Single or tandem autologous stem cell transplantation for treating Chinese patients with refractory/relapsed classical Hodgkin lymphoma

Abstract Background Autologous stem cell transplantation (ASCT) is the standard treatment strategy for refractory or relapsed classical Hodgkin lymphoma (R/R cHL). However, a single transplantation is insufficient to cure the disease because of unfavorable risk factors. Herein, we evaluated the outcomes of single or tandem ASCT in patients with R/R cHL, especially in high‐risk patients. Methods We retrospectively analyzed R/R cHL patients who underwent single or tandem ASCT between April 2000 and June 2021 at the Beijing Cancer Hospital and Peking University International Hospital. Results A total of 134 patients were enrolled. Patients were allocated to a favorable‐risk group (group A, n = 33), an unfavorable‐risk group (group B, n = 81) that underwent single ASCT, and an unfavorable‐risk group that underwent tandem ASCT (group C, n = 20). The median follow‐up time was 99 months (range, 91–107 months), and no treatment‐related deaths occurred after single or tandem ASCT. However, 27 patients (2 in group C) died during the follow‐up period. The groups A, B, and C had 5‐year progression‐free survival (PFS) rates of 77.05%, 45%, and 74.67%, respectively (p = 0.0014), and 5‐year overall survival (OS) rates of 89.85%, 76.06%, and 95%, respectively (p = 0.18). Neither the median PFS rates of groups A and C nor the OS rates of all groups were reached. Conclusions Our study discusses the advantages of tandem transplantation for high‐risk patients with R/R cHL.


| INTRODUCTION
Novel molecular therapeutic agents, such as brentuximab vedotin (BV), and immune checkpoint inhibitor, programmed death-1 (PD-1), such as nivolumab, enhance the prognosis of refractory or relapsed classical Hodgkin lymphoma (R/R cHL) and prolong survival after the failure of autologous stem cell transplantation (ASCT). [1][2][3][4][5][6][7] High-dose chemotherapy (HDCT) along with ASCT is the standard treatment strategy for cHL patients who fail first-line therapies or have a relapse. Approximately half of patients with chemotherapysensitive disease are cured with ASCT, 8 whereas the other half, especially those with unfavorable risk factors, ultimately die because of disease progression and complications without successful treatment. 9 The prognosis of patients with recurrence after ASCT is poor. [10][11][12][13] According to data from the European Society for Blood and Marrow Transplantation, the 5-year overall survival (OS) is only 32%. Unfavorable-risk factors include primary induction failure, extranodal involvement, B symptoms at relapse, disease relapse within 1 year of induction therapy, resistance to salvage therapy, and more than one salvage therapy regimen in patients with R/R cHL. [14][15][16] However, a single transplantation is insufficient for long-term control or to cure the disease because of unfavorable risk factors. Reducing recurrence after ASCT in high-risk patients remains an issue in current research.
The AETHERA study found that consolidation with BV after ASCT improved the 5-year progression-free survival (PFS) of R/R cHL patients with unfavorable risk factors in comparison to the placebo. 14,15 BV consolidation treatment after transplantation has been used as a guideline in high-risk patients. However, in China, few patients receive BV consolidation treatment because of its high cost. Checkpoint inhibitors are also important drugs in patients with R/R cHL. To date, only one small sample study has been reported; PD-1 blockade administered as consolidation therapy after ASCT requires more data to confirm its effectiveness and safety. 17 In addition to this new drug consolidation treatment, tandem transplantation was considered. However, the appropriate patients and opportunity for a second ASCT are unclear, since findings from previously published literature are limited. [18][19][20][21] Herein, we conducted a retrospective analysis to evaluate the outcomes of R/R cHL patients with at least one unfavorable prognostic factor after a single or tandem ASCT, and the significance of tandem transplantation in highrisk patients. Furthermore, we compared the survival of favorable-risk patients who received a single transplant in the same period.

