18 responders (72.2%) had a sustained response for ≥20 weeks. The median time to response was 4.1 weeks (range 3.90-8.10); ORR was 47.6% at 4 weeks and 75.0% at 24 weeks; best ORR was 80% for joints/fascia, 66.7% for the mouth, and 54.5% for skin. Overall, 57.1% of patients had clinically meaningful symptom improvement at least once; the median duration of symptom improvement was 22.2 weeks (range 4.0-51.3). Corticosteroid dose reductions were recorded for 57.1% of patients. Median failure-free and overall survival were not reached. Treatment-emergent adverse events occurred in 85.7% of patients (most commonly diarrhea, 19.0%), of which 38.1% were drug-related. There were no drug-related discontinuations or deaths. In summary, belumosudil 200 mg once daily as second or subsequent LOT in Japanese patients with steroid-dependent/steroid-resistant cGVHD was effective, with no new safety concerns.

Pretransplantation predictors of survival in nonremission acute myeloid leukemia treated with haploidentical transplantation using steroid-based GVHD prophylaxis.

Haploidentical hematopoietic cell transplantation (HCT) using glucocorticoids for acute graft-versus-host disease prophylaxis (GC-haplo) may become a curative treatment option for nonremission acute myeloid leukemia (AML). This retrospective study aimed to identify pre-HCT predictors of survival in a cohort of 97 nonremission AML treated with GC-haplo in Hyogo Medical University Hospital between 2010 and 2020. Relapse and primary induction failure included in 70 (72%) and 27 (28%) patients, respectively. Sixty-one patients (63%) had undergone previous HCT. Multivariate analysis revealed that ≤ 6 months' duration between first complete remission (CR1) and first relapse (Rel1) (CR1-Rel1 interval) (hazard ratio 2.11, 95% confidence interval [CI] 1.15-3.89, P = 0.016) and serum albumin before starting the conditioning treatment of ≤ 3.5 g/dL (hazard ratio 1.80, 95%CI 1.09-2.96, P = 0.022) as risk factors for overall survival. Among three groups categorized according to serum albumin and CR1-Rel1 interval, the best 3-year overall survival was observed in patients with albumin > 3.5 g/dL and CR1-Rel1 interval > 6 months or primary induction failure (50.2%, 95%CI 28.9%-68.3%, P < 0.001), revealing that survival could be predicted using albumin and past CR duration in patients with very high-risk AML not in remission before GC-haplo.

Individualized rabbit anti-thymocyte globulin dosing in adult haploidentical hematopoietic cell transplantation with high-risk hematologic malignancy: Exposure-response analysis and population pharmacokinetics simulations.

Hematopoietic cell transplantation (HCT) for hematologic malignancies with non-remission disease and/or prior post-transplant relapse have poor relapse-free survival. We previously demonstrated the efficacy of haploidentical reduced-intensity HCT regimen with glucocorticoid-based graft-versus-host disease (GVHD) prophylaxis. We recently showed a possible association between rabbit antithymocyte globulin (rATG) exposure and acute GVHD (aGVHD) risk, leading to hypothesize that optimization of rATG exposure may further improve this regimen. We retrospectively examined the exposure-response association of rATG and key clinical outcomes post haploidentical HCT. We subsequently developed an individualized rATG dosing that optimizes rATG exposure using a previously developed population pharmacokinetic model. Of the 103 patients analyzed, the median age was 47 years (range: 17-70) and majority had a non-remission disease prior to HCT (88%). rATG concentration on day 0 of HCT (Cday_0 ) was the strongest predictor of Grade 2-4 aGVHD through day +100. Patients with Cday_0 ≥ 20 µg/mL had an approximately 3-fold lower risk of Grade 2-4 aGVHD (hazard ratio [HR]: 0.32, 95% confidence interval [CI]: 0.16, 0.62) and Grade 3-4 aGVHD (HR: 0.33, 95% CI: 0.16, 0.68) as well as an approximately 2-fold lower risk of overall mortality (HR: 0.47, 95% CI: 0.28, 0.77) and relapse (HR: 0.50, 95% CI: 0.26, 0.94). In conclusion, this reduced-intensity haploidentical HCT regimen with exposure-optimized rATG may provide a promising option to patients undergoing high-risk HCT for hematologic malignancy. The developed rATG dosing warrant prospective validation.

