Clinical outcome of allogeneic stem cell transplantation in patients with b-cell lymphoid malignancies following treatment with targeted small molecule inhibitors

Abstract

We aimed to study the risks of graft-versus-host disease (GVHD), non-relapse mortality (NRM) and survival outcomes of allogeneic stem cell transplantation (alloSCT) in patients with chronic lymphocytic leukemia (n=17), Richter’s syndrome (n=14), or lymphoma (n=18) after small molecule inhibitors (SMIs). Patients had a median of 4 prior therapies, including ibrutinib (n=46; 94%), venetoclax (n=19; 39%), and idelalisib (n=6; 12%). Twenty-one (43%) had >1 SMI. P53 mutation was detected in 58% of patients. The 3-year overall and progression-free survival rates were 68% and 59%, respectively. The rates of grade II-IV and III-IV acute GVHD were 33% and 7%. The 1-year rates of chronic GVHD, NRM and relapse were 19%, 10% and 21%, respectively. Results were comparable to a historical control of patients who received alloSCT without a prior exposure to SMI. We conclude that a prior use of SMI does not impair the outcomes after alloSCT.

Introduction

The use of non-myeloablative allogeneic stem cell transplantation (alloSCT), which was adopted more than 20 years ago, has resulted in long-term remission in patients with chronic lymphocytic leukemia (CLL), Richter’s syndrome, mantle cell lymphoma (MCL), and follicular lymphoma (FL) that is refractory, has relapsed, or has progressed on conventional therapies [1–5]. The recent advent of small molecule inhibitors (SMIs) that target downstream B cell receptor signaling pathways, such as Bruton’s tyrosine kinase (BTK) inhibitors, phosphatidylinositol-3-kinase (PI3K) delta inhibitors, and BCL2-regulated intrinsic apoptotic pathway inhibitors, have significantly changed the treatment landscape of these diseases [6–10]. These molecules have shown robust and durable responses in patients with the above-mentioned lymphoid histologic types, leading to their incorporation in the frontline setting as well as in refractory or relapsed disease treatment strategies. However, studies have shown that a subset of patients with high-risk disease characteristics who have experienced disease progression on this pathway signaling inhibitors have a dismal prognosis [11–12]. Salvage treatment with cellular therapies such as alloSCT and chimeric antigen receptor T cell-based (CAR-T) therapy are under investigation [13].

Currently, there is a paucity of data on the clinical outcome of patients with CLL, Richter’s syndrome, MCL, and FL who have experienced disease progression or who have been treated with these SMIs prior to alloSCT. Therefore, in this study, we evaluated the safety and outcomes of alloSCT in these patients.

Patients and methods

Study design

This was a single-center retrospective study of patients with CLL, Richter’s syndrome, MCL, or FL who underwent alloSCT at The University of Texas MD Anderson Cancer Center (Houston, Texas) between January 2015 and December 2018 and who had either experienced disease progression prior to alloSCT or who had bridged to alloSCT with any of the novel SMIs (the BTK inhibitor ibrutinib, the PI3K delta inhibitor idelalisib, or the BCL2 inhibitor venetoclax).

We collected data on their baseline clinical characteristics, disease, treatment, and transplantation from the protected database in the Department of Stem Cell Transplantation and Cellular Therapy at MD Anderson. The institutional review board approved this study.

Preparative regimens, transplantation, and GVHD Prophylaxis

Eligibility criteria to receive an alloSCT at our center have been previously described [1,3,4]. These criteria included age 18 to 70 years and a diagnosis of resistant or relapsed CLL, Richter’s syndrome, MCL or FL that had failed the best conventional treatments available. The other inclusion criteria included an Eastern Cooperative Oncology Group performance status score of 0–2 and adequate liver function (bilirubin and liver enzyme concentrations up to 3 times the upper limit of normal), renal function (creatinine < 1.6 mg/dL), cardiac function (ejection fraction higher than 40%), and pulmonary function (higher than 40% of predictive value). Patients received an alternate donor only if 6/6 human leukocyte antigen-compatible sibling donor or unrelated donor were not available. The exclusion criteria included active central nervous system involvement with disease, pregnancy, breastfeeding, or known infection with human immunodeficiency virus, human T-lymphotropic virus, or hepatitis B or C virus. Additional exclusion criteria were the concurrent presence of other malignancies (with the exception of cutaneous squamous cell or basal cell carcinoma), uncontrolled infection and stroke or myocardial infarction within 6 months of study entry.

