Standards of outcome reporting in surgical oncology: a case study in esophageal cancer.

BACKGROUND

Multimodal strategies before surgery are often used to improve outcomes, but disease progression (precluding surgical resection) and inoperability at planned surgery still occur following neoadjuvant treatment. The standards of reporting of these outcomes have not previously been considered. This study examined reporting of rates of progression to surgical resection and inoperability at planned surgery following neoadjuvant treatment in surgical oncology, using esophageal cancer as a case study.

METHODS

A systematic review identified randomized trials and prospective nonrandomized studies reporting short-term outcomes of neoadjuvant treatment and surgery for esophageal cancer.

RESULTS

Of 4,763 abstracts, 224 papers were retrieved and 76 studies included (8 randomized trials and 68 cohort studies of 19,441 esophagectomies). Articles reported outcomes of preoperative chemotherapy (n = 33, 43.4 %), chemoradiotherapy (n = 13, 17.1 %), or both within one paper (n = 18, 23.7 %) and 12 (15.8 %) did not specify the type of neoadjuvant treatment. Also, 20 papers (26.3 %) reported numbers of patients not progressing to surgery after neoadjuvant treatment (with rates of nonprogression ranging between 2.0 and 35.3 %). In addition, 24 papers (31.6 %) reported rates of inoperability at planned surgery (with inoperability rates ranging between 0 and 26.2 %). More randomized controlled trials (RCTs) than observational studies reported nonprogression (4 randomized and 16 nonrandomized studies, 95 % CI -9.6 to 62.6 %, p = 0.108) and inoperability (6 randomized trials and 18 observational studies, 95 % CI 16.8-80.3 %, p = 0.005). Some 17 and 10 articles provided reasons for the observed rates of nonprogression and inoperability, respectively.

CONCLUSIONS

Reporting rates of progression to surgery after neoadjuvant treatment and inoperability at planned surgery for esophageal cancer were poor, limiting data synthesis and comparisons. It is suggested that core outcome sets for trials in surgical oncology are developed with inclusion of these important endpoints. Collaboration between medical and surgical oncologists is necessary to achieve this.