| Patients
A total of 134 patients with R/R cHL with or without unfavorable prognostic factors, who underwent single or tandem ASCT between April 2000 and June 2021 at the Peking University Cancer Hospital & Institute and the Peking University International Hospital, were included in the analysis. Pathological diagnosis was confirmed in all patients by examination of biopsy materials according to the criteria of the WHO Classification of Hematological Malignancies. Data from the follow-up period were retrospectively analyzed. This study was conducted in compliance with the Declaration of Helsinki and was approved by the Ethics Committee of the Peking University Cancer Hospital and the participating center. Written informed consent for participation was not required for this study, in accordance with national legislation and institutional requirements.
Unfavorable risks included cancer relapse within 1 year of induction therapy, primary induction failure, B symptoms at relapse, resistance to salvage therapy, more than one salvage therapy regimen before transplant, and extranodal involvement at relapse. [14][15][16] Patients without (n = 33) and with (n = 81) at least one unfavorable risk factor underwent a single ASCT, whereas the remaining patients (n = 20) with at least one unfavorable risk factor underwent tandem ASCT. The decision on the type of transplant was made by the researchers. The number of stem cells collected was recorded. The baseline characteristics of the patients in groups B and C were similar. Patients in the single and tandem ASCT groups were infused with 1.75 × 10 6 and 4.57 × 10 6 CD34 + cells/kg (median values) of autologous peripheral blood progenitor cells, respectively. The preparatory regimen, stem cell apheresis procedures, and cryopreservation of stem cells were conducted in accordance with the institutional practice guidelines, and the data were retrospectively analyzed.

| Evaluation of efficacy and follow-up
Responses were assessed before and after 6 weeks of ASCT according to the criteria of Lugano. 22 Positron emission tomography and computed tomography (PET-CT) scans were obtained and analyzed using the Deauville criteria. 23 Follow-up CT was performed every 3-6 months for 2 years and then semi-annually or annually thereafter, and the scans were analyzed. OS was defined as the date of the first conditioning treatment regimen until death from any cause or the date of the last follow-up. PFS was defined as the date of the first administration of the conditioning treatment regimen until cancer progression or the latest follow-up date.

| Statistical analysis
IBM SPSS Statistics software (Version 24.0, IBM Corp) and R version 4.1.2 (https://www.r-proje ct.org) were used for statistical analysis. The chi-squared test or one-way analysis of variance was performed to determine the significance of difference. PFS and OS rates were assessed using the Kaplan-Meier method and log-rank tests. Univariate analysis was performed to analyze the baseline clinical characteristics and treatment programs to determine the significant association of prognostic variables with PFS and OS. Cox regression models were used to determine the significance of the influence of different prognostic variables identified using univariate and multivariate analyses. Statistical significance was set at p < 0.05.

| Baseline and clinical characteristics of patients
The median age of the patients (n = 134) with R/R cHL who underwent single or tandem ASCT was 26 years (range, 20-34 years). The sex ratio was 1:1.8 (women:men). The most common pathological subtype was nodular sclerosis (71%), and 88 patients (66%) were initially at an advanced stage of cancer. Baseline characteristics such as stage, ECOG performance status, and B symptoms of all populations are matched; while group C patients are younger. Sixty-three (47%) patients achieved complete response (CR), 51 (38%) achieved partial response (PR), and only 14 (10%) patients showed disease progression at the time of transplantation ( Table 1). The baseline characteristics of patients in groups B and C, including age, sex, pathological subtype, stage, and ECOG performance status are matched. Analyze the high-risk factors of groups B and C, the status at the time of recurrence, including extranodal involvement and B symptoms, is similar. There are also no statistical differences in the number of risk factors. Although a higher proportion of patients in group C were treated with intensive induction therapy such as BEACOPP, there were still more patients with primary drug resistance (25% vs. 75% p < 0.001) ( Table 2). Since the approval of use of PD-1 inhibitors in China in 2018 and that of BV in 2020, only eight patients have received these drugs as salvage therapy. Ninety-six (71.6%) patients underwent PET-CT, and 38 (28.4%) underwent CT examinations to evaluate their characteristics before ASCT. The median number of infused CD34 + cells was 2.01 cells/kg (range, 1.13-4.93 cells/kg). Stem cells were successfully transplanted into all the patients. The median time required for neutrophil (>0.5 × 10 9 /L) and platelet (>20 × 10 9 /L) recovery was 11 days (range, 10-12) and 11 days (range, 9-13), respectively.

