Rapid acquisition of multicellular drug resistance after a single exposure of mammary tumor cells to antitumor alkylating agents.

BACKGROUND: Clinical drug resistance is either intrinsic (de novo) or often acquired rapidly in conjunction with chemotherapy. By contrast, the selection of drug-resistant mutant cell lines in monolayer culture systems is usually a more protracted process. Sublines of mouse EMT-6 mammary tumor cells selected for resistance to various alkylating agents in vivo after serial passage into syngeneic mice manifest their resistance in vitro only when cultured as three-dimensional multicellular aggregates or spheroids. PURPOSE: We examined whether a single exposure of mouse EMT-6 or human MDA-MB-231 breast cancer cells to alkylating agents in vitro is sufficient for the induction of a resistance phenotype, which may be detected by re-applying the drugs to cells grown as three-dimensional aggregates. METHODS: Mouse EMT-6 and human MDA-MB-231 breast cancer cells cultured as three-dimensional aggregates were exposed to a single dose of alkylating agent for 1-5 days. Aggregates were dispersed, and cells were plated as monolayer cultures for up to 8 weeks to allow for recovery. Colony-forming ability was assessed after a subsequent alkylating-agent exposure of cells cultured as monolayers or three-dimensional aggregates. RESULTS: A single in vitro exposure to 12.5-microM cisplatin (CDDP) for 5 days or 25 microM 4-hydroperoxycyclophosphamide (4-O2H CTX) for 1 or 3 days without changing the medium was sufficient to induce transient but substantial resistance in EMT-6 cells as determined by clonogenic assays. Such resistance was not detected when monolayer cell cultures were used. The concentration of 4-O2H-CTX and the length of time the cells remained in three-dimensional culture after initial exposure to this drug was associated with the degree of subsequent drug resistance of cells grown as three-dimensional cultures. Furthermore, this acquired resistance after a single drug exposure was accompanied by changes in the three-dimensional architecture of the cell aggregates, which now formed much more compact multicellular spheroids. Similarly, a single exposure to 4-O2H-CTX was enough to bring about resistance in MDA-MB-231 cells detectable only in three-dimensional cultures, as well as the change in three-dimensional architecture. CONCLUSIONS: Rapid acquisition of resistance likely represents a physiologic mechanism of adaptation operative at the multicellular level rather than a stable genetic change and may be one of the reasons for the rapid development of drug resistance acquired by tumors in vivo. IMPLICATIONS: In vivo drug exposure may result in transient and low levels of drug resistance that may nevertheless be clinically relevant.

Extent of vascularization as a prognostic indicator in thin (< 0.76 mm) malignant melanomas.

Angiogenesis in malignant melanoma (MM) was evaluated by comparing mean vessel number (MVN) in Spitz's nevi (SN), thick and thin MMs that metastasized, and thick and thin MMs with > or = 10-year survival. Vessels were identified with antibodies against factor VIII-related antigen (FVIII) and CD34 in 37 MMs (17 < or = 1.9 mm and 20 > or = 4.0 mm) with > or = 10-year follow-up and 10 SN from children (< or = 9 years old). Fields (x250) with the highest vessel density were counted by independent observers blinded to clinical outcome. There were no differences in MVN between SN versus MMs (P = 1.0), but the distribution of vessels was much more uniform in SN. Seven MM pairs (> or = 5.5 mm) and five pairs (< or = 0.75 mm) were matched by sex, age, site, stage, and primary treatment (paired t-test). In the pairs > or = 5.5 mm, there was no correlation with MVN with either metastasis or death (FVIII P = 0.98; CD34 P = 0.85). Among the thin paired lesions, high MVN (FVIII = 46, CD34 = 39) was significantly related not only to metastasis (FVIII P = 0.04, CD34 P = 0.03) but also to death (FVIII P = 0.04, CD34 P = 0.05). MVN does not separate SN versus MM nor predict outcome in thick (> or = 4.0 mm) MMs; however, high MVN (> or = 42 average) is predictive of metastasis and death in MMs < or = 0.75 mm. Larger matched studies are indicated to confirm this observation.