HLA haploidentical stem cell transplantation from HLA homozygous donors to HLA heterozygous donors may have lower survival rates than haploidentical transplantation from HLA heterozygous donors to HLA heterozygous donors: a retrospective nationwide analysis.

In HLA haploidentical stem cell transplantation, patients and donors usually share one HLA haplotype and have one different HLA haplotype (hetero-to-hetero). However, there are rare cases of transplantation from HLA homozygous donors to heterozygous recipients (homo-to-hetero), resulting in mismatches only in the graft-versus-host direction. We previously reported that homo-to-hetero transplants have a lower survival rate in a mouse model than hetero-to-hetero transplants due to stronger graft-versus-host disease (GVHD) but inferior graft-versus-leukemia effect. To examine whether homo-to-hetero transplant effects also occur in humans, we retrospectively compared the results of 59 homo-to-hetero and 4,539 hetero-to-hetero cases in the Japanese transplant registry data. The results showed no statistical difference between the homo-to-hetero and hetero-to-hetero groups in the cumulative incidences of neutrophil engraftment (83.1% vs 89.0%), acute GVHD II-IV (36.8% vs 38.8%), III-IV (16.8% vs 17.4%), chronic GVHD (32.7% vs 30.7%), relapse (52.9% vs 49.0%), and non-relapse mortality (31.6% vs 28.2%). In contrast, overall survival was significantly lower in the homo-to-hetero group than in the hetero-to-hetero group (12.6% vs 26.2%, p = 0.0308). The inferior effect of homo-to-hetero transplantation on overall survival remained significant in multivariate analyses.

Outcomes of allogeneic hematopoietic cell transplantation under letermovir prophylaxis for cytomegalovirus infection.

Cytomegalovirus (CMV) infection is a major infectious complication following allogeneic hematopoietic cell transplantation (allo-HCT). Although letermovir (LMV) prophylaxis dramatically reduces the incidence of early clinically significant CMV (csCMV) infection, it remains unclear whether it has a beneficial effect on nonrelapse mortality (NRM) and overall survival (OS). Herein, we evaluated the impact of LMV prophylaxis on posttransplant outcomes using the registry database of the Japanese Society for Transplantation and Cellular Therapy. Adult patients who underwent allo-HCT between 2017 and 2019 were analyzed (n = 6004). LMV prophylaxis was administered to 1640 patients (LMV group) and it significantly reduced the incidence of csCMV infection compared with those not administered LMV prophylaxis (15.4% vs 54.1%; p < 0.01). However, it did not improve the 1-year NRM (hazard ratio [HR], 0.93; p = 0.40) and OS (HR, 0.96; p = 0.49). In the LMV group, 74 patients had breakthrough csCMV infection and showed inferior NRM (HR, 3.44; p < 0.01) and OS (HR, 1.93; p = 0.02) compared with those without infection. After completing LMV prophylaxis, 252 patients had late csCMV infection and showed inferior NRM (HR, 1.83; p < 0.01) and OS (HR, 1.58; p < 0.01). Our findings suggest that managing breakthrough and late csCMV infections is important for improving long-term outcomes.

Outcomes of allogeneic hematopoietic stem cell transplantation for relapsed or refractory diffuse large B-cell lymphoma.

Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is a currative treatment modality for diffuse large B-cell lymphoma (DLBCL) because of the intrinsic graft-versus-lymphoma effect. However, limited information is available regarding which patients with relapsed or refractory DLBCL are likely to benefit from allo-HSCT. We retrospectively analyzed data from 1268 DLBCL patients who received allo-HSCT. The overall survival and progression-free survival (PFS) rates were 30.3% and 21.6% at 3 years, respectively. Multivariate analysis revealed that stable or progressive disease at transplantation, male patient, poorer performance status at transplantation, and shorter intervals from previous transplantation were associated independently with a lower PFS. Four prognostic factors were used to construct a prognostic index for PFS, predicting 3-year PFS of 55.4%, 43.7%, 20.4% and 6.6%, respectively. The prognostic model predicted relapse rates following allo-HSCT accordingly (P < 0.0001), whereas did not predict transplantation-related mortality (P = 0.249). The prognostic index can identify a subgroup of DLBCL patients who benefit from allo-HSCT and it is worthwhile to evaluate whether this model is also applicable to patients undergoing allo-HSCT in cases of relapse after chimeric antigen receptor engineered T-cell therapy, although the application of allo-HSCT has been declining with the increase of novel immunotherapies.

