13  New Horizons in Transplantation and Cellular Therapies in Asia and Beyond

13.1 Session Overview

Session Details
Session New Horizons in Transplantation and Cellular Therapies in Asia and Beyond
Speakers Robert Negrin, MD; William Hwang, MBBS
Affiliations Stanford University, Stanford, CA, USA; National Cancer Centre & Singapore General Hospital, Singapore
Time Day 2, 3:45–4:30 p.m.

The session is framed around three practical questions in contemporary allogeneic transplantation and cellular therapy: how to reduce post-transplant relapse, how to choose induction chemotherapy in the transplant-eligible fit patient, and how to make CAR T-cell therapy more effective and more accessible [slide p.3]. Each topic is anchored in a case vignette of a fit adult with newly diagnosed AML and expanded into data from recent trials—the EBMT HaploMUD randomized study, the PARADIGM phase 2 study of azacitidine-venetoclax induction, and emerging CAR T-cell platforms including Orca-T plus donor-derived CD19/22 CAR T cells, KLN-1010 in vivo BCMA CAR T, CARMEN CLL-1 CAR T in AML, and obecabtagene autoleucel in severe refractory SLE [slide p.4, p.6, p.15, p.20, p.31, p.42].

13.2 Speaker Spotlight

Robert Negrin, MD is a Professor of Medicine at Stanford University and a past president of the American Society for Transplantation and Cellular Therapy (ASTCT). His research program focuses on the biology of graft-versus-host disease (GVHD), with pioneering work on regulatory T cell (Treg) therapy for GVHD prevention. He has been instrumental in translating immunologic insights into clinical GVHD prophylaxis strategies.

William Hwang, MBBS is a senior consultant hematologist at the National Cancer Centre Singapore and Singapore General Hospital. He has been a driving force in developing transplant and cellular therapy infrastructure across Southeast Asia, including the expansion of unrelated donor registries in the region and the introduction of CAR-T cell therapy to Singapore. His work has focused on improving transplant access and outcomes in ethnically diverse Asian populations with limited matched donor availability.

13.3 What’s New in 2025–2026

The session is built around a prototypical case: a 46-year-old man with AML in CR1, normal cytogenetics but several ELN Tier 1 mutations, MRD-positive after induction, fit with no comorbidities, an only child with no HLA-matched sibling [slide p.5]. From this starting point, three questions structure the talk: Is he a transplant candidate? What is the best donor? And what is the best bridge and post-transplant consolidation strategy? [slide p.3, p.5].

13.3.1 Topic 1 — Reducing Relapse After Transplant: HaploMUD

Relapse remains the leading cause of treatment failure after allogeneic transplantation, and the choice between a matched unrelated donor (MUD) and a haploidentical related donor is a central practical decision [slide p.4].

  • The EBMT-labelled HaploMUD trial (Kröger et al.) was a randomized prospective European multicenter study comparing 10/10 MUD versus haploidentical donor transplantation in patients with acute leukemia (AML in CR1 ELN intermediate-II/high risk or CR2, ALL in CR1/CR2, and high-risk MDS) aged 18–70 with ECOG ≤2 [slide p.6, p.7]. A key design feature: GVHD prophylaxis was identical in both arms (post-transplant cyclophosphamide 50 mg/kg days +3/+4, tacrolimus through day +120, and MMF through day +35), with PBSC as the stem cell source [slide p.8].
  • 171 patients were enrolled and 158 randomized (78 MUD, 80 haplo); a small number crossed over due to donor unavailability, and per-protocol populations were 68 MUD vs. 72 haplo [slide p.9].
  • Outcomes were essentially superimposable across arms. EFS at 2 years was 59.7% (MUD ITT) vs. 63.7% (Haplo ITT), p=0.9; per-protocol 64.7% vs. 64.5% [slide p.10]. OS at 2 years was 67.8% vs. 77.9% (ITT, p=0.6) and 73.5% vs. 75.1% (per-protocol, p=0.8) [slide p.11]. GVHD/relapse-free survival (GRFS) curves overlapped [slide p.12].
  • The study was stopped for futility with respect to detecting a ≥10% reduction in relapse between donor sources. The authors concluded that haploidentical and MUD transplants with identical PTCy-based prophylaxis yield similar outcomes, a finding of particular relevance for regions where unrelated donors are “generally not available or very expensive to pursue” [slide p.13].
Donor Choice Is No Longer the Limiting Step

With identical PTCy-based prophylaxis, haploidentical and matched unrelated donors produced overlapping EFS, OS, and GRFS in the HaploMUD trial [slide p.10–p.13]. For Asia-Pacific programs where registry access is constrained by ethnic under-representation, geographic distance, and cost, a haploidentical family donor is a fully legitimate first-choice option rather than a fallback [editorial].