| Responses and survival
Seventy-six (66.7%) patients achieved CR, and 46 (22.8%) achieved PR after one course of ASCT (groups A and B). In group C, 40% of patients achieved CR before the onset of ASCT, and after the first and second transplants, their CR rates increased to 75% and 90%, respectively. Ninetysix and 38 patients underwent PET-CT scanning 6 weeks and CT scans 4 weeks after ASCT, respectively. In addition, 93 (69%) patients in both groups achieved CR.
During the follow-up period (median, 99 months; range, 91-107), 27 patients (2 in group C) died, including 25 deaths caused by lymphoma and 2 caused by pneumonia (in group B). Among the patients (n = 134) included in the survival analysis, the 5-year OS and PFS rates were 82.4% and 57.3%, respectively. A significant difference in PFS (p = 0.0246), but not OS, was found between groups B and C. The 2-year PFS rates in groups B and C were 53.6% and 80%, respectively. The 5-year PFS rates of groups A, B, and C were 77.05%, 45%, and 74.67%, respectively (p = 0.0014), and their 5-year OS rates were 89.85%, 76.06%, and 95%, respectively (p = 0.18) ( Figure 1). Neither the median PFS of groups A and C nor the median OS of all three groups were reached. The 5-year PFS rates of patients in the three groups who experienced CR, PR, or less than PR before ASCT were 73.69%, 49.07%, and 30%, respectively (p = 0.0016). Their 5-year OS rates were 94.4%, 76.04%, and 64.3%, respectively (p = 0.00086) (Figure 2).  In group A, seven patients relapsed and two patients were treated with PD-1 inhibitors. Among the 43 patients with recurrence in group B, 16 patients were given PD-1 inhibitors, two patients were given BV, and two patients were given both two. While in group C, one recurrent patient in five received BV and the other four patients received PD-1 inhibitors (most of them participated in clinical trials). In addition, there were two patients in group B who received allogeneic transplantation.
Pre-transplant data of the PET-CT scans of 96 patients were available for analysis. The pre-transplant 5-year OS and PFS rates of the PET-CT-negative patients were 95.5% and 72.32%, respectively, whereas those of the PET-CTpositive patients were 76.1% and 51.7%, respectively (OS, p = 0.0061; PFS, p = 0.0063). Ninety-six patients underwent post-transplant PET-CT. The 5-year OS and PFS rates of post-transplant PET-CT-negative patients were 95.46% and 76.95%, respectively, whereas those of the PET-CTpositive patients were 59.3% and 19.48%, respectively (OS, p < 0.0001; PFS, p < 0.0001). The 5-year PFS of patients with or without unfavorable risk factors was 51.05% and 77.05%, respectively (p = 0.012), and the 5-year OS rates were 89.95% and 80%, respectively (p = 0.3) (Figure 3). The number of previous chemotherapy regimens affected PFS, but not OS. Thus, the 5-year PFS rates of patients who underwent more than or less than two regimen lines were 34.4% and 64.62%, respectively (p = 0.00055).

