Goldie-Coldman model of resistance

James Goldie and Andrew Coldman adapted Luria and Delbruck's bacterial mutation analysis to tumours in 1979: resistance to a drug arises spontaneously at a rate per cell division, so the probability that a tumour of a given size harbours resistant cells is a function of its size and the mutation rate. Their conclusions were to start chemotherapy as early as possible and to alternate non-cross-resistant regimens to pre-empt resistance. Alternating schedules largely failed in trials, but the underlying logic survives in modern evolutionary models and in the case for adjuvant therapy of micrometastatic disease.