OGC P35 Reporting of core outcomes in gastric cancer surgery trials over the past 25 years

Abstract

Abstract Background Gastric cancer presents a significant health burden as the fifth most common malignancy and a significant cause of cancer-related mortality worldwide. A multi-modal approach has led to modest improvement in survival in recent years but surgery remains the cornerstone of curative treatment. A variety of surgical approaches are available, including newer minimally-invasive techniques: laparoscopic and robotic surgery. Determining the optimal approach requires appraisal of high-quality evidence, such as randomised controlled trials (RCTs) and systematic reviews with meta-analyses. However, heterogeneity within the literature hinders this, allowing little consensus. Surgical trials often report different outcomes and even the same outcome may be defined or measured differently. This challenge spans specialties and has been addressed by the development of core outcome sets (COS): a standardised set of outcomes that should be measured and reported as a minimum. The GASTROS group developed a COS for gastric cancer surgical trials. This process involved key stakeholders, patients and clinicians, better reflecting their priorities. These core outcomes are disease-free survival, disease-specific survival, surgery-related death, recurrence, completeness of tumour removal, overall quality of life, nutritional effects, and ‘serious’ adverse events. The aim of this review was to determine what proportion of trials report these core outcomes. Methods This review was an update of an earlier systematic review by our group spanning years 1996 to 2016. We systematically reviewed databases Evidence Based Medicine Reviews, MEDLINE, EMBASE and CINAHL from the years 2016 to 2021, specifically focusing on ‘type 2’ surgical RCTs: trials comparing two or more therapeutic surgical interventions. We separately reviewed systematic reviews of RCTs. The data extracted was outcomes reported, definitions, measuring tools used and when outcomes were measured. Results A total of 2213 abstracts were screened and 295 included. After full paper review this yielded 53 relevant articles. A further 47 articles from our previous systematic review were included. Twenty-two systematic reviews of RCTs were also identified. Of the 101 trials, no trial reported all core outcomes. Serious adverse events were reported by nearly all trials, 98.4%, but these were reported heterogeneously; as pooled and individual complications and graded using different severity tools. Other commonly reported outcomes were surgery-related death, 85.3%, and completeness of tumour resection, 73.8%. Survival outcomes were often reported as overall survival. Only 54.1% reported disease-free survival 54.10% and even fewer, 4.9%, reported disease-specific survival. Tumour recurrence was reported by just over a third of trials. Outcomes likely to be most important to patients, quality of life and nutritional effects, were reported least at 23% and 16.4% respectively. Over the past 25 years there has been no significant improvement in the reporting of core outcomes. An expectedly similar spread of outcome reporting was seen in systematic reviews of RCTs, with serious adverse events and surgery-related death reported most and quality of life, nutritional effects and survival outcomes reported least. Conclusions Surgical trials must report outcomes relevant to key stakeholders, importantly patients as well as surgeons. This review demonstrates that outcomes from the COS are poorly reported at present. The value of the COS needs to be disseminated to the wider research community in order to increase its uptake. As part of this, we intend to contact ongoing trials with the results of this review. Consistent reporting of core outcomes in future will allow for meaningful appraisal of evidence, ultimately improving clinical outcomes.