13.3.2 Topic 2 — Induction Chemotherapy in the Transplant-Eligible Fit Patient: PARADIGM

The second topic reframes the case as a 45-year-old fit man with AML, normal cytogenetics, several Tier 1 mutations, and an HLA-matched younger sister, asking: what is the preferred induction if the goal is to reach allogeneic HCT in CR1—7+3 idarubicin/cytarabine, venetoclax plus azacitidine, or FLAG-Ida? [slide p.14].

  • PARADIGM (Fathi et al.) is a phase 2 randomized multicenter study comparing azacitidine + venetoclax (Aza-Ven) with conventional intensive chemotherapy (7+3 or CPX-351) for newly diagnosed fit adults with AML [slide p.15, p.16]. Key exclusions included PML-RARA, CBF alterations, FLT3 ITD/TKD (≥5% VAF), NPM1 mutations if aged <60, BCR::ABL1 fusion, and mixed phenotype [slide p.16]. HCT was allowed at any time following disease response.
  • Patients on the Aza-Ven arm were significantly more likely to reach HCT: 60.8% (52/86) vs. 40.0% (34/86) on intensive chemo, p=0.009 [slide p.17].
  • Transplant itself exerted a significant protective effect on EFS, and after adjustment for HCT in univariable and multivariable models the Aza-Ven effect on EFS remained protective (HR 0.67, p=0.0302) [slide p.17]. The implication presented is that low-intensity azacitidine-venetoclax may be a legitimate bridge to allogeneic HCT in selected fit adults, and is not simply a less-effective alternative to 7+3 in this setting.

13.3.3 Topic 3 — Novel Approaches to CAR T-Cell Therapy

CAR T-cell therapy has transformed B-cell malignancies and myeloma, but all approved products are autologous—expensive, logistically challenging, and difficult to scale. The session surveys four strategies to expand reach: precision-engineered allogeneic grafts that deliver donor-derived CAR T cells together with an engineered transplant, in vivo CAR T generation, new targets in AML, and CD19 CAR T for autoimmune disease [slide p.18, p.19].

13.3.3.1 Orca-T plus donor-derived CD19/22 CAR T in high-risk B-ALL

Durable remission rates with autologous CD19 CAR T in adult B-ALL remain disappointing: in ZUMA-3 (brexu-cel) and FELIX (obe-cel), ~42–44% of patients had received HCT before CAR T and only ~18% proceeded to HCT after CAR T, and duration of remission curves plateau well below the level seen after consolidative transplant [slide p.21]. Stanford’s strategy combines the precision-engineered Orca-T graft (CD34+ HSPCs plus high-purity regulatory T cells, with conventional T cells infused on day +2) with a donor-derived bispecific CD19/22 CAR T cell product (“Allo-CAR-Transplant”, ACT) [slide p.22, p.24].

  • In the primary Orca-T program, the ITT primary endpoint was survival free of moderate-to-severe chronic GVHD: 78% with Orca-T vs. 38% with Tac/MTX at ~12 months post-HCT (HR 0.26, p<0.00001) [slide p.23].
  • In the phase 1 ACT trial (Orca-T + donor CD19/22 CAR T in high-risk adult B-ALL), 18 patients were enrolled, 17 successfully manufactured, and 16 received the combined product [slide p.25]. The high-risk population included Ph+ (35%), Ph-like (18%), KMT2A/MLL rearrangement (12%), TP53 (12%), complex karyotype (12%), MRD+ post-induction (76%), and relapsed/refractory disease (47%); 76% had received prior blinatumomab [slide p.25].
  • Conditioning was mostly fractionated TBI/Cy (88%); donors were 76% matched sibling and 24% MUD; CAR T was dosed at 1, 2, or 3 × 106/kg [slide p.24, p.25].
  • Safety and efficacy were remarkable: median neutrophil engraftment day 13, no primary graft failure, acute GVHD grade 1 in only 1/16 (6%), chronic GVHD mild/moderate in 3/16, CRS grade 1 in 81% and grade 2 in 19%, ICANS grade 1 in 1/16, and VOD in 1/16 [slide p.26]. Day 42 response was 88% CR MRD-negative and 12% CR MRD-positive [slide p.26].
  • With median follow-up of 571 days (range 312–1,278), OS and disease-free survival curves remained at 100% at the most recent landmark [slide p.26]. All patients achieved MRD negativity by flow cytometry, BCR-ABL PCR, and clonoSEQ sequencing [slide p.27]. CAR T cells expanded and persisted without prolonged cytopenia, and persistence was longer with donor-derived (allogeneic) than with autologous CD19/22 CAR T in matched comparisons from the Stanford program [slide p.28, p.29].