Details of the patients’ preparative regimens, supportive care, and infection and graft-versus-host disease (GVHD) prophylaxis, have been previously published [1,3,4]. The predominant preparative regimens used were non-myeloablative, consisting of 30 mg/m² of fludarabine, 130 mg/m² of bendamustine or 750 mg/m² of cyclophosphamide given intravenously on days −5 to −3 prior to transplantation, and 375 mg/m² of rituximab on day −13 and 1000 mg/m² on days −6, +1, and +8 (n = 23/9 [65% total]) [3,4]. Ten patients (20%) received reduced-intensity regimens of fludarabine, melphalan, and rituximab, and seven (14%) patients received busulfan, fludarabine myeloablative conditioning. Thirty (61%) patients received transplantations from histocompatible matched unrelated donors (MUDs) and 11 (22%) from matched siblings; eight (16%) patients who had no matched adult donors received transplantations from haplo-identical donors.

GVHD prophylaxis in matched-donor transplantations consisted of tacrolimus (0.015–0.03 mg/kg starting on day −2) and methotrexate (5 mg/m² on days 1, 3, and 6). Patients who received a transplantation from a MUD received an additional dose of methotrexate (5 mg/m² on day 11) and rabbit antithymocyte globulin (1 mg/kg intravenously on days −2 and −1). Methotrexate was substituted with mycophenolate mofetil after haplo-identical transplantation. In addition, patients with a haplo-identical donor received 50 mg/kg cyclophosphamide on days +3 and +4 after alloSCT. Tacrolimus tapering was initiated in patients with no active GVHD 6 months after alloSCT.

Clinical evaluation and gene sequencing analysis

Acute GVHD and chronic GVHD were graded using standard criteria [14,15]. We assessed patients’ disease status prior to alloSCT, as well as their response to alloSCT, using standard criteria for lymphoma and CLL [16,17]. Patients were evaluated 1, 3, 6, and 12 months after alloSCT; they were then evaluated every 6 months for up to 3 years and yearly thereafter.

Targeted next generation gene sequencing mutation was performed as previously described [18,19]. This was performed using either peripheral blood or marrow samples from patients with CLL or Richter’s syndrome at the time of their first evaluation at MD Anderson and who had at least 10% clonal CD19/CD5–positive cells as evaluated by flow cytometry. Analysis for 29 genes (ATM, BIRC3, BTK, CALR, CARD11, CD79A, CD79B, CHD2, CSMD3, CXCR4, DDX3X, EZH2, FAT1, FBXW7, KLHL6, LRP1B, MAPK1, MUC2, MYD88, NOTCH1, PLCG2, PLEKHG5, POT1, SF3B1, SPEN, TGM7, TP53, XPO1 and ZMYM3) was performed on a next generation sequencing-based panel, designated EndCLL Assay V1.

Statistical analysis

Clinical outcomes evaluated in this study were overall survival (OS), progression-free survival (PFS), relapse, non-relapse mortality (NRM), and GVHD. OS was computed from the date of alloSCT to the last known vital sign. Patients who were alive at the last follow-up date were censored. PFS was computed from the date of alloSCT to the date of disease progression or death (if the patient died without disease progression) or the last evaluation date. Patients who were alive and did not experience disease progression at the last follow-up date were censored. The Kaplan-Meier method was used to estimate OS and PFS. Differences in survival between groups were assessed using the log-rank test. Associations between measures of interest and OS and PFS were assessed using Cox proportional hazards regression models.

