Autologous BM and PB HPC are usually stored from weeks to months until reinfusion after myeloablative chemotherapy. HPC have been stored for up to 16 months at -90 degrees C, using a mixture of 5% DMSO, 6% hydroxyethyl starch (HES), and 4% HSA as a cryoprotectant. Long-term storage (LTS) has usually entailed rate-controlled freezing using 10% DMSO and preservation in liquid nitrogen. The effects of LTS at -90 degrees C on the in vitro cell recovery, viability, and colony-forming unit-granulocyte macrophage (CFU-GM) clonogenic potential of autologous HPC that were not transplanted was studied. Sixteen BM and sixteen PB HPC had been cryopreserved for a median of 53 months (range 27-71) and 35 months (range 26-78), respectively. Samples of frozen HPC were thawed after 48 h, and the nucleated cell count, viability by trypan blue exclusion, and culture for CFU-GM were obtained. Following LTS, the cells were thawed and examined using the same assays. No difference in the median percentage recovery of nucleated cells was found in either the BM or PB HPC between the samples stored for 48 h and after LTS (5.73 x 10(9) versus 5.61 x 10(9) and 6.20 x 10(9) versus 5.78 x 10(9), respectively). In addition, no difference in median percentage viability was found in either the BM or PB HPC sampled at 48 h and at the end of LTS (75% versus 74% and 75% versus 76%, respectively). Finally, the median number of CFU-GM cultured from BM HPC at 48 h was 2.41 x 10(5) (range 0.33-11.01 x 10(5)) and at the end of LTS was 1.93 x 10(5) (range 0.32-10.55), representing a median recovery of 93% (range 19%-308%). Similarly, the median number of CFU-GM cultured from PB HPC was 1.66 x 10(5) (range 0-50.57) and at the end of LTS was 0.93 x 10(5) (range 0-44.9), representing a median recovery of 80% (range 36%-165%). This difference in percentage recovery was not significant (p = 0.514). There was poor correlation between the number of nucleated cells harvested and the percentage recovery of nucleated cells, cell viability, or CFU-GM for either the BM or PB HPC. Similarly, there was poor correlation between the number of CFU-GM in the harvest and their percentage recovery following LTS for both BM and PB HPC. Finally, there was poor correlation between the storage time of the BM or PB HPC and the percentage recovery of nucleated cells, cell viability, and CFU-GM. These data suggest that LTS of HPC at -90 degrees C is not associated with decreased recovery of nucleated cells or in vitro viability and is associated with only a modest decrease in clonogenic potential. This indicates that storage of HPC at -90 degrees C for periods in excess of 3 years is possible.