20-Year Steady Increase in Survival of Adult Patients with Relapsed Philadelphia-Positive Acute Lymphoblastic Leukemia Post Allogeneic Hematopoietic Cell Transplantation

Allogeneic hematopoietic cell transplantation (allo-HCT) remains an important and potentially curative treatment modality for various hematologic malignancies including for patients with Philadelphia-positive (Ph⁺) acute lymphoblastic leukemia (ALL), particularly those in first complete remission (CR1) who remain minimal/measurable residual disease (MRD) positive, as well as those beyond CR1 (1). Unfortunately, disease relapse after transplant remains the main cause of failure of allo-HCT. In Ph⁺ ALL, posttransplant relapse occurs in up to 30% of transplanted patients (6–8), and in earlier studies, long-term overall survival (OS) was dismal (2).

In recent years, posttransplant pharmacologic interventions aimed at reducing the risk of relapse in Ph⁺ ALL are now widely applied, using either a prophylactic approach with tyrosine kinase inhibitor (TKI)-based maintenance therapy or a preemptive approach based on regular MRD monitoring (3, 4). Furthermore, new strategies for the management of posttransplant relapse have become available, including the use of newer generation TKI (5–12), monoclonal antibodies, such blinatumomab and inotuzumab ozogamicin, as well as CAR-T-cell therapy (13–17). Finally, increased availability of alternative donors (either well-matched unrelated or haploidentical family donors) facilitate second allo-HCT as salvage therapy.

In other settings, improvement in the management of posttransplant relapse has resulted in progressive increase over time in the survival of young patients with acute myeloid leukemia relapsing after allo-HCT, or Hodgkin lymphoma patients relapsing after auto-HCT (18, 19). The aim of this study was to evaluate, in the setting of Ph⁺ ALL, changes over time between 2000 and 2019, in patients' characteristics, risk factors, and clinical outcomes following relapse after allo-HCT. We used a large sample from the European Society for Blood and Marrow Transplantation (EBMT) registry.

This is a retrospective, registry-based, multicenter analysis. Data were provided and approved by the Acute Leukemia Working Party of the EBMT. The EBMT is a voluntary working group of more than 600 transplant centers that are required to report all consecutive HCTs and follow-ups once a year. Audits are routinely performed to determine the accuracy of the data. Since January 2003, all transplant centers have been required to obtain written informed consent prior to data registration with the EBMT, following the guidelines of the Declaration of Helsinki, 1975.

Eligibility criteria for this analysis included age ≥18 years, first allo-HCT for B-cell Ph⁺ ALL in CR1 and documented hematologic relapse after allo-HCT between 2000 and 2019. Patients only showing decreasing donor chimerism or cytogenetic/molecular relapse were excluded. Donor types included matched sibling donors (MSD) and unrelated donors (UD) regardless of HLA mismatch. Haploidentical donors (Haplo) were excluded because of small numbers. Cord blood transplants were excluded because of the missing opportunity for donor lymphocyte infusion (DLI) and second allo-HCT from the same donor for management of relapse. The stem cell source was bone marrow (BM) or G-CSF-mobilized peripheral blood (PB). Patients who received in vitro T-cell depletion (TCD) were excluded.

Variables collected included recipient age at transplant, recipient and donor gender, date of diagnosis, year of transplant, time from diagnosis to transplant, year of relapse and time from transplant to relapse, Karnofsky performance status (KPS) score at time of transplant, transplant-related factors including conditioning regimen, use of total body irradiation (TBI), graft-versus-host disease (GVHD) prophylaxis, donor type, stem cell source (BM or PB), patient and donor cytomegalovirus (CMV) status, and finally, the development of acute and chronic GVHD before relapse. Relapse-associated variables included the interval from allo-HCT to relapse, OS from relapse, and cause of death. For recipients of DLI, time from relapse to DLI was recorded. For recipients of a second allo-HCT, time from relapse to second transplant, conditioning intensity, donor type, and relapse after second allo-HCT were also recorded.