13.3.3.2 KLN-1010 — in vivo BCMA CAR T for relapsed/refractory myeloma (inMMyCAR, LBA-1)

KLN-1010 is a first-in-class approach that generates anti-BCMA CAR T cells in vivo by direct infusion of an envelope-modified, replication-incompetent, self-inactivating lentiviral vector [slide p.31, p.32, p.33]. A de-targeted VSV-G fusogen avoids LDL-expressing cells and liver “drug sinks,” and a CD3 single-chain variable fragment re-targets the vector to T cells, which are then transduced to express a fully-human 4-1BB/CD3ζ anti-BCMA CAR [slide p.33]. Potential advantages over ex vivo CAR T include elimination of preconditioning lymphodepletion, greatly simplified logistics, no ex vivo culture (potentially better T-cell fitness), and reduced cost of goods [slide p.32].

  • inMMyCAR is a first-in-human phase 1 dose-escalation study (NCT07075185) in RRMM after ≥3 lines of therapy (prior PI, IMiD, and CD38 mAb) at Australian sites (Peter MacCallum, The Alfred, Royal Prince Alfred), with dose levels of 6×106, 2×107, and higher IU/kg [slide p.34]. Endpoints include safety/RP2D, CAR T expansion and persistence, and IMWG response, MRD, DOR, and PFS.
  • Initial 4 patients: heavily pretreated (3–5 prior lines), high-risk cytogenetics (del17p, t(4;14), del1p32), several with extramedullary disease [slide p.35]. All CRS events were grade 1–2 with median onset day 10 and median duration 5.5 days; no ICANS and no delayed neurotoxicity were observed [slide p.35]. Dexamethasone and tocilizumab were used as supportive care.
  • In vivo generation reached Cmax values of vector copies/µg DNA (51,647–108,730) that were commensurate with approved ex vivo products such as ciltacabtagene autoleucel (CARTITUDE-1 median 47,806) [slide p.36]. Persistence was documented in both blood and bone marrow.
  • Deep, ongoing MRD-negative responses were observed in all first 4 patients, with CR/VGPR/PR reached at month 2–3 and MRD-negativity at 10-5 or 10-6; involved serum free light chains normalized by month 1 and soluble BCMA fell in parallel [slide p.37]. This is the first clinical demonstration that CAR T cells can be generated directly in a patient from an infused lentiviral particle, with clinical activity comparable to ex vivo products.

13.3.3.3 CLL-1 CAR T for relapsed/refractory AML (CARMEN)

CAR T for AML has been held back by the lack of targets that are expressed on AML blasts and leukemic stem cells but spared by normal hematopoietic stem cells [slide p.38]. CLL-1 (CD371) is highly expressed on AML blasts and leukemic stem cells while absent on normal HSCs, making it a promising bridge-to-transplant antigen [slide p.39].

  • An earlier phase 1 study in 38 r/r AML patients reported an ORR of 73.7%, MRD-CR 42%, and median OS 12.2 months [slide p.39].
  • The CARMEN phase 1 trial (NCT04219163) uses an autologous CLL-1 CAR T with a CD28 costimulatory domain and gammaretroviral transduction (7–9 day manufacturing), in transplant-eligible heavily pretreated patients with >30% CLL-1 expression on blasts; dose escalation to 1×108 cells/m2 with fludarabine/cyclophosphamide lymphodepletion [slide p.40].
  • In the first 14 patients (median age 46, median 3 prior therapies, 6 prior allo-HSCT): CRS in 86% (grade 3 in 2), ICANS in 14% (max grade 3), and IEC-HS (hyperinflammatory syndrome) in 2 patients on DL1; grade ≥3 infections included bacteremia, viremia, and fungemia [slide p.41].
  • Responses in 14 patients: 2 CRi, 1 MLFS, 1 PR, 6 stable disease, 4 progression; 3 patients proceeded to allo-HSCT and 1 remains alive in remission >2 years post-transplant. Take-home: CLL-1 CAR T induced deep remissions in heavily pretreated r/r AML, consolidation with allo-HSCT was feasible, and dose expansion is planned [slide p.42].

