Where is the evidence base for benefits of sentinel node biopsy in melanoma?BMJ 2013; 346 doi: http://dx.doi.org/10.1136/bmj.f675 (Published 06 February 2013) Cite this as: BMJ 2013;346:f675
- J Meirion Thomas, professor of surgical oncology1
Sentinel node biopsy (SNB) has been promoted by claims of a survival advantage after early lymphadenectomy in patients with a positive result. The pivotal paper states that SNB “identifies patients with nodal metastases whose survival can be prolonged by immediate lymphadenectomy.”1 An earlier paper by Morton and colleagues states, “Our data suggest a significant therapeutic benefit for immediate dissection based on identification of tumour-involved sentinel node.”2 Accurate statistical analysis3 does not support those statements, and the claim that SNB offers a disease-free survival advantage.1 An appeal on this matter to the National Institutes of Health in 2007 was upheld. Morton was asked to include distant disease-free survival as an endpoint in the next update of the trial.3 The final results of the first Multicenter Selective Lymphadenectomy Trial (MSLT-1) could have been published in May 2010. Why has this not happened?
Other central claims, such as sentinel node status being essential for accurate staging and early lymphadenectomy facilitating regional disease control,4 are equally inaccurate.
This pretence is sustained by vested interest and citation distortion. None of the main protagonists of SNB, who hold powerful positions in academia, learned societies, and journals, especially in the US, have ever cited the main publication challenging the statistical accuracy of MSLT-1.3 This has allowed presentation and publication selection, resulting in a biased message. Many protagonists have put all their clinical and research eggs in the SNB basket, and discrediting the subject would be highly inconvenient. To quote Upton Sinclair, “It’s difficult for a man to understand something if his salary depends on him not understanding it.” Torjesen is correct.5 Publish the final results of MSLT-1.
Cite this as: BMJ 2013;346:f675
Competing interests: None declared.
The full response is at www.bmj.com/content/346/bmj.e8645/rr/625654.