Background: Linear and parallel are the two leading models of metastatic progression. In this study we propose a simple way to differentiate between them. While the linear model predicts accumulation of genetic and epigenetic alterations within the primary tumor by founder cells before spreading as waves of metastases, the parallel model suggests preclinical distribution of less advanced disseminated tumor cells with independent selection and expansion at the ectopic sites. Due to identical clonal origin and time of dispatching, linear metastases are expected to have comparable diameters in any specific organ while parallel metastases are expected to appear in variable sizes.
Methods and findings: Retrospective revision of chest CT of oncological patients with lung metastases was performed. Metastasis number and largest diameters were recorded. The sum number of metastases with a similar diameter (c) and those without (i) was counted and the linear/parallel ratio (LPR) was calculated for each patient using the formula (∑c-∑i)/(∑c+∑i). A LPR ratio of 1 implies pure linear progression pattern and -1 pure parallel. 12,887 metastases were measured in 503 patients with nine malignancy types. The median LPR of the entire group was 0.71 (IQR 0.14-0.93). In carcinomas of the pancreas, prostate, and thyroid the median LPR was 1. Median LPRs were 0.91, 0.65, 0.60, 0.58, 0.50 and 0.43 in renal cell carcinomas, melanomas, colorectal, breast, bladder, and sarcomas, respectively.
Conclusions: Metastatic spread of thyroid, pancreas, and prostate tumors is almost exclusively by a linear route. The spread of kidney, melanoma, colorectal, breast, bladder and sarcoma is both linear and parallel with increasing dominance of the parallel route in this order. These findings can explain and predict the clinical and genomic features of these tumors and can potentially be used for evaluation of metastatic origin in the individual patient.