Analysis of risk factors for fatal renal complications after allogeneic hematopoietic cell transplantation.

Various complications can influence hematopoietic cell transplantation (HCT) outcomes. Renal complications can occur during the early to late phases of HCT along with various factors. However, studies focusing on fatal renal complications (FRCs) are scarce. Herein, we analyzed 36,596 first allogeneic HCT recipients retrospectively. Overall, 782 patients died of FRCs at a median of 108 (range, 0-3,440) days after HCT. The cumulative incidence of FRCs was 1.7% and 2.2% at one and five years, respectively. FRCs were associated with older age, male sex, non-complete remission (non-CR), lower performance status (PS), and HCT comorbidity index (HCT-CI) associated with renal comorbidity in multivariate analysis. The risk factors within 100 days included older age, multiple myeloma, PS, and HCT-CI comorbidities (psychiatric disturbance, hepatic disease, obesity, and renal disease). Older age and male sex were risk factors between 100 days and one year. After one year, HCT-CI was associated with the presence of diabetes and prior solid tumor; total body irradiation was identified as a risk factor. Non-CR was a common risk factor in all three phases. Furthermore, acute and chronic graft-versus-host disease, reactivation of cytomegalovirus, and relapse of underlying disease also affected FRCs. Systematic follow-up may be necessary based on the patients' risk factors and post-HCT events.

Outcomes in human T-cell leukemia virus type I carriers after hematopoietic stem cell transplantation for diseases other than adult T cell leukemia/lymphoma: a Japanese national survey.

Background: Human T-cell leukemia virus type I (HTLV-1) is a retrovirus known to cause adult T-cell leukemia/lymphoma (ATL). There are few reports on hematopoietic stem cell transplantation (HSCT) for HTLV-1 carriers with diseases other than ATL. Methods: A total of 25,839 patients (24,399 adults and 1440 children) with pre-transplant HTLV-1 serostatus information recorded in the Japanese National Survey Database who had undergone their first HSCT were analyzed. We investigated the overall survival (OS), transplant-related mortality (TRM), and disease-related mortality (DRM) after HSCT in relation to HTLV-1 serologic status. Findings: Three hundred and forty-eight patients were HTLV-1 antibody carriers. The number of HTLV-1 carriers and noncarriers among adult patients who received allogeneic HSCT (allo-HSCT) or autologous HSCT (auto-HSCT) was 237/15,777 and 95/8920, respectively, and was 16/1424 among pediatric patients who received allo-HSCT. No pediatric HTLV-1 carrier recipients undergoing auto-HSCT were identified. There were no significant differences between HTLV-1 carriers and non-carriers regarding stem cell source, disease risk, or HCT-CI score prior to allo-HSCT. Multivariate analysis of OS (P = 0.020) and TRM (P = 0.017) in adult patients showed that HTLV-1 positive status was a significant prognostic factor. In children, TRM was significantly higher (P = 0.019), but OS was not significantly different. In adult patients who underwent auto-HSCT, HTLV-1 positive status was not a significant prognostic factor. In adult allo-HSCT patients, cytomegalovirus reactivation was significantly more common in HTLV-1 carriers (P = 0.001). Interpretation: HTLV-1 antibody positivity was shown to have a poor prognosis in OS and TRM after allo-HSCT in adult patients and in TRM after allo-HSCT in pediatric patients. Funding: This work was supported in part by the practical research programs of the Japan Agency for Medical Research and Development (AMED) under grant number 17ck0106342h0001.

Improved Long-Term Net Survival after Allogeneic Hematopoietic Cell Transplantation in Patients with Hematologic Malignancies over Two Decades.