NRM was computed from time of alloSCT to last known vital signs, while relapse and GVHD were computed from date of alloSCT to date of event. The cumulative incidences of relapse, NRM and acute GVHD were determined using the competing risks method. The competing risk for relapse was death and for NRM was relapse. Patients who were alive and who did not experience relapse at the last follow-up date were censored. For acute GVHD, the competing risks were relapse and death, while those patients who did not experience the event of interest, did not experience relapse, and were still alive at the last follow-up date were censored. Differences in cumulative incidences between groups were assessed using Gray’s test [20], while associations between measures of interest and cumulative incidence outcomes were determined using proportional sub-distribution hazards regression models [21]. Statistical analyses were performed using SAS 9.4 software for Windows (SAS Institute, Inc., Cary, NC). All statistical tests used a significance level of 5%. No adjustments for multiple testing were made.

Results

Patient, disease, and treatment characteristics

This study group included 49 patients (CLL = 17; Richter’s = 14; MCL = 13; and FL = 5) treated at MD Anderson. The demographic and baseline disease characteristics are summarized in Table 1. Patients were heavily pre-treated and had high-risk features. Most notably, 19 of 33 patients (58%) had TP53 mutations, and 15 of 27 (56%) had acquired other mutations, including seven (26%) with BTK or PLCG2 and four (15%) with FBXW7 and SPEN (Table 2). In addition, 10 of 12 (83%) of patients with MCL had elevated Ki-67 (≥ 30%), and 60% of patients with FL experienced relapse < 2 years after induction chemo-immunotherapy. Six patients (12%) had received programmed cell death protein 1 (PD-1) inhibitor during their disease, two other patients (4%) had failed to experience a clinical response to prior chimeric antigen receptor T (CAR T) cell therapy, and eight patients (16%) had received obinutuzumab.

Table 1.

Patient and clinical characteristics

Characteristic	Groups				P
	CLL (N = 17)	Richter’s (N = 14)	MCL (N = 13)	FL (N = 5)
Age, years
Median	60	60	56	56	0.22
Range	45–71	26–71	39–69	40–61
Gender, no. (%)
Male	11(65)	8 (57)	11 (85)	2 (40)	0.26
Female	6 (35)	6(43)	2 (15)	3(60)
HCT-CI
Median	3	2	2	3	0.78
Range	(0–6)	(0–9)	(0–5)	(0–8)
> 3, n (%)	7 (41)	5 (36)	3 (23)	2 (40)
Prior chemotherapy lines
Median	3	4	3	5	0.15
Range	2–7	1–9	2–9	2–11
Prior auto-SCT, no. (%)	0	0	4(31)	1(20)	0.01
Prior anti-PD1, no. (%)	1(6)	5(36)	0	0	0.02
Prior CAR T cell, no. (%)	1(6)	1(7)	0	0	1.0
Disease status at alloSCT, no. (%)					0.026
CR	4(24)	5(36)	9(69)	4(80)
PR	10(59)	4(29)	1(8)	0
Refractory	3(18)	5(36)	3(23)	1(20)
BM + at alloSCT, no. (%)	13(76)	10(71)	3(23)	1(20)	0.005
PET status, no. (%)					0.12
Negative	8(67)	4(31)	6(75)	4(80)
Positive	4(33)	9(69)	2(25)	1(20)
Not done	5	1	5	0
Donor relation, no. (%)					0.66
MRD	2(12)	4(29)	3(23)	2(40)
MUD	13(76)	7(50)	8(62)	2(40)
Haploidentical	2(12)	3(21)	2(15)	1(20)

Abbreviations: CLL, chronic lymphocytic leukemia; MCL, mantle cell lymphoma; FL, follicular lymphoma; HCT-CI, hematopoietic cell transplantation comorbidity index; PD-1, programmed cell death protein 1; CAR T, chimeric antigen receptor T cell; alloSCT, allogeneic stem cell transplant; CR, complete remission; PR, partial remission; BM, bone marrow; MRD, matched related donor; MUD, matched unrelated donor.

Table 2.

Disease risks and mutation profile.

