High-dose chemotherapy with autologous peripheral blood stem cell transplantation (auto-PBSCT) offers a valid treatment option in cancer patients (pts) with advanced, relapsed or high-risk (HR) disease. Auto-PBSCT is in general a safe procedure, however, some pts at risk may have a dismal outcome due to treatment related mortality (TMR), early relapse/progression (ER) and/or failure of blood cell reconstitution (FBR). Risk factors predisposing for early events after PBSCT have not yet been clearly assessed. From a total of 796 pts having been transplanted over a 10 year (y) period (6/93–5/03), we determined 50 pts (group A, m:f=29:21), who had primary FBR or died within 3 months after PBSCT. As a control group, 50 pts who had undergone PBSCT over the same period, were randomly selected (group B; m:f=25:25). Pre-PBSCT-parameters associated with early complications were examined by univariate analysis. Median age at transplant was similar in group A (52y) and B (54y). Group A vs. B had underlying lymphoma in 31 vs. 17, solid tumors in 13 vs. 21 and leukemias in 6 vs. 12 pts, respectively. Notable differences were a lower performance status, more pretreatment chemotherapy (CX)-cycles, higher LDH levels, lower platelet counts, hypocellular bone marrow (BM) and lower colony forming units (CFU) in group A. Treatment before PBSCT in group A consisted of a median number of 7 CX (group B: 4 CX) cycles. Median number of retransfused CD34+-cells were similar (group A 4.1 vs. group B 3.7×10e6/kg bw). Total CFUs/1.5e5, determined by methylcellulose culture of PBSC grafts in group A were decreased with 128 vs. 187 (group B), and BFU-E, CFU-GM, and CFU-GEMM were 51 vs. 79, 55 vs. 87 and 5 vs. 4, respectively. Hypocellular BM was observed in group A and B in 46% vs. 4% of pts (odds ratio 20.44 [p<0.001]), platelet counts were 98 vs. 170×10e9/L (odds ratio 13.5 [p<0.001]), respectively. Advanced stage at transplant was present in 90% vs. 78%, and residual BM-infiltration in 36% vs. 58% in group A and B. Age, number of CX-cycles and CD34+-transfused cells showed no significantly increased risk. Of pts in group A, 21 had early TRM (group A1: sepsis and multi-organ failure in 81%, fatal bleeding in 14%, CX-toxicity in 5%; death after median of 23 days [d]). 26 pts (52%) died of early progression (group A2: after a median of 62d) and 3 pts (6%) had FBR (group A3). Median BFU-E, CFU-GM and CFU-GEMM were similar in group A1 and A2 and correlated with comparable CD34+-numbers. However, in group A1, 29% reached no WBC and 81% no platelet engraftment, whereas in group A2 only 12% and 23% did not reach WBC- and platelet reconstitution, respectively. Engraftment in group B (control pts) was regular (median WBC and platelet reconstitution on d+10). With a TRM of 2.64% and FBR in 0.38% in this HR-pt cohort, auto-PBSCT is in general a well tolerable treatment option. Nevertheless, distinct risk factors for early complications can be determined and should be considered. We identified BM-hypocellularity and a decreased platelet count <100×10e9/L as the strongest prognostic parameters. Our analysis also suggests, that other pre-transplant parameters, including CFUs - more reliably than CD34+-cell numbers - provide additional valuable information on engraftment and relevant events after auto-PBSCT.

Disclosure: No relevant conflicts of interest to declare.

Author notes


Corresponding author