Intra-patient comparison of microarchitecture of tumour negative lymph nodes from oesophageal cancer patients – Results from the MRC Oe02 trial

Background Regional lymph node (LN) status is a key prognostic factor in oesophageal cancer (OeC). Tumour-derived antigens can activate immune reactions in LNs, potentially reflecting the host’s anti-tumour immune response. It remains unclear whether this response is homogeneous across all tumour negative LNs (LNneg) within individual OeC patients. Purpose To investigate the hypotheses: (1) the host anti-tumour immune response is similar in all LNneg from an individual OeC patient reflected in a similar microarchitecture in all LNneg; and (2) immune response measured in the largest LNneg can represent that of all LNnegs. Methods (y)pN0 patients from the Oe02 trial with at least two LNneg were included. Microarchitectural LN features (germinal centres (GermC), lymphocytes outside GermCs (lymphocytes), histiocytes) were morphometrically quantified. Linear mixed-effects models, intraclass correlation coefficients (ICC) and Bland-Altman plots were used to determine systematic bias, reliability/variability and agreement of LNneg microarchitecture measurements. Results Linear mixed-effects models showed no systematic bias in LNneg microarchitectural features within a patient. The ICC revealed moderate variability for lymphocytes (ICC: 0.39; 95 %CI: 0.01– 0.61, p = 0.02)) and GermC (ICC: 0.50; 95 %CI: 0.22–0.68, p < 0.001), and high variability for histiocytes (ICC: 0.07 (95 %CI: −0.45–0.40, p = 0.38). Bland-Altman plots showed that 5.0 % of GermC, 5.0 % of histiocytes and 8.5 % of lymphocyte measurements were outside the 95 % limits of agreement. Conclusions This is the first study to systematically assess agreement of microarchitectural features in LNneg within an individual (y)pN0 OeC patient. The absence of systematic bias supports using largest LNneg as surrogate for OeC patient’s overall anti-tumour immune response.