Allogeneic (allo-) hematopoietic cell transplantation (HCT) has evolved as a curative therapy for hematologic malignancies and diseases, with practice changes over the past 2 decades. This study aimed to evaluate the change in 5-year net survival (NS) of allo-HCT recipients in a population-based cohort over the past 2 decades, which allows the estimation of a more HCT-specific long-term survival rate by considering background mortality changes. This study included 42,064 patients with hematologic malignancies who underwent their first allo-HCT in Japan between 2000 and 2018 and were reported to the Transplant Registry Unified Management Program. We compared the 5-year NS after allo-HCT in 4 consecutive HCT periods (2000 to 2004, 2005 to 2008, 2009 to 2012, and 2013 to 2018). The 5-year NS of the latest period was estimated using the period analysis method. Adjusted excess hazard ratios (EHRs) for 5-year NS over the HCT period were analyzed using an EHR model. In addition to the analysis of all hematologic malignancies, adjusted 5-year NS for each major hematologic malignancy, including acute myelogenous leukemia, acute lymphoblastic leukemia (ALL), myelodysplastic syndrome, adult T cell leukemia/lymphoma, chronic myeloid leukemia (CML), and malignant lymphoma, was analyzed. The probability of adjusted 5-year NS after HCT improved significantly over time: 35% in 2000 to 2004, 39% in 2005 to 2008, 45% in 2009 to 2012, and 49% in 2013 to 2018. The adjusted EHRs were .90 (95% confidence interval [CI], .86 to .93) in the 2005 to 2008 period, .77 (95% CI, .74 to .80) in the 2009 to 2012 period, and .65 (95% CI, .63 to .68) in the 2013 to 2018 period, with the 2000 to 2004 period as the reference. The 5-year NS improved among all hematologic malignancies, with a significant improvement in CML and ALL. The changes in 5-year NS from the 2000 to 2004 period to the 2013 to 2018 period ranged from 46% to 66% in CML and from 41% to 59% in ALL. In addition to the large improvement of 1-year NS, smaller but continued improvement in NS between 1 and 5 years after transplantation was observed. NS at 5 years conditional on being alive at 1 year increased from 64% in 2000 to 2004 to 73% in 2013 to 2018. Even after subtracting the background mortality in the general population, we found a significant improvement in long-term allo-HCT-specific survival rates for patients with hematologic malignancies over the past 2 decades in Japan.

Impact of GVHD on lymphoma progression: Nationwide study from Japanese Society for Transplantation and Cellular Therapy.

The graft-versus-lymphoma (GVL) effect and its association with acute and chronic GVHD (aGVHD, cGVHD) has not been comprehensively elucidated. We retrospectively analysed 2204 Japanese patients with non-Hodgkin lymphomas (NHLs; indolent B-NHLs, n = 689; aggressive B-NHLs, n = 720; mature T/NK-NHLs, n = 795) receiving a first allo-HSCT in 2003-2017. Pre-transplant lymphoma control showed complete response (CR) in 759 and non-CR in 1445. We assessed the impact of aGVHD/cGVHD on lymphoma progression and other outcomes. Although aGVHD/cGVHD showed no statistical impact on lymphoma progression in the overall cohort, their impact was clear in certain groups: Grade I-II aGVHD in CR patients (HR, 0.63; 95% CI, 0.43-0.91), especially in mature T/NK-NHL (HR, 0.46; 95% CI, 0.26-0.83) and extensive cGVHD in patients with mature aggressive B-NHLs (HR, 0.55; 95% CI, 0.31-0.97). In total, limited cGVHD was associated with superior survivals (progression-free survival: HR, 0.71; 95% CI, 0.56-0.90), whereas severe GVHDs showed negative impacts on them. Our results support the presence of GVL effects differentially associated with GVHD in different lymphoma subtypes/controls. Meanwhile, it was also suggested that we should manage GVHDs within a limited activity, considering the negative impact of severe GVHDs. As pre-transplant lymphoma control remains a strong factor influencing transplant outcomes, improving its management is an important issue to be addressed.

Population Pharmacokinetics of Total Rabbit Anti-thymocyte Globulin in Non-obese Adult Patients Undergoing Hematopoietic Cell Transplantation for Hematologic Malignancy.