13.3.3.4 Obecabtagene autoleucel (obe-cel) in severe refractory SLE — CARLYSLE

Building on the Erlangen experience, the CARLYSLE phase 1 study (Roddie et al.) is evaluating obe-cel—a fast off-rate CD19 CAR T originally developed for B-ALL (FELIX)—in severe, refractory SLE [slide p.43]. Eligible patients were adults 18–65 or adolescents 12–<18 with 2019 EULAR/ACR SLE, SLEDAI-2K ≥8 and ≥1 major organ involvement, refractory to hydroxychloroquine + corticosteroids plus ≥2 of immunosuppressants, B-cell agents, or cytokine inhibitors; biopsy-confirmed active lupus nephritis was required for patients with renal involvement [slide p.44].

  • Dose levels were 50×106 and 100×106 CAR T cells, with fludarabine/cyclophosphamide lymphodepletion; 7 adults recruited at 50M (6 infused) and 3 adults + 3 adolescents at 100M (5 infused) [slide p.45]. Baseline characteristics reflected severe refractory disease: median 10 years from SLE diagnosis, median SLEDAI-2K 17–18, 100% lupus nephritis in the 50M cohort, and prior exposure to B-cell therapies and multiple immunosuppressants [slide p.46].
  • Obe-cel was well tolerated with no ICANS and no grade ≥2 CRS in any patient; CRS occurred in 50% of the 50M cohort and 100% of the 100M cohort, all grade 1 [slide p.47]. One case of liver injury (possibly CAR- or prophylaxis-related) was considered a DLT and fully resolved.
  • Five of 6 (83%) infused adults in the 50M cohort achieved DORIS remission with a median onset of 5.1 months; 3/6 (50%) achieved a complete renal response and 1 achieved a partial renal response [slide p.48]. SLEDAI-2K fell substantially over time, anti-dsDNA antibodies decreased, and complement C3 recovered by month 1 in most patients.

13.3.4 Bringing It Together

Across three topics, the arc is that (i) donor selection has been democratized by PTCy-based prophylaxis so that haploidentical transplant is a legitimate first choice—of particular relevance outside well-resourced registries [slide p.13]; (ii) low-intensity azacitidine-venetoclax induction can increase the fraction of fit patients who actually reach HCT while remaining EFS-protective after adjustment for transplant [slide p.17]; and (iii) novel cellular therapies—precision-engineered allogeneic grafts with donor CAR T, in vivo CAR T generation, CLL-1 CAR T for AML, and CD19 CAR T for SLE—are beginning to address the twin bottlenecks of efficacy and access that have constrained the first generation of autologous products [slide p.50].

13.4 Clinical Pearls

Five Key Takeaways
  1. Multiple donor sources yield similar outcomes after allogeneic HCT. With identical PTCy-based prophylaxis, the EBMT HaploMUD randomized trial showed no clinically meaningful difference in EFS, OS, or GRFS between 10/10 MUD and haploidentical donors—the study was stopped for futility in detecting a ≥10% relapse reduction [slide p.10–p.13].
  2. Azacitidine–venetoclax can be a legitimate bridge to transplant in fit adults. In the PARADIGM phase 2 trial, Aza-Ven induction produced a significantly higher HCT rate than intensive chemotherapy (60.8% vs. 40.0%, p=0.009) and remained protective for EFS (HR 0.67, p=0.0302) after adjustment for transplant [slide p.17].
  3. Orca-T plus donor-derived CD19/22 CAR T is active and remarkably safe in high-risk adult B-ALL. 16 patients treated with the combined precision-engineered graft and donor CAR T reached 88% CR MRD-negative at day 42 with minimal acute GVHD, mostly grade 1–2 CRS, and 100% OS/DFS at a median follow-up of 571 days [slide p.25, p.26].
  4. In vivo CAR T generation is clinically feasible. KLN-1010, a T-cell-targeted lentiviral particle, produced deep, ongoing MRD-negative responses in all first 4 RRMM patients with CAR expansion Cmax comparable to approved ex vivo products, no preconditioning lymphodepletion, and only grade 1–2 CRS [slide p.35–p.37].
  5. CD19 CAR T and CLL-1 CAR T extend cellular therapy to new indications. Obe-cel produced DORIS remission in 5/6 severe refractory SLE patients with no ICANS or grade ≥2 CRS (CARLYSLE) [slide p.47, p.48], while CARMEN showed CLL-1 CAR T can induce deep remissions and bridge r/r AML patients to curative allo-HSCT [slide p.42].
Clinical Pearl: Choosing the Right Donor in 2026

