Compositional stability of artificial heterogeneous tumors in vivo: use of mitomycin C as a cytotoxic probe.

A major part of the overall response of solid cancers to cytotoxic treatments will be due to the differential sensitivities of the neoplastic cell subpopulations present. To quantitatively investigate this, artificial heterogeneous human colon xenograft tumors were constructed using two clonally related cell lines (A and D) which were mixed to create compositions of approximately 9:1 or 1:9 A:D cells. Then, the pure A, D, or admixed tumors were challenged with mitomycin C which kills A cells more efficiently than D cells by a factor of about 2.3 as determined by in vitro survival curve inactivation slopes. This difference was also exhibited in vivo by the post-mitomycin C cell survival responses determined by excision assay and by the shapes of the regrowth curves of the pure A and D or admixed tumors. By approximately 30 days after treatment, artificial heterogeneous tumors had reached a new stable cellular composition which could be quantitatively predicted based on the individual survival of A and D cells from pure clonal tumors measured 24 h after treatment. Thus, in this model system, cytotoxic treatment of heterogeneous neoplasms produced predictable and stable, albeit altered, percentage admixtures of subpopulations. This result is consistent with the observed decreased clinical responsivity of primary tumors to sequential courses of therapy due to selection of preexisting resistant subpopulations.