Factors Influencing Relapse after Allogeneic Hematopoietic Cell Transplantation (HCT) Following Reduced Intensity Conditioning (RIC) in Patients with AML and MDS.

Relapse (RI) after HCT especially after RIC remains the major risk factor affecting outcome. Factors influencing relapse in patients (pts) with AML/MDS undergoing HCT after RIC were analysed.

Patients and methods: From July, 1998- December, 2006, 156 consecutive pts [86 m/70 f; median age 61 (range 21–75) years (y)] with AML, n=138 (88%) and high-risk MDS, n=18 (12%) treated within the OSHO studies received HCT after RIC [2Gy TBI ± fludarabine 30 mg/m²/day on 3 days] followed by immunosuppression with mycophenolate mofetil and cyclosporine. Donors were MRD in 40 (26%) and MUD in 116 (74%) pts. Stage of disease was CR1, n= 99 (63%), CR2, n=18 (12%), and >CR2, n=39 (25%). Cytog. was intermediate-, and high-risk in 100 (64%), and 49 (31%) respectively. DCC in unsorted and flow-sorted T (CD3+)-, and CD34+-bone marrow cells at days (d) 28, 56, 84, and at 3 months interval thereafter was monitored by FISH for the XY chromosome in gender mismatched or PCR for polymorphic micro satellite regions in gender matched HCT.

Results: 141 (90%) pts engrafted. OS, DFS, RI, and NRM at 5 y were 38%, 36%, 47%, and 30% respectively. In multivariate analysis, RI for pts in CR1 was significantly lower compared to pts transplanted beyond CR1 (p=0.005). For the entire cohort, type of donor had no influence on RI but for pts in CR1, RI with MUD and MRD was 29% and 60% respectively (p=0.008). NRM did not correlate with the type of donor. Interestingly, high-risk cytog. had no negative impact on outcome. OS, DFS, and RI for pts with high-risk cytog. transplanted in CR1 were 43%, 48%, and 38% respectively compared to 38%, 43%, and 45% for pts with intermediate-risk cytog. (p=0.65). Incidence of acute (grade I-IV) and chronic (limited and extensive) GvHD was 55%, and 58% respectively. GvHD correlated with an improved OS (p=0.0003) and DFS (p=0.003) but was not associated with a higher NRM (p=0.1) compared to pts with no GvHD. Also, RI was markedly influenced by GvHD (p=0.002). RI in pts with no GvHD (n=42), acute GvHD only (n=34), limited chronic GvHD (n=33), and extensive chronic GvHD (n=30) was 63%, 41%, 29%, and 18% respectively. In pts with chronic GvHD, RI was significantly lower compared to pts with no GvHD (p<0.0001). Further, CD34+ DCC ≥90% at d 28 strongly correlated with outcome at 5 y. OS and DFS in pts with CD34+ DCC ≥90% at d 28 (group I) were 50%, and 47% respectively compared to pts with CD34+ DCC <90% at d 28 (group II) (OS 20%, DFS 15%) (p<0.0001). CD34+ DCC ≥90% at d 28 was highly predictive of early (≤100 d) and late (>100 d) haematological relapse. Probability of RI in group I was 29% compared to 76% in group II (p<0.0001). On the other hand, DCC in unsorted- and CD3+-cells at d 28 did not correlate with OS, DFS nor predict relapse.

Conclusions: HCT after RIC offers long-term OS and DFS in AML/MDS pts. RI was lowest in pts transplanted from MUD in CR1. High-risk cytog. had no negative impact on RI. GvHD correlated with an improved outcome. CD34+ DCC ≥90% at d 28 was highly predictive of relapse after HCT. Careful monitoring of CD34+ DCC could identify pts at risk of relapse and might allow therapeutic interventions as tapering of immunosuppression.