Characteristics	No. (%)
CLL/Richter’s patients (n = 31)
Deletion 17p	23 (74)
Complex cytogenetics	15 (48)
Unmutated VH	15 (48)
Mutations present	19/27 (70)
TP53 alone	4 (22)
TP53, FBXW7	1 (6)
TP53, XPOL	1 (6)
TP53, ATM	1 (6)
TP53, SF3B1	1 (6)
TP53, BTK	1 (6)
TP53, BTK, POT1	1 (6)
TP53, BTK, SPEN	1 (6)
TP53, BTK, CHD2	1 (6)
TP53, MUC2, SPEN, ZMYM3	1 (6)
TP53, BTK, NOTCH1, KDR	1 (6)
TP53, ATM, PLCG2, SPEN	1 (6)
ATM	1 (6)
ATM, MUC2, BIRC3, BTK, SF3B1	1 (6)
SF3B1	1 (6)
SF3B1, XPO1, BCOR, JAK1, KMT2A	1 (6)
MCL patients (n = 13)
TP53 mutation, no./tested (%)	4/6 (67)
Ki-67 ≥ 30%	10 (77)
Blastoid histology	6 (46)

Abbreviations: CLL, chronic lymphocytic leukemia; V_H, immunoglobulin heavy chain variable region gene; MCL, mantle cell lymphoma.

Pre-transplantation SMI treatment included ibrutinib (n = 46 [94%]), venetoclax (n = 19 [39%]), and idelalisib (n = 6 [12%]), as shown in Table 3. On average, ibrutinib was used as a third-line therapy; only six patients received ibrutinib as first-line treatment. Eleven patients were re-treated with ibrutinib in combination with monoclonal antibodies, chemotherapy, or another class of SMI. Patients received venetoclax as second-line therapy (n = 6), third-line therapy (n = 3), or more (n = 10). Similarly, five of six (83%) patients received idelalisib as an advanced line of therapy. Overall, 43% of patients were treated with ≥ two types of SMI prior to alloSCT, and 29 (51%) were refractory to at least one. None of the patients received routine maintenance therapy after transplant.

Table 3.

Small molecule inhibitors treatment

SMI type and status at transplantation, no. (%)	Value
Ibrutinib	46 (94)
Status at transplant
Sensitive	15 (33)
Refractory	25 (54)
Discontinued because of adverse events	6 (13)
Venetoclax	19 (39)
Status at transplant
Sensitive	12 (63)
Refractory	6 (32)
Discontinued because of adverse events	1 (5)
Idealisib	6 (12)
Status at transplant
Refractory	4 (67)
Discontinued because of adverse events	2 (33)
Number of SMI treatments received
1	28 (57)
2	14 (29)
3	7 (14)
Refractory to ibrutinib and sensitive to venetoclax	10 (20)
Median time from last SMI treatment to alloSCT, days (range)	34 (1–1149)
Time from last SMI treatment to alloSCT
≤ 2 weeks	10 (20)
> 2 weeks	39 (80)
Median duration of SMI treatment, months (range)	9 (2.0–55)
Median time from diagnosis to first SMI treatment, months (range)	34 (1.0–308)

Abbreviations: SMI, small molecule inhibitor; alloSCT, allogeneic stem cell transplantation.

Engraftment

All patients experienced engraftment after receiving unmanipulated donor grafts. Median donor T cells was 100% at both 30- and 100-days post alloSCT. Neutrophil counts recovered to more than 0.5 × 10⁹/L a median of 9 days after alloSCT (range, 0–22 days). Platelet counts recovered to more than 20 × 10⁹/L after a median of 10.5 days (range, 0–32 days). Sixty percent of patients who received the bendamustine, fludarabine, rituximab (BFR) regimen did not experience severe neutropenia (neutrophil count < 0.5 × 10⁹/L) or thrombocytopenia (platelets < 20 × 10⁹/L). Time to engraftment was not significantly different between patients who stopped SMIs within 14 days prior to transplantation and those who stopped earlier (neutrophils > 0.5 × 10⁹/L at 8.5 [0–22] days vs 10 [0–19] days, P = .42; platelets > 20 × 10⁹/L at 5 [0–28] days vs 11 [0–32] days; P = .52).