With PTCy-based prophylaxis, the traditional donor hierarchy (matched sibling > matched unrelated > haploidentical) is being flattened. A readily available haploidentical family member with PTCy may be preferable to a matched unrelated donor with conventional prophylaxis, particularly when time-to-transplant is critical. Consider haploidentical donors early in the search process rather than as a last resort.

Asia-Pacific Perspective

The Asia-Pacific region faces unique challenges in transplant and cellular therapy: limited matched unrelated donor registries (particularly for underrepresented ethnic groups), geographic distances to certified treatment centers, and cost barriers that place CAR-T therapy out of reach for many patients. The convergence of PTCy-based haploidentical transplant and future allogeneic CAR-T products offers the greatest potential to bridge the access gap. Singapore, Korea, Japan, and India are leading the regional expansion of both transplant and CAR-T programs.

13.5 Key References

  1. Kröger N, Sanz J, Stelljes M, et al. Matched unrelated vs. haploidentical donor for allogeneic stem cell transplantation in patients with acute leukemia with identical GVHD prophylaxis — a randomized prospective European trial (HaploMUD). Presented at ASH 2025 [slide p.6, p.13].
  2. Fathi AT, Perl AE, Fell GG, et al. Results from PARADIGM — a phase 2 randomized multi-center study comparing azacitidine and venetoclax to conventional induction chemotherapy for newly diagnosed fit adults with AML. Presented at ASH 2025 [slide p.15, p.17].
  3. Meyer EH, et al. Orca-T engineered allogeneic transplant with low GVHD. Blood. 2025 [slide p.22].
  4. Spiegel JY, Patel S, Muffly L, et al. CAR T cells with dual targeting of CD19 and CD22 in adult patients with recurrent or refractory B cell malignancies: a phase 1 trial. Nat Med. 2021;27(8):1419–1431 [slide p.22].
  5. Shah BD, Ghobadi A, Oluwole OO, et al. KTE-X19 for relapsed or refractory adult B-cell acute lymphoblastic leukaemia: phase 2 results of the single-arm, open-label, multicentre ZUMA-3 study. Lancet. 2021;398(10299):491–502 [slide p.21].
  6. Roddie C, Sandhu KS, Tholouli E, et al. Obecabtagene autoleucel in adults with B-cell acute lymphoblastic leukemia (FELIX). N Engl J Med. 2024 [slide p.21].
  7. Molina A, et al. Superior efficacy and persistence of Orca-T + donor-derived CD19/22 CAR T versus autologous CAR19/22 in high-risk adult B-ALL. Presented at ASH 2025 [slide p.20, p.29].
  8. Harrison SJ, Ho PJ, Lim S-L, et al. MRD-negative outcomes following a novel in vivo gene therapy generating anti-BCMA CAR T cells in patients with RRMM: preliminary results from inMMyCAR, the first-in-human phase 1 study of KLN-1010 (LBA-1). Presented at ASH 2025 [slide p.31, p.37].
  9. Zhang H, et al. CLL-1 CAR T-cell therapy in relapsed/refractory AML. J Hematol Oncol. 2025 [slide p.39].
  10. Roddie C, Leandro M, Parker B, et al. Obecabtagene autoleucel (obe-cel) in patients with severe, refractory systemic lupus erythematosus: initial results from the phase 1 CARLYSLE study. Presented at ASH 2025 [slide p.43, p.48].
  11. Mackensen A, Müller F, Mougiakakos D, et al. Anti-CD19 CAR T cell therapy for refractory systemic lupus erythematosus. Nat Med. 2022;28(10):2124–2132 [editorial background for SLE CAR T].