BACKGROUND AND OBJECTIVES: Rabbit anti-thymocyte globulin (rATG), a therapeutic polyclonal antibody against human T cells, is commonly used in conditioning therapy prior to allogeneic hematopoietic cell transplantation (HCT). Previous studies successfully developed an individualized rATG dosing regimen based on "active" rATG population PK (popPK) analysis, while "total" rATG can be a more logistically favorable alternative for early HCT outcomes. We conducted a novel popPK analysis of total rATG. METHODS: Total rATG concentration was measured in adult human-leukocyte antigen (HLA) mismatched HCT patients who received a low-dose rATG regimen (total 2.5-3 mg/kg) within 3 days prior to HCT. PopPK modeling and simulation was performed using nonlinear mixed effect modeling approach. RESULTS: A total of 504 rATG concentrations were available from 105 non-obese patients with hematologic malignancy (median age 47 years) treated in Japan. The majority had acute leukemia or malignant lymphoma (94%). Total rATG PK was described by a two-compartment linear model. Influential covariate relations include ideal body weight [positively on both clearance (CL) and central volume of distribution], baseline serum albumin (negatively on CL), CD4+ T cell dose (positively on CL), and baseline serum IgG (positively on CL). Simulated covariate effects predicted that early total rATG exposures were affected by ideal body weight. CONCLUSIONS: This novel popPK model described the PK of total rATG in the adult HCT patients who received a low-dose rATG conditioning regimen. This model can be used for model-informed precision dosing in the settings with minimal baseline rATG targets (T cells), and early clinical outcomes are of interest.

Clinical characteristics of late-onset interstitial pneumonia after allogeneic hematopoietic stem cell transplantation.

Non-infectious pulmonary complications after allogeneic hematopoietic stem cell transplantation (HSCT) remain fatal. In particular, information regarding late-onset interstitial lung disease predominantly including organizing pneumonia and interstitial pneumonia (IP) is limited. A retrospective nationwide survey was conducted using data collected from the Japanese transplant outcome registry database from 2005 to 2010. This study focused on patients (n = 73) with IP diagnosed after day 90 post-HSCT. A total of 69 (94.5%) patients were treated with systemic steroids, and 34 (46.6%) experienced improvement. The presence of chronic graft-versus-host disease at the onset of IP was significantly associated with non-improvement of symptoms (odds ratio [OR] 0.35). At the time of last follow-up (median, 1471 days), 26 patients were alive. Of the 47 deaths, 32 (68%) were due to IP. The 3-year overall survival (OS) and non-relapse mortality (NRM) rates were 38.8% and 51.8%, respectively. In the multivariate analysis, the predictive factors for OS were comorbidities at IP onset (hazard ratio [HR]: 2.19) and performance status (PS) score of 2-4 (HR 2.77). Furthermore, cytomegalovirus reactivation requiring early intervention (HR 2.04), PS score of 2-4 (HR 2.63), and comorbidities at IP onset (HR 2.90) were also significantly associated with increased risk of NRM.

Risk factors and outcome of Stenotrophomonas maltophilia infection after allogeneic hematopoietic stem cell transplantation: JSTCT, Transplant Complications Working Group.

Stenotrophomonas maltophilia (S. maltophilia) is an aerobic nonfermenting Gram-negative bacillus widely distributed in the environment that has inherent multidrug resistance to beta-lactam and carbapenem antibiotics. S. maltophilia infection (SMI) is known as an important fatal complication following allogeneic hematopoietic stem cell transplantation (HSCT), but its clinical characteristics have not been well clarified. A retrospective study to identify the incidence, risk factors, and outcomes of SMI after allogeneic HSCT was performed using the database of the Japanese nationwide registry, including 29,052 patients who received allogeneic HSCT in Japan between January 2007 and December 2016. A total of 665 patients developed SMI (sepsis/septic shock, 432; pneumonia, 171; other, 62). The cumulative incidence of SMI at 100 days after HSCT was 2.2%. Among risk factors identified for SMI (age ≥ 50 years, male, performance status 2-4, cord blood transplantation [CBT], myeloablative conditioning, Hematopoietic Cell Transplant-Comorbidity Index [HCT-CI] score 1-2, HCT-CI score ≥ 3, and active infectious disease at HSCT), CBT was the strongest risk factor (hazard ratio, 2.89; 95%CI, 1.94-4.32; p < 0.001). The survival rate at day 30 after SMI was 45.7%, and SMI before neutrophil engraftment was significantly associated with poor survival (survival rate 30 days after SMI, 40.1% and 53.8% in patients with SMI before and after engraftment, respectively; p = 0.002). SMI is rare after allogeneic HSCT, but its prognosis is extremely poor. CBT was a strong risk factor for SMI, and its development prior to neutrophil engraftment was associated with poor survival.