Survival

With a median follow-up among survivors of 36 months (range, 17 to 72 months), the 3-year OS and PFS rates were 68% (95% CI, 52% to 79%) and 59% (95% CI, 43% to 71%), respectively (Figure 1A). There was no statistically significant difference in OS (Figure 1B) or PFS by histologic subtype (Figure 1C) (Table 2). Conversely, a significant difference in OS and PFS was observed for disease refractoriness (progressing or non-responding to any treatment of chemotherapy or SMI) at alloSCT (Figures 1D and 1E), having a positron emission tomography- positive status at the time of transplant, achieving CR after transplant and experiencing grade III to IV acute GVHD (table 2). In multivariate analysis, disease refractoriness before transplant and severe acute GVHD remained significant for PFS (Table 2). No factor retained its independent predictor status for OS in multivariate analysis. Among the CLL cohort, we also found no difference in PFS or OS between patients with abnormal mutations and those in whom the mutation status was unknown, nor with the number of mutations present (Figure 1F).

Figure 1.

(A) Kaplan-Meier estimate of overall survival (OS) and progression-free survival (PFS) of the entire study group. (B) Kaplan-Meier estimate of overall survival (OS) by histologic type. (C) Kaplan-Meier estimate of OS by disease status prior to alloSCT.

GVHD

The rates of grade II to IV and III to IV acute GVHD were 33% and 7%, respectively (Figure 2A). The cumulative incidences of overall and extensive chronic GVHD were 19% and 16%, 1 year after alloSCT, respectively (Figure 2B). In the univariate analysis, a myeloablative conditioning regimen of busulfan and fludarabine was associated with a higher incidence of grade II to IV acute GVHD (hazards ratio [HR] 3.57; 95% confidence interval [CI], 1.20–10.68; p = 0.010), while prior exposure to PD-1 inhibitor was not found to have a significant association (p = 0.29). Similarly, prior SMI discontinuation (within 2 weeks vs earlier) prior to alloSCT was not significantly associated with incidence of grade II to IV acute GVHD (p = 0.30).

Figure 2.

(A) Cumulative incidence of acute and (B) chronic graft-versus-host-disease (GVHD). (B) Cumulative incidence of non-relapse mortality (NRM) by age group (≤ or > 60 years). (C) Cumulative incidence of non-relapse mortality (NRM) by hematopoietic cell transplantation comorbidity index (HCT-CI).

NRM and relapse

The cumulative incidence rates of NRM for the entire group at day 100, 1 year and 3 years after alloSCT were 2%, 10%, and 14%, respectively. Patients who had grade III to IV acute GVHD (HR 9.06, 95% CI 2.46–33.34; p < 0.001) had a higher rate of NRM than did patients without those features. Age (≤ vs > 60 years) (Figure 2C) and hematopoietic cell transplantation-comorbidity index (HCT-CI) score (Figure 2D) were not found to have a statistically significant impact on NRM.

Eight (16%) patients experienced relapse. After progression post alloSCT, three patients received venetoclax, one received ibrutinib+ obinutuzumab, and one received CAR T without response. Three patients did not receive any treatment because of rapid progression (2) and infection (1).

The 3-year cumulative relapse rates for CLL, Richter’s syndrome. MCL, and FL patients were 24%, 21%, 46% and 0%, respectively (p = 0.45). On univariate analysis, disease refractoriness at the time of alloSCT (HR 7.69; 95% CI 2.24–26.44; p = 0.0001) and not achieving CR after transplant (HR 0.28, 95% CI 0.08–0.95; p = 0.042) were associated with a higher incidence of relapse.

Fourteen patients died at the time of this analysis: 9 from disease relapse, 2 from infection, 2 from acute GVHD and 1 from chronic GVHD.

Discussion

When compared to the results of prior published studies from our institution of patients who underwent alloSCT after chemo-immunotherapy alone, without SMIs, our findings demonstrate that the use of these agents prior to transplantation is safe, with no delayed engraftment, increased risk of acute or chronic GVHD, or increased NRM rate.