Myeloablative conditioning (MAC) was defined as a regimen containing either TBI with a dose greater than 6 Gy, a total dose of oral busulfan greater than 8 mg/kg, or a total dose of intravenous Bu greater than 6.4 mg/kg. All other regimens were defined as reduced intensity conditioning (RIC). The diagnosis and grading of acute (20) and chronic GVHD were performed by transplant centers using standard criteria (20). Hematologic relapse was defined by recurrence of blasts in the PB or infiltration of the BM by ≥ 5% blasts. Second allo-HCT was defined as infusion of donor PB or BM stem cells, following MAC or RIC, with immunosuppression for GVHD prevention.

Patient, disease, and transplant-related characteristics for the four cohorts were compared using χ² statistics for categorical variables and the Kruskal–Wallis test for continuous variables.

The primary endpoint was the probability of OS after relapse. Secondary endpoints encompassed causes of death within two years post relapse, cumulative incidence of second allo-HCT after relapse and OS after second allo-HCT. Surviving patients were censored at last contact, and all events were censored at two years after the starting point (date of relapse or date of second transplant).

Probabilities of OS were calculated using the Kaplan–Meier estimates. Cumulative incidence functions (CIF) was used to estimate the incidence of salvage with second allo-HCT, death being a competing event. All probabilities are given in percentage at two years.

Univariate analyses were performed using Gray's test for CIF and the log-rank test for OS. For all univariate analyses, continuous variables were categorized, and the median was used as a cutoff point.

A Cox proportional hazards model was used for multivariate regression. All variables differing between the four time periods, associated with OS in univariate analysis with a nonrestrictive P value of 0.10 and variables known as potential prognostic factors, were included in the multivariate model. Continuous variables included in the Cox model were not categorized. All factors were tested for the proportional hazards assumption. Multivariate results are expressed as a hazard ratio (HR) with a 95% confidence interval (CI). All tests were two sided. The type-1 error rate was fixed at 0.05 for determination of factors associated with time-to-event outcomes. All analyses were performed using SPSS 26.0 (SPSS Inc.) and R version 4.1.0 [R Core Team (2021). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/.)

Data will be available upon request by emailing the corresponding author at [email protected]

We identified 899 patients with a median age at transplant and at relapse of 44 and 45.4 years, respectively. Overall, 116 patients relapsed between 2000 and 2004, 225 between 2005 and 2009, 294 between 2010 and 2014, and 264 between 2015 and 2019. Patient characteristics (Table 1) revealed a progressive increase in patient age at transplant (from 40.6 to 46.1 years; P = 0.007). Regarding transplant characteristics (Table 2) over the four time periods, there was a progressive increase in the use of matched UD (from 34.5% between 2000 and 2004 to 56% between 2015 and 2019; P = 0.0002), PB stem cells (from 60.3% to 84.5%; P < 0.0001), RIC (from 16% to 34.5%; P = 0.004), and in vivo TCD (from 28% to 62%; P < 0.0001) as well as a progressive decrease in the use of TBI (from 73% to 53%; P = 0.0002), respectively. Acute GVHD grade II–IV and chronic GVHD had occurred before relapse in 21% and 21% of patients, respectively.

Median follow-up for surviving patients was 56 months. For the entire cohort, the two-year OS after relapse was 41.5% (95% CI, 38–44.9). Importantly, in univariate analysis, the two-year OS after relapse increased from 27.8% for patients relapsing between 2000 and 2004 to 31.7% for 2005 and 2009, 44.5% for 2010 and 2014 and 54.8% for 2015 and 2019 (P = 0.001; Fig. 1). Original disease was the cause of death in 68.5% of patients, followed by infections (14.3%) and GVHD (8.9%; Table 3). A notable change in the cause of death was observed over time with original disease decreasing from 72.2% for patients relapsing between 2000 and 2004 to 50% for 2015 and 2019, whereas infections increased from 8.2% to 30.6% for the same periods (Table 3).

On univariate analysis (Table 4), OS after relapse was positively affected by the year of relapse, a longer time from transplant to relapse, and patient CMV negativity. Similarly, the cumulative incidence of a subsequent transplant was significantly affected by patient age, the year of relapse, the intensity of conditioning, use of TBI in first transplant, and the time from first transplant to relapse (Table 4).

