The prognostic impact of the tumour stroma fraction: A machine learning-based analysis in 16 human solid tumour types

Abstract

Background: The development of a reactive tumour stroma is a hallmark of tumour progression and pronounced tumour stroma is generally considered to be associated with clinical aggressiveness. The variability between tumour types regarding stroma fraction, and its prognosis associations, have not been systematically analysed.

Methods: Using an objective machine-learning method we quantified the tumour stroma in 16 solid cancer types from 2732 patients, representing retrospective tissue collections of surgically resected primary tumours. Image analysis performed tissue segmentation into stromal and epithelial compartment based on pan-cytokeratin staining and autofluorescence patterns.

Findings: The stroma fraction was highly variable within and across the tumour types, with kidney cancer showing the lowest and pancreato-biliary type periampullary cancer showing the highest stroma proportion (median 19% and 73% respectively). Adjusted Cox regression models revealed both positive (pancreato-biliary type periampullary cancer and oestrogen negative breast cancer, HR(95%CI)=0.56(0.34-0.92) and HR(95%CI)=0.41(0.17-0.98) respectively) and negative (intestinal type periampullary cancer, HR(95%CI)=3.59(1.49-8.62)) associations of the tumour stroma fraction with survival.

Interpretation: Our study provides an objective quantification of the tumour stroma fraction across major types of solid cancer. Findings strongly argue against the commonly promoted view of a general associations between high stroma abundance and poor prognosis. The results also suggest that full exploitation of the prognostic potential of tumour stroma requires analyses that go beyond determination of stroma abundance.

Funding: The Swedish Cancer Society, The Lions Cancer Foundation Uppsala, The Swedish Government Grant for Clinical Research, The Mrs Berta Kamprad Foundation, Sweden, Sellanders foundation, P.O.Zetterling Foundation, and The Sjöberg Foundation, Sweden.

Fig. 1

Stroma fraction in solid cancer types. (a) The empirical cumulative distribution plots of the stroma fractions in 16 cancer entities. Each dot represents an individual sample, red horizontal lines indicate the median value of stroma fraction in the respective cancer type. (b) Kaplan–Meier plots illustrate the overall survival in each cohort. Stroma fraction (SF) dichotomized into stroma rich (red) and stroma poor (blue) groups (cut-off = median stroma fraction). The p-values refer to the log-rank test. (c) Forest plots of multivariable Cox regression models representing hazard ratios of the association between stroma fraction and overall survival. For each tumour type, a square indicates the hazard ratio and the line on either side of the square demonstrates 95% confidence interval. A higher hazard ratio indicates that a higher stroma fraction is associated with a shorter survival. Co-variables were selected individually for each tumour type, depending on their availability and known clinical importance (Detailed information is shown in Suppl. Fig). The tumour cohorts are described in detail in the Supplementary Material. **The proportional hazards assumption was violated in PACi and therefore the illustrated hazard ratio should be interpreted as the mean value over time; corrected model, split into time segments before and after 70 weeks, demonstrated by grey colour.