HPB O01 Colorectal Liver Metastases and optimal follow-up: a systematic review

Abstract Background Colorectal cancer (CRC) is the third most common cancer worldwide. Approximately 42 000 new cases of CRC are diagnosed each year in the UK (1) (2). At diagnosis many patients (20–25%) already have colorectal liver metastases (CLRM) and further patients develop CRLM after their primary cancer is detected. Evidence suggests that survival rates for patients undergoing hepatectomy for CRLM are improving, however disease recurrence is still a problem with up to 70% recurring (3,4). Repeat resection has been shown to benefit patients with survival comparable to original hepatectomy (5,6). This highlights the importance of having a sensible and robust follow up plan for patients after hepatectomy for CRLM. On review of the literature, currently there is no clear consensus or recommendation surrounding the methods, intensity and duration of follow up for these patients. The aim of this systematic review was to report contemporary follow-up practices after curative hepatic resection for colorectal liver metastases and identify any potential oncological benefits. Methods A systematic review was conducted using PRISMA guidelines. Inclusion and exclusion criteria were set and searches were conducted between inception until 1st August 2020. Articles reporting follow up and outcomes for patients undergoing hepatectomy for CRLM were included. Data extraction was undertaken by two independent authors (AJ/AF) by reviewing full texts, any discrepancy was resolved by the senior author (SP). Outcomes included general demographics, method, frequency and duration of follow up. Reported survival and recurrence data was also collected (overall, disease-free and progression-free survival at 1, 3 and 5 years). Any further details of follow up recorded and these were classified as secondary outcomes, these included co-ordination with local hospitals, cost efficiency, patient reported outcomes and adaptability. Cost standardisation was performed using the Bank of England inflation calculator to account for the different years papers were published. Quality assessment was also performed using MINORS and CHEERS assessment methods. Results Twenty-two articles, totalling 8,384 included patients, published between 1994 and 2017 were included for data extraction. The most commonly used method of follow up was a CT scan (n=13), followed by physical examination (n=12). The most commonly used biochemical marker was CEA (n=11). The intensity of follow up was greater during first two years for majority of the articles. There was variation in the reported outcome measures, most commonly reported was overall survival, which ranged from 39% to 78.1% in the seventeen articles that reported it. The range of 1, 3, and 5 year survival were 63–94.7%, 45–65.1%, and 20–68% respectively. Co-ordination with local hospitals was reported in six articles, cost efficiency in six articles and patient reported outcomes were only described in three. The quality assessment using MINORS tool showed range of 4–10 for non-comparative and 5–16 for comparative studies and the CHEERS assessment for cost analysis showed a failure to meet many of the criteria. Conclusions This review has demonstrated the low quality evidence base on which surveillance after liver resection is based. Also shown is the large variation between the studies with regards to method and frequency of follow up. Based on this review and taking into account two previous systematic reviews in 2004 and 2012, we can conclude that further retrospective data will not be able to adequately answer questions around survival benefits of surveillance. Therefore, high quality, prospective studies focussing on endpoints that matter to patients should be the focus of future research.