Q: Although you have practiced as a surgeon in the British National Health Service (NHS) for most of your career, how is it that you’ve been such an outspoken critic of the “21st C cures act” that was passed through congress last year? What possible relevance might this have for your NHS?
A: All first year medical students should be taught the raison d’être for the practice of their chosen profession. Simply stated this distills down to three principles: the maintenance of health (well-being) for those who are free of disease and, for those with life threatening disease, the improvement of quality of life (QoL) and length of life (LoL). Therefore for patients diagnosed with cancer, QoL and LoL should be the primary outcome measures for all randomized clinical trials (RCTs) of innovative treatment. All other outcome measures have to be considered surrogates that may or may not translate into improvements in the primary outcomes. These may be used for legitimate reasons, such as aborting a trial early if surrogates all point in the wrong direction, but more often than not these surrogates are used to replace the primary outcome measures in order to fast track the adoption of what look like “promising” new interventions.
At its best, even significant improvements in cancer specific mortality might be abrogated by deaths directly related to the toxicities of treatment. At its worst is the assumption that some third or fourth level indices of “activity” of the novel treatment are accepted as evidence sufficiently compelling to bring the treatment to market.
In a recent JAMA Internal Medicine study, Prasad et al used meta-analysis to study the association of surrogate end points and overall survival in oncology. In 52% of examined studies the correlation was low; 25% showed medium correlation: only 23% showed a high level. Well over half the time surrogate end points failed to impact the gold standard of overall survival. This disconnect will have profound influences on safety, efficacy, and cost of oncology drugs that is already reaching fever pitch after the passing of the U.S. Congressional “21st Century Cures” bill last year.
In an accompanying JAMA Internal Medicine editor’s note, Rita F. Redberg MD MSc commented, “We must reduce drug approvals based on unreliable surrogates and change practice when critical studies show no survival benefit… In our rush to find new effective treatments, we should not harm our patients with ineffective toxic ones.”
We recognize that surrogate end points continue to figure prominently in oncology studies, usually promoted as meaningful outcomes in desperate situations where it is better to have cancers respond than not respond, and it is better to live progression-free than with progression. But without QoL measures to account for the offsetting toxicity of such gains, and without financial data to account for the cost, we really don’t know what has been achieved for patients who ended up surviving for identical amounts of time.
As someone who has practised in the British NHS all his professional life why should that worry me? Sadly ill-considered lunacies from the USA often follow the prevailing winds to our shores. Last year Lord Saatchi tried to drive through the House of Lords a “Medical Innovations” bill that was considered a “Quacks Charter”, based on the same fundamental misconceptions. It was a close call but fortunately it came to nothing.