Richard Lehman’s journal review—15 May 2017

Richard Lehman reviews the latest research in the top medical journals

richard_lehmanNEJM  11 May 2017  Vol 376

Tracking cancer evolution

Since Vinay Prasad and Bishal Gayawali blazed onto the oncology scene, it’s been a relief for me to leave the subject in their capable hands. Through their close examination of the original research, they are able to tease out the true advances from the hype, and bring some sense of reality and indeed morality to the stenchy world of oncobillionairism. As Vinay put it in a tweet the other day, “I’ve no problem that much basic research is unlikely to result in therapies in near future. I do have a problem when many pretend otherwise.” Here’s a complex original study of patterns of evolution within non-small-cell lung cancer. Commendably, it does not end with a welter of speculation about exciting therapeutic possibilities. The final sentence reads: “In the analysis presented here, we provide a census of driver events in early-stage NSCLC in relation to clonality and show that chromosomal instability is not only a significant driver of parallel evolution, but also a predictor of poor outcome.” So the main gain here is scientific and prognostic. For an actual person diagnosed with NSCLC today there is no immediate benefit. Quite the opposite: if I had the condition I would not want to know the chromosomal instability of my tumour. However, if someone wanted to sample my cancer in the context of truly scientific, truly shared research for patient benefit in the future, I would gladly consent, replying “moriturus te salutat” (“One who is about to die wishes you good fortune”, as the gladiators said to Caesar). On the other hand, if I was asked by someone who wanted to make a lot of money out of the data, I might just be tempted to use my last gladiatorial opportunity for homicide.

JAMA  9 May 2017  Vol 317

Non-small-cell lung cancer treatment

So now let’s cross from the NEJM to JAMA and look at the latest randomised clinical trial in non-small-cell lung cancer. Not many trials of this kind appear in JAMA these days, and those that do tend to be negative. I take this as an accolade to the journal’s editorial policy, as outlined in an article by the editors last week, and I suspect it has cost them a lot of reprint revenue from pharma-sponsored studies. Anyway, let’s go back to those poor human beings with advanced NSCLC. Their mean overall survival in this trial was 8 months. During this time they were given docetaxel plus placebo, with a 45% rate of adverse effects (grade 3 or above), or docetaxel plus selumetinib, with a 67% rate of severe adverse events. So we now know that among patients with previously treated advanced KRAS-mutant NSCLC, addition of selumetinib to docetaxel does not provide benefit over docetaxel alone. This is disappointing, since logically a mitogen-activated protein kinase (MEK) inhibitor might have been expected to work in the presence of this mutation. But perhaps those chromosome instabilities had gone too far by this point, and lethal clones were multiplying in all directions. Maybe by this stage palliative care would be kinder than docetaxel, and more life-prolonging.

FDA approved does not mean safe

Last week, an article in The BMJ led by a Yale medical student showed how few drugs and devices fast-tracked by the US Food and Drug Administration ever received proper validation by subsequent studies. Here’s another FDA-related paper from the same team this time the lead author is Nick Downing, who previously got two articles into the leading journals as a Yale medical student. They sure look after them good over there at Yale. How unlike the FDA, which seems unable to do anything much to protect the interests of the American people: “Among 222 novel therapeutics approved by the FDA from 2001 through 2010, 71 (32.0%) were affected by a postmarket safety event. Postmarket safety events were more frequent among biologics, therapeutics indicated for the treatment of psychiatric disease, those receiving accelerated approval, and those with near–regulatory deadline approval.” I guess I should explain, for those old enough to remember when words kept to their original meaning, that a “safety event” means a danger event severe enough to warrant public airing. This amazingly thorough review only stopped looking at the end of February 2017. From now on the USA is just one big safety event. One can only hope that the Trump Event falls short of total national catastrophe, but gets close enough to it to act as a warning for the rest of the century.

JAMA Intern Med  May 2017  Vol

Thyroid “cancer”

In a lively commentary on the US Preventive Services Task Force’s decision to recommend against routine screening for thyroid cancer, Gil Welch writes, “How would I summarize the effect of thyroid cancer screening? Massive overdiagnosis with no change in mortality.” About the massive overdiagnosis there can be no argument. But I was given pause for thought about mortality by an article in JAMA a few weeks ago.