A convolutional neural network-based model that predicts acute graft-versus-host disease after allogeneic hematopoietic stem cell transplantation.

BACKGROUND: Forecasting acute graft-versus-host disease (aGVHD) after allogeneic hematopoietic stem cell transplantation (HSCT) is highly challenging with conventional statistical techniques due to complex parameters and their interactions. The primary object of this study was to establish a convolutional neural network (CNN)-based prediction model for aGVHD. METHOD: We analyzed adult patients who underwent allogeneic HSCT between 2008 and 2018, using the Japanese nationwide registry database. The CNN algorithm, equipped with a natural language processing technique and an interpretable explanation algorithm, was applied to develop and validate prediction models. RESULTS: Here, we evaluate 18,763 patients between 16 and 80 years of age (median, 50 years). In total, grade II-IV and grade III-IV aGVHD is observed among 42.0% and 15.6%. The CNN-based model eventually allows us to calculate a prediction score of aGVHD for an individual case, which is validated to distinguish the high-risk group of aGVHD in the test cohort: cumulative incidence of grade III-IV aGVHD at Day 100 after HSCT is 28.8% for patients assigned to a high-risk group by the CNN model, compared to 8.4% among low-risk patients (hazard ratio, 4.02; 95% confidence interval, 2.70-5.97; p < 0.01), suggesting high generalizability. Furthermore, our CNN-based model succeeds in visualizing the learning process. Moreover, contributions of pre-transplant parameters other than HLA information to the risk of aGVHD are determined. CONCLUSIONS: Our results suggest that CNN-based prediction provides a faithful prediction model for aGVHD, and can serve as a valuable tool for decision-making in clinical practice.

Hematopoietic stem cell transplantation (HSCT) is a procedure used in patients to reestablish blood cell production. It involves the transplant of cells from a donor to the patient. In some patients the transplanted cells damage cells within the patients. This is called graft-versus-host disease (GVHD). We developed a computational code that can predict the likelihood a person will develop GVHD soon after HSCT. Using this computer program will enable doctors to better identify those at risk of GVHD and initiate treatments when required.

Prognostic impact of complex and/or monosomal karyotypes in post-transplant poor cytogenetic acute myeloid leukaemia: A quantitative approach.

To evaluate the prognostic impact of complex karyotype (CK) and/or monosomal karyotype (MK) in combination with various clinical factors on allogeneic stem cell transplantation (HSCT) outcomes of patients with acute myeloid leukaemia (AML), we analysed the registry database of adult AML patients who underwent allogeneic HSCT between 2000 and 2019 in Japan. Among 16 094 patients, those with poor cytogenetic risk (N = 3345) showed poor overall survival (OS) after HSCT (25.3% at 5 years). Multivariate analyses revealed that CK and/or MK (hazard ratio [HR], 1.31 for CK without MK; 1.27 for MK without CK; and 1.73 for both), age at HSCT ≥50 years (HR, 1.58), male sex (HR, 1.40), performance status ≥2 (HR, 1.89), HCT-CI score ≥3 (HR, 1.23), non-remission status at HSCT (HR, 2.49), and time from diagnosis to HSCT ≥3 months (HR, 1.24) independently reduced post-HSCT OS among patients with poor cytogenetic risk AML. A risk scoring system based on the multivariate analysis successfully stratified patients into five distinct groups for OS. This study confirms the negative effects of CK and MK on post-HSCT outcomes, and offers a powerful risk scoring system for predicting prognoses after HSCT among AML patients with unfavourable cytogenetics.