On multivariate analysis (Table 5), factors that positively influenced OS were a longer time from transplant to relapse (P = 0.0006) and the year of relapse (HR 0.71; P < 0.033 for patients relapsing from 2005 to 2009; HR 0.51; P < 0.0001 for patients relapsing from 2010 to 2014, and HR = 0.37; P < 0.0001 for patients relapsing from 2015 to 2019) and negatively affected by patient age at relapse (P = 0.034). Other patient, donor, and transplant characteristics had no significant effect on OS.

For 537 patients with available data, DLI was given after relapse to 260 (48%) patients after a median of 79.5 days from relapse (IQR 36.8–162.2; Table 6). Trends over time showed a nonsignificant decrease in the use of DLI from 59% for patients relapsing between 2000 and 2004 to 44% for 2015 and 2019, together with a nonsignificant increase in the time from relapse to DLI from 58 to 102 days for the same periods (Table 6).

A second allo-HCT was performed in 115 patients (Table 6). The cumulative incidence of second allo-HCT within two years after relapse was 22%, 13%, 10%, and 16% for the four time periods (P = 0.009; Tables 4 and 5). The cumulative incidence of second allo-HCT was significantly higher following late relapse (19.7% for patients relapsing after more than 7.1 months versus 8.1% for those with an early relapse; P = 0.001; Tables 4 and 5). RIC was utilized in 66 (59%) patients. For 90 patients with available data, the same donor as for first allo-HCT was used in 23 (26%) patients (Table 6).

Overall, second allo-HCT resulted in a two-year OS from the date of second transplant of 35.9% (95% CI, 26.5–45.4%). Trends over time showed a progressive decrease in two-year relapse incidence from second transplant from 74% for 2000–2004, to 54% for 2005–2009, 46% for 2010–2014, and 33% for 2015–2018, respectively (P = 0.03; Table 6).

This retrospective analysis of a homogeneous cohort of 899 patients with hematologic relapse after allo-HCT for Ph⁺ ALL in CR1, analyzed trends in patient characteristics and outcomes over the last two decades. Overall, the two-year OS from relapse was 41.5%, with the original disease being the primary cause of death. Over time, we observed a steady increase in two-year survival from 27.8% to 54.8% (P = 0.001) despite a significant increase in patient age at relapse from 44 to 56 years (P < 0.001).

Historically, patients with Ph⁺ ALL relapsing post allo-HCT had a dismal prognosis with a two-year OS not exceeding 15% (21, 22). Younger age at relapse, longer duration of first response, achievement of second remission on salvage treatment and the performance of allo-HCT were the main factors positively influencing the long-term OS of relapsing Ph⁺ ALL patients (21).

The observed improvement in OS in our study can be explained in several ways, all of which most likely play a certain role in achieving higher responses to post-relapse treatments: first, RIC was used more often in recent years whereas TBI was used less frequently. Hence, patients were less heavily pretreated and may have been able to tolerate more intensive treatments for relapse. Further, patients relapsing more recently had received more T-cell–depleted grafts. Therefore, in these patients, the graft-versus-leukemia (GVL) effect may have been less exploited after the first allo-HCT, rendering the disease more potentially sensitive to GVL-based treatments for posttransplant relapse, such as the use of DLI.

Unlike acute myeloid leukemia, where relapses are likely due to immune function dysregulation, including loss of expression of an HLA haplotype, or an increased expression of inhibitory checkpoint ligands (23–25), the majority of posttransplant relapses in Ph⁺ ALL are the result of the emergence of ABL kinase domain (KD) mutations (26, 27). Almost 70% of imatinib-treated patients are positive for an ABL KD mutation at time of relapse, with T315I mutation being the most frequent (26). The recent wider use of second-generation TKIs, mainly dasatinib and nilotinib in the first-line setting, has resulted in a higher percentage of T315I-mutant relapses in up to 65% of cases (26). Those patients harboring the T315I mutation can benefit from ponatinib, a known third-generation TKI that became available in 2014 for the treatment of Ph⁺ ALL (28).