In contrast with Gil’s graph, this reached the conclusion that “Among patients in the United States diagnosed with thyroid cancer from 1974-2013, the overall incidence of thyroid cancer increased 3% annually, with increases in the incidence rate and thyroid cancer mortality rate for advanced-stage papillary thyroid cancer. These findings are consistent with a true increase in the occurrence of thyroid cancer in the United States.” Both could be correct, depending on definitions and the scale of the subgroup mortality increase. But the elephant in this room is the histological diagnosis of thyroid cancer. How is it that in 2017 we can follow the genomic patterns of so many cancers with exquisite precision and yet utterly fail to distinguish between harmless and progressive forms of thyroid “cancer”? And it’s the same for PSA-detected prostate cancer or intraductal carcinoma of the breast.

Unequal in life and death

With an election looming that may spell the end of the NHS, fracture with Europe, economic decline and growing inequality, take a look over the Atlantic. Geographical disparity in life expectancy in the USA increased steadily between 1980 and 2014. There is now a 20-year gap in life expectancy between the highest and lowest performing counties there. Socioeconomic and race/ethnicity factors may account for 60% of this, and healthcare provision for another 27%. Life expectancy in the last five years has already declined in the UK, but not in an even manner. You have seen the future, and it kills.

The Lancet 13 May 2017  Vol 389

Dupilumab for eczema

Over the last two years, dupilumab has acquired the status of a breakthrough drug for atopic eczema. Its manufacturer, Sanofi and Regeneron, now expects to make at least $4billion from it over the next few years, and NICE has already given it early access status, though its general UK approval and pricing have still to be fixed. In the USA, the figure of $30K per patient per annum is doing the rounds. Dupilumab is a fully human monoclonal antibody that binds to interleukin 4Ra, a component of the interleukin-4 and interleukin-13 receptors, which inhibits their signalling and downregulates type 2 immunity. From the LIBERTY AD SOLO and LIBERTY AD CHRONOS trials (don’t ask me what these mad acronyms stand for) it’s clear that dupilumab quickly reduces the most troublesome symptom of eczema: itching. From the latter trial it’s also clear that the drug is safe over one year and in fact reduces the incidence of skin infection. But in other aspects it is pretty impossible to judge its effectiveness, given that patients were free to use topical corticosteroids at the same time and the flow-chart shows a complex pattern of successive drop-outs up to and beyond the open-label follow-up phase. Here is an agent which may bring relief to millions of people but at enormous cost to health systems. Almost all the evidence comes from studies funded by the manufacturer, in which “funders participated in the conception and design of the study, analysis and interpretation of the data, drafting and critical revision of the report, and gave approval to submit.” I know I say this every week, but is that the way things would operate in a world which prioritized equitable and affordable healthcare for everyone?

The weekend non-effect

Among the many ironies of the latest study of the “weekend effect” in NHS hospitals is that the Department of Health funded it and paid for it to be open access. Well done Tim Peto, the Oxford University NHS consultant who brought this off. Too late to undo the irreparable damage of course, but his team have shown that: “Adjustment for routine test results substantially reduced excess mortality associated with emergency admission at weekends and public holidays. Adjustment for patient-level factors not available in our study might further reduce the residual excess mortality, particularly as this clustered around midday at weekends. Hospital workload was not associated with mortality. Together, these findings suggest that the weekend effect arises from patient-level differences at admission rather than reduced hospital staffing or services.” If I bump into David Cameron at the next Chipping Norton Farmers’ Market, I must remember to try and explain residual confounding and case-mix to him, amongst banter about the fresh seafood that Jack has brought up from Selsey this time.

The BMJ 13 May 2017  Vol 357

NSAIDs and infarcts, again

When rofecoxib (Vioxx) was withdrawn from sale in 2004, a lot of us became interested in the cardiovascular safety of other oral non-steroidal anti-inflammatory drugs. The evidence at that point was scrappy, but by 2008 it was clear that diclofenac was associated with a quadrupling of risk for myocardial infarction. Two things happened around then: Neal Maskrey at the UK National Prescribing Centre set about disseminating the evidence via prescribing advisers to every GP practice, resulting in a very large drop in prescriptions for oral diclofenac; while at the same time a different government agency authorized over-the-counter sale of diclofenac tablets to the public. The prescribing centre was wound down and became part of NICE, while OTC sales of diclofenac continued to 2015. Move on to 2017, and here is an systematic review of the evidence including a bayesian meta-analysis of individual patient data. “All NSAIDs, including naproxen, were found to be associated with an increased risk of acute myocardial infarction. Risk of myocardial infarction with celecoxib was comparable to that of traditional NSAIDS and was lower than for rofecoxib. Risk was greatest during the first month of NSAID use and with higher doses.” Unmasking the cardiac risk of NSAIDs would make for a great lecture on the erratic dissemination of evidence-based medicine.

