Transplantation outcomes
After a median follow-up of 40 months (95% CI: 36.1–45.7) from transplant, 103 patients had died. The graphical representation of the main results can be seen in Fig. 1(A-F). The estimated OS at 1, 3, and 5 years were 81% (77–85%), 71% (66–76%), and 63% (56–69%), respectively. The estimated 3-year OS rate was 79% (71–88%), 37% (8–66%), 72% (64–79%) and 62% (47–77%) for allo-HCT from MSD, MMRD, MUD and MMUD, respectively. Causes of death were GVHD (n = 32), relapse/progression (n = 28), infection (n = 20), organ failure/toxicity (n = 8), secondary malignancy (n = 5), other (n = 8), and unknown (n = 2).
The cumulative incidence of primary and secondary graft failure by one year was 2% (0–3%) and 6% (3–8%), respectively. Median time to neutrophil engraftment (> 0.5 x 109/L) was 17 days (95% CI: 16–17) and to platelet engraftment (> 20 x 109/L) was 23 days (95% CI: 21–25). Cumulative incidence was 27% (23–32%) for grade II-IV and 12% (9–16%) for grade III-IV acute GVHD at 100 days. Cumulative incidence of any grade and extensive chronic GVHD at 1 year was 40% (35–46%) and 23% (19–28%), respectively.
The estimated risk of NRM at 1, 3, and 5 years was 16% (12–20%), 22% (17–27%) and 26% (21–32%), respectively. Cumulative incidence of relapse/progression at 1, 3, and 5 years was 11% (7–14%), 15% (11–19%), and 17% (13–22%), respectively. The probability of PFS after 1, 3, and 5 years was 73% (69–78%), 63% (58–69%) and 56% (50–63%), respectively. Lastly, the probability of GRFS after 1, 3, and 5 years was 48% (42–53%), 39% (34–45%), and 36% (29–42%), respectively.
Prognostic factors for transplantation outcomes in CALR -mutated MF patients
Factors associated with the main outcomes in the univariable analysis are shown in Table 2. As can be seen, a high comorbidity index (47% (31–63%) vs. 66% (59–73%); p = 0.013) and a PMF diagnosis compared with post-ET MF (55% (46–64%) vs. 79% (70–87%); p = 0.008) correlated with inferior 5-year OS rate. These two factors were associated with higher NRM rate at 5 years after transplant (HCT-CI High: 38% (24–53%) vs. Low/intermediate: 24% (17–30%), p = 0.019; PMF: 32% (24–40%) vs. post-ET MF: 13% (7–20%), p = 0.009), with no difference in relapse incidence. Busulfan-based conditioning was associated with improved 5-year OS rate as compared to melphalan-based or other types (71% (64–79%) vs. 51% (33–69%) vs. 48% (34–62%), respectively; p < 0.001), mainly due to a significantly lower 5-year NRM rate (18% (12–25%) vs. 38% (21–55%) vs. 39% (26–52%), respectively; p < 0.001).
In multivariate analysis, older age (HR: 1.37, 95% CI: 1.08–1.74; p = 0.009) was the only factor significantly associated with a worse OS, whereas PMF (HR: 1.58, 95% CI: 0.96–2.62; p = 0.07) and transplant from a mismatched donor (HR: 1.55, 95% CI: 0.95–2.52; p = 0.08) had a near-to-significant trend for decreased OS. MAC was significantly associated with improved GRFS (HR: 1.45, 95% CI: 1.02–2.05; p = 0.04)(Table 3).
Comparison of transplantation outcomes in CALR - and JAK2 -mutated MF patients
A comparative analysis was performed to evaluate the influence of driver mutation type on transplantation outcomes. The study included a total of 250 patients with CALR-mutated MF and 1022 patients with JAK2-mutated MF who underwent allo-HCT from an HLA matched donor between 2010 and 2019. The main patient characteristics at the time of allo-HCT are shown in Table 4. As can be seen, the only significant differences between the two groups were a higher patient age, a higher prevalence of PMF and a shorter interval from diagnosis to transplantation in the JAK2-mutated cohort. Figure 2(A-E) provides a visual comparison of outcomes between the two groups. Notably, the estimated 3-year and 5-year OS rate was significantly better in CALR-mutated patients (74% (68–80%) vs. 57% (54–60%) by 3 years and 63% (55–72%) vs. 50% (46–53%) by 5 years, respectively, p < 0.001). The leading causes of death were GVHD (31.3%), followed by relapse/progression (28.4%) and infections (20.9%) in CALR-mutated patients, whereas the most frequent cause of death was relapse/progression (32%), followed by GVHD (29.3%) and infections (23.9%) in JAK2-mutated patients, although these figures were not statistically different (Supplementary Table S1). Figure 3 shows the OS curves based on molecular subtype in patients with PMF and post-ET MF separately, demonstrating that the improved OS associated with CALR mutation was evident regardless of the MF type.
The cumulative incidence of primary and secondary graft failure by one year was 1% (0–3%) and 6% (3–9%) for CALR-mutated patients and 4% (2–5%) and 4% (3–5%) for JAK2-mutated patients, respectively. Cumulative incidence was 25% (19–30%) vs. 28% (25–31%) for grade II-IV and 11% (7–15%) vs. 14% (11–16%) for grade III-IV aGvHD at 100 days for the CALR-mutated and JAK2-mutated cohorts. Cumulative incidence of any grade and extensive chronic GVHD at 1 year was 39% (33–46%) vs. 35% (31–38%) and 22% (17–28%) vs. 21% (18–23%), respectively, for the CALR-mutated and JAK2-mutated groups. Overall, no significant difference in the incidence of acute or chronic GVHD was observed between the groups (Supplementary Table S2).
The estimated risk of NRM at 1, 3, and 5 years was significantly lower in CALR-mutated patients (15% (10–19%) vs. 24% (21–26%), 20% (14–25%) vs. 30% (27–33%), and 26% (18–34%) vs. 33% (30–37%), respectively, p = 0.003). The cumulative incidence of relapse/progression at 1, 3, and 5 years was also lower in CALR-mutated patients (12% (8–16%) vs. 15% (13–18%), 16% (11–21%) vs 21% (19–24%), and 16% (11–21%) vs. 25% (22–28%), respectively, p = 0.024). The probability of PFS after 3 and 5 years was 64% (58–71%) vs. 48% (45–52%) and 58% (50–66%) vs. 41% (38–45%), respectively; p < 0.001). Finally, the probability of GRFS after 3, and 5 years was 40% (33–47%) vs. 30% (27–33%), and 37% (29–44%) vs. 25% (22–28%), respectively, for the CALR-mutated and JAK2-mutated cohorts (p = 0.004). Detailed results are shown in Supplementary Table S2.
Finally, Table 5 shows the impact of molecular subtype on the main transplantation outcomes adjusting for other potentially relevant factors. Of note, the CALR mutation was found to be an independent prognostic factor that was associated with improved OS, PFS and GRFS.