We previously reported on the outcomes of 63 patients with CLL who underwent alloSCT after fludarabine, cyclophosphamide, rituximab (FCR) conditioning regimen at our center, with tacrolimus and mini-methotrexate as GVHD prophylaxis [1]. The 3-year OS and PFS rates were 56% and 27%, respectively. The NRM rate at 2 years was 23% and the incidence rate of grade II to IV acute GVHD was 23%. More patients in the current study had 17p deletion (74% vs 24%) and complex karyotype (48% vs 11%). The mutation status was unknown in the control FCR study. Moreover, 16% of patients in the current study received an alternative donor transplantation vs none in the historical FCR study.

In the MCL cohort of the current study, the 3-year OS and PFS rates were 53% and 46%, respectively, where 77% of patients had elevated Ki-67. These results are comparable to the 75 patients who were transplanted pre-SMI era at our center, where the 3-year OS and PFS were 63% and 45%, respectively [5]. The small number of FL patients in this study precludes comparison to previous reports from our institution. Overall, we found no statistically significant difference in outcomes by histologic subtype.

The introduction of novel SMIs, either as frontline therapy or for refractory or relapsed disease, has led to a paradigm shift in therapy for b-cell malignancies, but SMIs have their own drawbacks. Treatment with these agents is usually continuous and indefinite, leading to a higher cost of healthcare. In addition, a significant proportion of patients discontinue treatment because of the development of resistance or intolerance [11,12]. A previous retrospective study of 320 CLL patients treated with ibrutinib at our center showed that 28% discontinued ibrutinib after a median follow-up of 38 months because of intolerance (32%), disease progression (21%), or Richter’s transformation (10%). Patients who discontinued because of disease progression or Richter’s transformation had dismal outcomes, with median survival durations of 16 months and 2.3 months, respectively [11].

Our study was the first to analyze the outcomes of alloSCT in the presence of various CLL mutations that had developed in part because of prior SMI therapies. In a study by Woyach et al [22], baseline karyotypic complexity, presence of del (17) (p13.1), and age < 65 years were risk factors for disease progression. Among patients who experienced relapse in that study, acquired BTK or PLCG2 mutations were found in 85% that were detected a median of 9.3 months before relapse. Seven of the 27 (26%) CLL patients who were tested in our study had acquired mutations in BTK or in PLCG2, the protein immediately downstream of BTK. In addition, four (15%) patients acquired mutations in FBXW7 and SPEN. While the significance of these mutations is not fully understood, emerging data suggest that mutations in FBXW7 and SPEN are associated with inferior clinical outcomes in patients with relapsed CLL who have been treated with lenalidomide and rituximab [23].

The EBMT Chronic Malignancies and Lymphoma Working Parties has reported on the use of ibrutinib prior to alloSCT [24]. That study included 70 patients (58 with CLL and 22 with MCL). Unlike in our study, patients received one SMI, ibrutinib, and 73% of patients were ibrutinib sensitive at the time of alloSCT. The authors found no effect on treatment-related mortality, time to engraftment, or risk of GVHD compared to their findings in prior reports in ibrutinib-naïve patients. Contrary to our study, they reported a higher than expected relapse rate of 30% at 1 year in CLL. In addition, the EBMT study was missing information about important prognostic features, including the HCT-CI score as well as Ki-67 level, and blastoid histologic type in MCL patients. The use of alemtuzumab and antithymocyte globulin in a significant proportion of patients with CLL (10% and 58%, respectively) could explain the higher-than-expected relapse rate in the EBMT study. In another multicenter study, Roeker et al found that HCT-CI score of >0 was predictor on survival [25]. The authors acknowledged the limitations related to the small sample size, and heterogeneity of alloSCT strategies between centers. None of these studies included information on acquired mutations.