| DISCUSSION
cHL is highly treatable with initial chemotherapy. However, 10%-35% of patients experience disease progression. 24,25 Treatment of patients with R/R cHL remains a challenge despite the use of novel molecular therapies, even after BV and immune checkpoint inhibitors have been approved for the treatment of R/R patients undergoing ASCT. However, the role of ASCT in the treatment of R/R cHL has been well-established. Approximately half of patients with chemotherapy-sensitive disease are cured with ASCT, 8 whereas the other half, especially those with unfavorable risk factors, exhibit disease progression and have poor outcomes. 9 The following unfavorable factors are defined as high-risk: primary induction failure, extranodal involvement, B symptoms at relapse, disease relapse within 1 year of induction therapy, resistance to salvage therapy, and more than one salvage therapy regimen before transplantation. [14][15][16] Attempts are being made to improve the survival of high-risk patients through new drug consolidation or sequential transplantation. Our study discusses the advantages of tandem transplantation for high-risk patients with R/R cHL. This kind of treatment is not only effective but also accessible and affordable. The AETHERA study found that early consolidation with BV after ASCT improved the 5-year PFS of R/R cHL patients with unfavorable risk factors in comparison to placebo (59% vs. 41%, respectively; HR 0.521). 14,15 BV consolidation treatment after transplantation has been used as a guideline in high-risk patients. However, in China, few patients receive the treatment because of its high cost. Checkpoint inhibitors are also important drugs for treating patients with R/R cHL after ASCT. Two recently published related phase II trials, CheckMate 205 and KEYNOTE-087, have demonstrated median PFS rates of 14.7 and 13.7 months, respectively. 3,4 However, the efficacy duration of the checkpoint inhibitors remains far from satisfactory. In a phase II study with a small sample size, 17 30 patients received PD-1 blockade treatment after ASCT for eight cycles. Overall improvement in PFS at 18 months after ASCT from the estimated historical results was in the range of 60%-82%; therefore, the author concluded that checkpoint inhibitor consolidation is a successful method for R/R cHL patients post-ASCT. This phase II study has some limitations. First, the sample size was small, and this was not a randomized controlled study. Second, not all the patients belonged to high-risk groups, and 87% of them met the eligibility criteria for the high-risk AETHERA study. Thus, the follow-up time was short. PD-1 blockade administration consolidation therapy after ASCT requires more data to confirm its effectiveness and safety. In addition to the treatment involving consolidation with new drugs, tandem transplantation was also considered as an important regimen. Limited data are available regarding the use of tandem therapy in unfavorable R/R cHL patients; however, the H96 trial by the LYSA/ SFGM-TC study group had a large sample size and long follow-up duration. 19 This multicenter phase II trial included 245 primary refractory or first-relapse cHL patients eligible for single or tandem ASCT, and confirmed the feasibility of tandem transplantation for high-risk patients. Nevertheless, the H96 trial had some limitations. PET was not included in the assessment criteria in the beginning of the trial, and only patients with first-line treatment failure were included. Patients with resistance to salvage therapy or persistent disease before transplantation were not included. The 5-year PFS rate was 95% in our study, whereas the 10-year PFS rate was 47% in the H96 trial. However, direct comparisons cannot be drawn between the two studies because the baseline information and follow-up duration were inconsistent. The results of this study demonstrated that R/R cHL patients with an unfavorable prognosis achieved long-term remission or cure after undergoing tandem ASCT, with favorable treatment outcomes. Forty-six R/R cHL patients with high-risk factors were included in another study, of which 41 received tandem ASCT and reported a 5-year PFS of 49%, 20 while we observed a 74.67% 5-year PFS rate in our study. The following differences were noted between the two studies: the duration when the study was conducted (our study being more recent); the definition of high-risk patients (our inclusion criteria comprised more high-risk factors and were widely acceptable); and the survival rate of the patients (which showed greater improvement in our study). However, the sample size of both the studies was relatively small. Our results were also superior to those of the SWOG 0410 trial, in which 82 patients with R/R cHL received tandem transplantation (5-year PFS, 55%). 