Grey areas and evidence gaps in the management of rectal cancer as revealed by comparing recommendations from clinical guidelines

Background: While the management of nonmetastatic and oligometastatic rectal cancer has rapidly evolved over the last few decades, many grey areas and highly debated topics remain that foster significant variation in clinical practice. We aimed to identify controversial points and evidence gaps in this disease setting by systematically comparing recommendations from national and international clinical guidelines. Methods: Twenty-six clinical questions reflecting practical challenges in the routine management of nonmetastatic and oligometastatic rectal cancer patients were selected. Recommendations from the ESMO, NCCN, JSCCR, Australian and Ontario guidelines were extrapolated and compared using a 4-tier classification system (i.e., identical/very similar, similar, slightly different, different). Overall agreement between guidelines (i.e., substantial/complete disagreement, partial disagreement, partial agreement, substantial/complete agreement) was assessed for each clinical question and compared against the highest level of available evidence by using the chi(2) statistic test. Results: Guidelines were in substantial/complete agreement, partial agreement, partial disagreement, and substantial/complete disagreement for 8 (30.8%), 2 (7.7%), 7 (26.9%), and 9 (34.6%) clinical questions, respectively. High level of evidence supported clinical recommendations in 3/10 cases (30%) where guidelines were in agreement and in 10/16 cases (62.5%) where guidelines were in disagreement (chi(2) = 2.6, p = 0.106). Agreement was frequently reached for questions regarding diagnosis, staging, and radiology/pathology pro forma reporting, while disagreement characterised most of the treatment-related topics. Conclusions: Substantial variation exists across clinical guidelines in the recommendations for the management of nonmetastatic and oligometastatic rectal cancer. This variation is only partly explained by the lack of supporting, high-level evidence.