Thus, the probability of achieving a complete remission after salvage treatment is increasing with the availability of second- and third-generation TKIs, selected based on the underlying ABL KD mutation. A recent EBMT registry study confirmed the efficacy of second-/third-generation TKIs in 140 patients with Ph⁺ ALL, suffering from persistent MRD positivity (n = 6), molecular relapse (n = 23), or hematologic relapse (n = 111) following allo-HCT (12). Treatment included dasatinib in 104, nilotinib in 18, or ponatinib in 18 patients. Response rates were 71%, with a two- and five-year OS of 49% and 33%, respectively (12). OS was comparable among patients treated for MRD positivity or for hematologic relapse. These results compared favorably to an earlier study by the EBMT, where Spyridonidis and colleagues reported a two-year OS of 13% among 465 patients with relapsed ALL of all subtypes after allo-HCT between 1995 and 2000, including 157 with Ph⁺ ALL, 61% receiving imatinib ± chemotherapy (29), suggesting that the consequent use of a second-/third-generation TKI might have substantially contributed to the superior results, given that therapies applied in addition to TKI (chemotherapy, DLI, second allo-HCT) had not changed over time. Unfortunately, we did not have enough details on treatment modalities of relapse to prove this. However, a diminishing frequency of leukemia-related death over the years suggests a better disease control, although at the possible expense of a relative increase in other causes of mortality.

Besides the year of relapse, multivariate analysis confirmed the predictive value of the time from transplant to relapse. The duration of first response is usually associated with a higher response to salvage treatment (21, 30). More therapeutic options may be made available to patients with late relapse, including a second allo-HCT. However, we could not support this hypothesis due to missing information on applied treatments. In addition, we could not assess whether those with a longer time from transplant to relapse received any type of prophylactic TKI maintenance that may have also influenced the choice of salvage treatment.

Some limitations of our retrospective study must be considered. These include the lack of information on MRD status prior to overt hematologic relapse, of relevance if patients were receiving preemptive treatment with TKI prior to their hematologic relapse posttransplant. As discussed above, the lack of detailed information on the treatment of posttransplant relapse (besides second allo-HCT) and DLI in a considerable percentage of patients is another limitation, including the use of blinatumomab, inotuzumab ozogamicin and CAR-T cell therapies. Similarly, we lacked reliable data on maintenance therapy, once a second remission had been achieved. This unfortunately precluded the definition of the role of different innovations in the observed improvement in outcome among younger patients, which, however, was not the main focus of this study investigating overall trends in relapsed patients over time.

In summary, this study represents the largest analysis to date assessing trends over time and predictive factors for outcome of relapsed Ph⁺ ALL after allo-HCT. We observed a major progressive improvement in OS from posttransplant relapse for patients with Ph⁺ ALL, likely reflecting the efficacy of posttransplant salvage strategies. These large-scale real-world data can serve as a benchmark for future studies in this setting.

No disclosures were reported.

A. Bazarbachi: Conceptualization, data curation, formal analysis, supervision, investigation, methodology, writing–original draft, writing–review and editing. M. Labopin: Conceptualization, data curation, formal analysis, validation, investigation, methodology, writing–review and editing. M. Aljurf: Investigation, writing–review and editing. R. Niittyvuopio: Investigation, writing–review and editing. M. Balsat: Investigation, writing–review and editing. D. Blaise: Investigation, writing–review and editing. I. Yakoub-Agha: Investigation, writing–review and editing. A. Grassi: Investigation, writing–review and editing. H.C. Reinhardt: Investigation, writing–review and editing. S. Lenhoff: Investigation, writing–review and editing. P. Jindra: Investigation, writing–review and editing. J. Passweg: Investigation, writing–review and editing. I.A. Dalle: Investigation, writing–review and editing. M. Stadler: Investigation, writing–review and editing. B. Lioure: Investigation, writing–review and editing. P. Ceballos: Investigation, writing–review and editing. E. Brissot: Investigation, writing–review and editing. S. Giebel: Investigation, writing–review and editing. A. Nagler: Investigation, writing–review and editing. C. Schmid: Conceptualization, investigation, writing–review and editing. M. Mohty: Conceptualization, resources, data curation, formal analysis, investigation, writing–review and editing.

There is no funding source for this project.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.