Our study’s inherent limitations are its retrospective nature, small sample size, and lack of prospective follow-up status of on acquired mutations post-alloSCT. It has the advantage of being based at a single institution with standardized GVHD prophylaxis and standardized duration and withdrawal of immunosuppression after alloSCT. We believe that a median follow-up period of 3 years is enough to draw conclusions about the safety, NRM, and GVHD and the risk of early relapse of alloSCT after SMI.

In conclusion, our results demonstrate that exposure to SMIs does not increase the rates of graft failure, NRM, GVHD, or relapse after alloSCT. The best clinical outcomes were observed in patients with treatment-sensitive disease, regardless of the type of SMI or chemoimmunotherapy used to achieve it. Additional studies are needed to validate these observations.

Table 4.

Univariate OS and PFS after allogeneic transplant

Covariate	OS			PFS
	HR	95% CI	P	HR	95% CI	P
Age
<60 years	1.00			1.00
≥60 years	1.09	0.41–2.90	.87	0.92	0.40–2.13	.84
Diagnosis
CLL	1.00			1.00
Richter’s	1.72	0.46–6.42	.42	1.68	0.56–5.07	.35
MCL	1.93	0.54–6.88	.31	1.60	0.54–4.79	.40
FL	0.76	0.08–6.83	.81	0.93	0.19–4.61	.92
Disease status at transplant
CR	1.00			1.00
PR	2.28	0.61–8.52	.22	2.62	0.83–8.27	.10
Refractory	5.02	1.45–17.35	.011	7.24	2.43–21.6	<.001a
TP53 mutation
Absent	1.00			1.00
Present	0.67	0.17–2.68	.57	0.95	0.30–3.00	.93
No. of prior therapies	1.01	0.81–1.27	.91	1.06	0.88–1.26	.54
HCT-CI
≤ 3	1.00			1.00
>3	1.27	0.46–3.51	.64	1.40	0.60–3.28	.44
PET status at transplant
Negative	1.00			1.00
Positive	3.01	0.90–10.02	.07	3.29	1.21–8.95	.020
Donor relation
MRD	1.00			1.00
MUD	1.63	0.45–5.85	.42	1.01	0.36–2.81	.99
Haploidentical	1.14	0.19–6.86	.89	0.82	0.19–3.41	.78
Conditioning type
BFR	1.00			1.00
FCR	1.93	0.49–7.58	.35	2.12	0.64–6.98	.22
FM	0.83	0.17–4.02	.82	0.94	0.25–3.48	.93
FBu	1.47	0.37–5.75	.58	2.73	0.90–8.22	.08
Last SMI type prior transplant
Ibrutinib	1.00			1.00
Venetoclax	0.53	0.15–1.94	.34	0.94	0.35–2.52	.90
Idelalisib	2.10	0.46–9.56	.34	2.41	0.68–8.52	.17
Time from last SMI to transplant
≤ 2 weeks	1.00			1.00
> 2 weeks	0.50	0.17–1.46	.21	0.65	0.25–1.67	.37
Achieving CR post-transplant	0.25	0.09–0.71	.009	0.20	0.07–0.58	.003
Acute grade 2–4 GVHD	1.50	0.56–4.05	.42	1.57	0.65–3.78	.31
Acute grade 3–4 GVHD	3.72	1.03–13.47	.045	4.95	1.54–15.90	.007a
Chronic extensive GVHD	0.73	0.15–3.63	.70	0.93	0.24–3.62	.92

Abbreviations: OS, overall survival; PFS, progression-free survival; HR, hazard ratio; CI, confidence interval; CLL, chronic lymphocytic leukemia; MCL, mantle cell lymphoma; FL, follicular lymphoma; CR, complete remission; PR, partial remission;; HCT-CI, hematopoietic cell transplantation comorbidity index; PET, positron emission tomography; MRD, matched related donor; MUD, matched unrelated donor; BFR, bendamustine, fludarabine, rituximab; FCR, fludarabine, cyclophosphamide, rituximab; FM, fludarabine, melphalan; Bu, busulfan; SMI, small molecule inhibitor; GVHD, graft-versus-host disease.

^aRemained significant by multivariable analysis