Stem Cell

The salvage role of allogeneic hematopoietic stem-cell transplantation in relapsed/refractory diffuse large B cell lymphoma – Scientific Reports

Time Stamp: October 14, 2023 8:00 PM
Source Node: 136556

Baseline characteristics

A total of 52 adult patients with R/R DLBCL were identified and enrolled. The median age at diagnosis was 45 years (range, 17–63), with marginal male predominance (n = 30, 57.7%). 36 patients (69.2%) were identified as having an advanced-stage disease (Ann Arbor stage III–IV), whereas 15 (28.8%) suffered from B symptoms. Thirteen patients (25.0%) had BM involvement of DLBCL at the time of diagnosis, and 8 of 13 (61.5%) had complex karyotypes. All enrolled patients were classified as low (n = 15, 28.8%), low-intermediate (n = 13, 25.0%), high-intermediate (n = 14, 26.9%), and high (n = 10, 19.2%) risks based on international prognostic index (IPI) risk classification. Four patients had a double-hit mutation (7.7%), and according to Han’s criteria, 32 and 13 patients were categorized as ABC type (61.5%) and GCB type (25.0%), respectively. The baseline characteristics at the time of diagnosis are summarized in Table 1.

Table 1 DLBCL characteristics at diagnosis (n = 52).

Full size table

Allo-HSCT characteristics

Notably, most enrolled patients had previously undergone intensive treatments, and the median number of chemotherapy lines before allo-HSCT was 4 (range, 2–6). Sixteen patients (30.8%) had previously undergone auto-HSCT, and the median time from diagnosis and auto-HSCT to allo-HSCT was 27.1 months (range, 6.2–117.7 months) and 18.5 months (range, 6.5–44.8 months), respectively. The disease status at the time of transplantation was CR (n = 14, 26.9%), PR (n = 16, 30.8%), and active disease (n = 22, 42.3%). Donor types included matched sibling (n = 18, 34.6%), haploidentical (n = 18, 34.6%), matched unrelated (n = 11, 21.2%), and unrelated donors with one allele mismatch (n = 5, 9.6%). The conditioning regimen comprised MAC (n = 2, 3.9%) and RIC (n = 50, 96.1%). Half of the enrolled patients were ABO-type matched (n = 26, 50.0%), 27 (51.9%) were in donor-to-recipient sex mismatch, and 44 (84.6%) were both donor and recipient cytomegalovirus (CMV) IgG seropositive. Table 2 presents the demographic information of the patients who underwent allo-HSCT.

Table 2 Allo-HSCT characteristics (n = 52).

Full size table

Response rate, survival outcomes, and complications

Over a median follow-up period of 38.3 months (range, 1.9–112.0), the estimated 5-year OS and EFS were 38.4% (95% CI, 24.7–51.8) and 30.6% (95% CI, 18.8–43.3), respectively. The estimated 5-year CIR, NRM, and GRFS were 36.7% (95% CI, 23.6–49.8), 32.7% (95% CI, 20.3–45.6), and 15.1% (95% CI, 6.9–26.2), respectively. Figure 1 shows the 1-year (100 days outcomes in CIR and NRM) and 5-year survival outcomes. Compared to the active-disease group, the remission-achieved group showed a significantly superior rate of CR in the first 3 months after allo-HSCT (76.7% vs. 36.4%, p = 0.003) and at the last follow-up (50.0% vs. 13.6%, p = 0.006) (Fig. 2). Moreover, except for NRM (26.7% vs. 40.9%, p = 0.217), the clinical outcomes of OS (54.1% vs. 15.6%, p = 0.001), EFS (46.4% vs. 9.1%, p < 0.001), and GRFS (22.9% vs. 4.6%, p = 0.002) were significantly superior in the remission-achieved group with lower CIR (26.9% vs. 50.0%, p = 0.047). In detail (Fig. 3), among patients in the remission-achieved group before allo-HSCT (n = 30), 23 achieved CR, one achieved PR, four experienced disease relapse after allo-HSCT and died due to disease progression, and three died after engraftment (two from bacterial septic shock and one from veno-occlusive disease [VOD]) without relapse. In contrast, among patients in the active-disease group before allo-HSCT (n = 22), eight achieved CR. However, only two remained in CR, four experienced disease relapse, and two died due to septic shock during disease remission. All eight of the remaining patients who achieved PR eventually relapsed, and six died after engraftment either due to allo-HSCT-related complications (one with grade IV hemorrhagic cystitis combined with renal failure, one with VOD, and two with grade IV acute hepatic GVHD complicated with liver failure) or infection (two with CMV pneumonia).

The overall cumulative incidence of grade 3 to 4 acute GVHD and moderate-to-severe chronic GVHD that required steroid pulse therapy was 17.3% (95% CI, 8.5–28.8%) and 29.1% (95% CI, 17.2–42.0%), respectively. The active-disease group had a significantly higher incidence of grade 3–4 acute GVHD than the remission-achieved group before allo-HSCT (6.7% vs. 31.8%, p = 0.023). However, no significant differences in moderate-to-severe chronic GVHD incidence (37.1% vs. 18.2%, p = 0.172) were observed between the two groups. The allo-HSCT-related complication incidence, response rate, and survival outcomes in the remission-achieved and active-disease groups before allo-HSCT are presented in Table 3.

Table 3 Allo-HSCT complications and clinical outcomes (n = 52).

Full size table

Prognostic factor analysis

The results of the univariate analysis for OS, EFS, CIR, NRM, and GFRS are presented in Supplementary Tables 1 and 2. In the multivariate analysis, a shorter interval from diagnosis to allo-HSCT (median of < 27.1 months), which reflects relatively rapid disease progression, showed significantly poor OS (hazard ratio [HR], 3.92; 95% CI, 1.83–8.43; p < 0.001) and EFS (HR, 2.65; 95% CI, 1.28–5.46; p = 0.008). Complex karyotypes in BM involving DLBCL were also associated with poor OS (HR, 1.42; 95% CI, 1.06–1.90; p = 0.018) and NRM (HR, 1.41; 95% CI, 1.02–1.94; p = 0.040). Active disease before allo-HSCT was associated with significantly lower EFS (HR. 2.50; 95% CI, 1.21–5.17; p = 0.014), GFRS (HR, 2.54; 95% CI, 1.37–4.72; p = 0.003), and higher CIR (HR, 3.03; 95% CI, 1.14–8.02; p = 0.026). The MAC regimen for allo-HSCT was associated with significantly higher CIR (HR, 11.3; 95% CI, 3.33–38.1; p < 0.001), and patients with previous autologous stem-cell transplantation (ASCT) had significantly better GRFS (HR, 0.50; 95% CI, 0.26–0.98; p = 0.043). The results of the multivariate analysis are presented in Fig. 4.