Guidelines you can really use

While we’re on to the erratic dissemination of evidence-based medicine, let me do another old-man-looks-back job on knee arthroscopy. This gained an enormous vogue from the 1990s onwards. You could spot lesions and treat them and often the symptoms got better afterwards. Or you could just wash the joint out and point out little bits of debris in the fluid. The trouble came when some orthopaedic surgeons started to do sham-controlled trials. To their credit, they have become very good at these. And given the results, I don’t think many patients with knee osteoarthritis or meniscal tears would opt for arthroscopic surgery. I’m truly delighted that The BMJ is continuing to develop interactive infographics which show the true effects of interventions like arthoscopy, called Rapid Recommendations. These are fast approaching the point where they could be shared with patients.

Montgomery: shared decision making as the law

Sharing knowledge and choices about interventions has become a legal requirement throughout the UK since March 2015. The case of Montgomery v Lanarkshire Health Authority was heard in the Supreme Court of the UK and has been written about extensively ever since, though rarely invoked in actual court actions. Here’s a very useful summary of the current position. I like the sentence “In reality, medical decision making involves a nuanced negotiation of information.” Dumping every kind of possible risk on the patient just to cover your back legally is the opposite of true shared decision making.

Plant of the Week: Paeonia rockii

“To ride in the mountains of North China when the tree peonies are in bloom is to taste paradise” wrote EH Wilson in 1913. He was talking about the sight and scent of whole valleys filled with P rockii, thought to be the common ancestor of all the innumerable tree peonies hybridized over a millennium in China and Japan. In “Chinese” Wilson’s time, it grew as a scrub plant over hundreds of square miles and was a popular local source of firewood.

It is thought that the first example of rockii arrived in Europe on the ship Hope captained by James Prendergast in 1802. But the flouncier hybrids which flooded Europe after the Opium Wars in the mid-nineteenth century drove it out of known Western cultivation until interest was renewed by Wilson’s account:

“Through the foaming shallows of the copse I plunged, and soon was holding my breath with growing excitement as I neared my goal, and it became more certain that I was setting my eyes on Paeonia Moutan as a wild plant. The event itself justified enthusiasm, but all considerations of botanical geography vanish from one’s mind in the first contemplation of that amazing flower, the most overpoweringly superb of hardy shrubs. Here in the brushwood it grew up tall and slender and straight, in two or three unbranching shoots, each one of which carried at the top, elegantly balancing, that single enormous blossom, waved and crimped into the boldest grace of line, of absolute pure white, with featherings of deepest maroon radiating at the base of the petals from the boss of golden fluff at the flower’s heart. Above the sere and thorny scrub the snowy beauties poise and hover, and the breath of them went out upon the twilight as sweet as any rose. For a long time I remained in worship…”

On journeys from 1922 onwards, the Austrian-born Joseph Rock set out from the Arnold Arboretum in Massachusetts to retrace Wilson’s footsteps, but it was not until the 1930s that he sent back seed from wild or half-cultivated tree peonies in Gangsu. It was a time of great civil unrest in China and accounts vary of where these seeds originated and where they ended up.

Through a mix-up at the supplier, we got five tiny dried up bits of root from an alleged specimen of P rockii about six years ago. We’ve grown them on patiently ever since, giving away three to friends. There was never any certainty about their true origin, so it was with joy that we beheld the first flowers on two of them appear this year. Chinese Wilson’s description is spot on. For a long time we remained in worship…


  • kidmugsy

    “With an election looming that may spell the end of the NHS”: what, another one? This feeble trope has been trotted out at every election for decades. One might almost think that vested interests are protecting their pockets, their egos, or their power-bases.

  • George

    There’s a parallel NSAID story in the New Zealand experience; use of Ibuprofen, and intramuscular injection of diclofenac (then prescription only) were associated with what was then a rare, novel (and lethal) disease, necrotizing fasciitis. A government warning was issued in 2000.
    In 2004, a different branch of government approved Ibuprofen for OTC sale, diclofenac followed around this time.
    Of course paracetamol was always an OTC drug and was responsible for ~200 deaths a year in New Zealand prior to the introduction of Parvolex and glutathione, whereas a year often went by without a fatal heroin or morphine overdose.