21 These differences may be due to variability in the inclusion criteria of patients and sample size. Previous studies have also discussed the conditioning regimen for transplantation of R/R cHL patients. The choice of conditioning regimen for ASCT II has traditionally been based on institutional experience. Several regimens such as BEAM, CBV, BuCy, and TBIcontaining are considered standard and routinely used for lymphoma patients. A large study from CIBMTR showed 26 that patients with HL receiving BEAM had superior overall survival compared to all other regimens, probabilities of 3y-OS were: BEAM 79%, CBV 68%, BuCy 65%, and TBI 47% (p < 0.001). In our study, preference will be given to BEAM for ASCT I, other classical conditioning regimens are an option for ASCT II, CBV in particular.
In the present study, the 5-year PFS of unfavorablerisk patients who received a single ASCT was 45%, similar to that of the placebo group in the AETHERA study. 14,15 Moreover, we observed a 5-year PFS rate of 75% in patients with unfavorable risk factors who were treated with tandem ASCT, which was higher than that in patients treated with BV consolidation. On comparison, tandem ASCT was superior to BV consolidation after a single course of ASCT in patients with unfavorable risk factors. There is a report of anti-PD-1 consolidation on small sample size recently, with a 2-year PFS and OS rates of 79% and 87%, respectively. 27 Compare with our study, the 2-year PFS and OS rates were 80% (95% CI 64-100) and 95% (95% CI 86-100), respectively. There is also report of the two drugs combination, BV combined with anti-PD-1 was given for consolidation of high-risk patients after transplantation. At a median follow-up of 29.9 months, the 18-month PFS was 94% (95% CI 84-98). 28 In China, the cost of BV consolidation treatment is at least five times that of ASCT, and tandem transplantation has considerable practical significance for unfavorable-risk R/R cHL patients.
Nevertheless, the present study was not a randomized study that compared tandem ASCT with BV consolidation. Furthermore, the baseline characteristics of the populations in different studies were not similar. Our study was retrospective and did not directly compare various treatment regimens. Additionally, the sample size was small. Therefore, whether tandem ASCT can replace BV consolidation in these patients remains unclear. However, tandem transplantation costs less and has higher acceptance in China. In future, we plan to conduct a randomized controlled study with a larger sample size to confirm the practical significance of tandem ASCT.
Nevertheless, tandem ASCT is a viable option for treating patients with unfavorable prognosis in the present era of new molecular therapeutic agents. Treatment strategies such as consolidation with new drugs, tandem-autoSCT or auto-alloSCT, [29][30][31] and radiation therapy 32 can improve patient outcomes after ASCT. The clinical trials designed by us in the future will randomly group unfavorable patients who have already undergone ASCT1 to receive ASCT2, BV, immune checkpoint inhibitors, radiation, or allogeneic stem cell transplantation.
The results of the current trial demonstrate a riskadapted strategy for patients with R/R cHL. In our study, it was confirmed that tandem ASCT improved the PFS of unfavorable-risk patients compared with single transplantation, but there was no significant difference in overall survival. A higher proportion of patients in single ASCT group were subsequently treated with BV and PD1-blockers (46.5% vs. 40%), In addition, there were two patients in single ASCT group who received allogeneic transplantation. This may be the reason for the lack of OS differences across groups in our study. In addition, our study focused on the prognostic significance of PET/ CT results. Pre-ASCT PET-CT findings are critically important for identifying prognosis in patients with R/R cHL. [33][34][35][36][37] For example, a previous study revealed that the findings from PET scan images were associated with PFS in patients with R/R cHL, which is consistent with the data from our study. Here, the 5-year PFS rates of patients who achieved CR or PR before ASCT were 73.69% and 49.07%, respectively, although the 5-year PFS rates of patients with SD and PD were 30%. Therefore, patients may achieve a PR or better response before undergoing ASCT.

| CONCLUSION
In the era of new drugs, the aim of this multicenter retrospective study with a long follow-up duration was to improve the survival of unfavorable-risk R/R cHL patients in