Gastrostomy versus nasogastric tube feeding for chemoradiation patients with head and neck cancer: the TUBE pilot RCT

Background Approximately 9000 new cases of head and neck squamous cell cancers (HNSCCs) are treated by the NHS each year. Chemoradiation therapy (CRT) is a commonly used treatment for advanced HNSCC. Approximately 90% of patients undergoing CRT require nutritional support via gastrostomy or nasogastric tube feeding. Long-term dysphagia following CRT is a primary concern for patients. The effect of enteral feeding routes on swallowing function is not well understood, and the two feeding methods have, to date (at the time of writing), not been compared. The aim of this pilot randomised controlled trial (RCT) was to compare these two options. Methods This was a mixed-methods multicentre study to establish the feasibility of a RCT comparing oral feeding plus pre-treatment gastrostomy with oral feeding plus as-required nasogastric tube feeding in patients with HNSCC. Patients were recruited from four tertiary centres treating cancer and randomised to the two arms of the study (using a 1 : 1 ratio). The eligibility criteria were patients with advanced-staged HNSCC who were suitable for primary CRT with curative intent and who presented with no swallowing problems. Main outcome measures The primary outcome was the willingness to be randomised. A qualitative process evaluation was conducted alongside an economic modelling exercise. The criteria for progression to a Phase III trial were based on a hypothesised recruitment rate of at least 50%, collection of outcome measures in at least 80% of those recruited and an economic value-of-information analysis for cost-effectiveness. Results Of the 75 patients approached about the trial, only 17 consented to be randomised [0.23, 95% confidence interval (CI) 0.13 to 0.32]. Among those who were randomised, the compliance rate was high (0.94, 95% CI 0.83 to 1.05). Retention rates were high at completion of treatment (0.94, 95% CI 0.83 to 1.05), at the 3-month follow-up (0.88, 95% CI 0.73 to 1.04) and at the 6-month follow-up (0.88, 95% CI 0.73 to 1.04). No serious adverse events were recorded in relation to the trial. The qualitative substudy identified several factors that had an impact on recruitment, many of which are amenable to change. These included organisational factors, changing cancer treatments and patient and clinician preferences. A key reason for the differential recruitment between sites was the degree to which the multidisciplinary team gave a consistent demonstration of equipoise at all patient interactions at which supplementary feeding was discussed. An exploratory economic model generated from published evidence and expert opinion suggests that, over the 6-month model time horizon, pre-treatment gastrostomy tube feeding is not a cost-effective option, although this should be interpreted with caution and we recommend that this should not form the basis for policy. The economic value-of-information analysis indicates that additional research to eliminate uncertainty around model parameters is highly likely to be cost-effective. Study limitations The recruitment issues identified for this cohort may not be applicable to other populations undergoing CRT. There remains substantial uncertainty in the economic evaluation. Conclusions The trial did not meet one of the three criteria for progression, as the recruitment rate was lower than hypothesised. Once patients were recruited to the trial, compliance and retention in the trial were both high. The implementation of organisational and operational measures can increase the numbers recruited. The economic analysis suggests that further research in this area is likely to be cost-effective. Future work The implementation of organisational and operational measures can increase recruitment. The appropriate research question and design of a future study needs to be identified. More work is needed to understand the experiences of nasogastric tube feeding in patients undergoing CRT. Trial registration Current Controlled Trials ISRCTN48569216. Funding This project was funded by the National Institute for Health Research (NIHR) Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 22, No. 16. See the NIHR Journals Library website for further project information.