Impact of postoperative chemotherapy on survival for oesophagogastric adenocarcinoma after preoperative chemotherapy and surgery.

Oesophageal (gullet) and gastric (stomach) cancers are exceptionally hard to treat¹. Unfortunately, the United Kingdom has the highest incidence of  oesophageal adenocarcinoma in the world², meaning that more than 9000 patients every year are faced with this diagnosis. For most patients, by the time their cancer produces symptoms, it will have already spread, meaning curative treatment cannot be considered. For the minority (about 1/3)³ who have cancer confined to the oesophagus or stomach and local lymph nodes, a gruelling multi-pronged course of treatment is usually necessary, incorporating multiple rounds of chemotherapy and a highly invasive operation to remove the oesophagus and/or stomach. This takes a number of months to complete and recovery can take well over a year.

Multiple clinical studies have proven the benefit of chemotherapy and surgery over surgery alone ^4–7, but only about 20% of patients derive a survival benefit from the chemotherapy and we cannot predict before the start of treatment who they will be^8,9. Perioperative chemotherapy (a course of chemotherapy, followed by surgery, followed by a further course of chemotherapy) is currently the treatment of choice for these patients, but because recovery from surgery takes many months, even in the best clinical trials, with highly controlled and sanitised conditions, less than half of patients actually received chemotherapy after their surgery^5,7. Consequently, we don’t know if giving additional chemotherapy after surgery actually makes any difference to survival.

This leaves a fundamental unanswered question that clinicians and patients wrestle when considering treatment options:

• Should we give patients chemotherapy after surgery when they have already received chemotherapy before their operation?

In our recent study published in the BJS, we aimed to answer this question using pragmatic, real-world population-level data from the National Oesophagogastric Cancer Audit (NOGCA) of England and Wales. We set out to quantify the benefit, if any, of giving postoperative chemotherapy in this setting.

The NOGCA is a mandated registry of patients diagnosed with oesophageal and gastric cancers in England and Wales and has been collating data since 2012. It is a robust and mature dataset, and this is combined with other national sources of cancer data to enhance its richness. Recently, it has been used to conduct outcome modelling studies for oesophageal and gastric cancer patients^10,11, and the available data was sufficiently detailed to answer this question.

In the absence of randomisation of treatment, we adjusted for differences between patients who received post-operative chemotherapy and those who did not using a technique called ‘propensity score weighting’^12,13. Essentially, this creates a cohort of patients where patient, treatment and disease characteristics were well matched between those who did and did not receive chemotherapy after surgery.

We included 4,139 patients who underwent planned perioperative chemotherapy (platinum-based triplet or FLOT) and surgery for adenocarcinoma of the oesophagus or stomach, of whom 1,539 (38.5%) received post-operative chemotherapy (fewer than in the randomised studies, reflecting real-world practice). We found that an important determinant of the receipt of postoperative chemotherapy was if the patient suffered complications from surgery – with patients who did not receive chemotherapy having a substantially higher rate of all cause complications (37.0 vs 24.9%, p<0.001), pulmonary (chest) complications (17.0 vs 11.3%, p<0.001) and anastomotic leak (disruption of the join between oesophagus and stomach/small bowel) (7.4 vs 2.1% p<0.001). Patients who received additional chemotherapy were also younger and fitter, but importantly, tumour biology and staging were similar between the groups.

What was the answer to our research question? Well, we found that administering postoperative chemotherapy after preoperative chemotherapy and surgery resulted in a substantially improved overall survival from a median of 50.4 months to 62.7 months (p=0.004), a clinically meaningful amount. Putting this into context of recent practice-changing trials in oesophageal and stomach cancer reveals a similar improvement in survival to that yielded by the most effective chemotherapy regimen, FLOT-4⁷ and the addition of post-operative immunotherapy to standard-of-care treatments in the Checkmate 577¹⁴ trial.

Our study provides the strongest current evidence for postoperative chemotherapy in patients who have received preoperative chemotherapy and surgery for oesophagogastric adenocarcinoma, with a substantial survival benefit seen after adjustment for all possible factors.

We found that potentially modifiable factors influence if patients receive this treatment, specifically surgical complications, which could feasibly be reduced by increasing minimally invasive approaches or via centralisation of services/expertise. This study informs practice on a day-to-day basis in the UK and worldwide.