NEJM  20-27 Aug 2015  Vol 373

726   We start with a basket trial. Say you are in a supermarket and put lots of brown things in your basket—bread, a joint of lamb, a tin of brown beans, some kiwi fruit, and a shirt. Now, out of scientific curiosity, you decide to dip the contents of your basket in anti-brown, a corrosive substance that binds to anything that’s brown. You find that the bread dissolves completely, the lamb turns blue, the kiwi fruit explodes, the shirt shrinks and the beans remain unaffected. The actual basket trial described here was only slightly more sophisticated. It gathered together 120 patients with various forms of non-melanoma cancers, all of which expressed the oncogene BRAF V600. They were all given a selective oral inhibitor of the BRAF V600 kinase, vemurafenib. The short-term responses were then observed. “Preliminary vemurafenib activity was observed in non–small-cell lung cancer and in Erdheim–Chester disease and Langerhans’-cell histiocytosis… There were anecdotal responses among patients with pleomorphic xanthoastrocytoma, anaplastic thyroid cancer, cholangiocarcinoma, salivary-duct cancer, ovarian cancer, and clear-cell sarcoma and among patients with colorectal cancer who received vemurafenib and cetuximab.”

737   Now let’s move to a carefully focussed randomised controlled trial in a single stage of a single cancer: precise clinical science rather than basketing. Here is something that will really change practice. Metastatic, hormone-sensitive prostate cancer is common and eventually lethal. At the moment, the best we can hope for is postponement of death, and this trial shows that can be achieved by adding docetaxel to standard androgen deprivation therapy. “After a median follow-up of 28.9 months, the median overall survival was 13.6 months longer with ADT plus docetaxel (combination therapy) than with ADT alone (57.6 months vs. 44.0 months; hazard ratio for death in the combination group, 0.61; 95% confidence interval [CI], 0.47 to 0.80; P<0.001). The median time to biochemical, symptomatic, or radiographic progression was 20.2 months in the combination group, as compared with 11.7 months in the ADT-alone group.”

795   In 1998, when I started writing these reviews, effective triple antiretroviral therapy for HIV was only a few years old and people were arguing about the best point to start it. They still are. But in the light of the latest INSIGHT START trial it seems clear that it is a bad idea to wait until the CD4+ has gone below 500, let alone 350. In a total of 4685 patients were followed for a mean of 3.0 years a policy of starting at a point above 500 produced clearly better outcomes than waiting until 350 or the development of symptoms. The primary composite end point was any serious AIDS-related event, serious non–AIDS-related event, or death from any cause.

834   In the 1990s I also did a little study of B-type natriuretic levels in patients presumed to have heart failure in the community. This uncovered a couple of people who on further investigation turned out to have pulmonary hypertension, an even more lethal diagnosis. In this condition the BNP comes mostly from the right ventricle and can be used to monitor treatment, which is the case in this trial of ambrisentan with tadalafil for pulmonary hypertension. Most of us are familiar with tadalafil (Cialis), while ambrisentan is a new endothelin-A–receptor antagonist. Together, they slightly slow the progress and mitigate the effects of this nasty condition, but don’t hold much promise otherwise.

845   In Britain there is a sharp distinction between physicians and surgeons. Surgeons are supermen, and refuse to accept the title “doctor” for fear of being confused with lower forms of life who merely talk to people and prescribe drugs. In North America, though, surgeons are often referred to as “physicians” and that is the case in this study of the effect of sleep deprivation on performance the next day. I will hand you over to Druin Burch, a friend and consultant physician, who has beaten me to the task of commenting on this study: “An intriguing study from Canada showing that outcomes for daytime surgery were identical regardless of whether or not the surgeon had been working through the night. There are a couple of possible explanations, both pleasant for the physician to consider. Firstly, the obvious one: surgery is very simple and any fool can do it, even if they’re half asleep. Secondly, surgeons are insensitive to hardship and should be made to work very much harder than their physician colleagues. I regard this paper as evidence-based proof of the superior mental performance required of those in physicianly specialities. It is a welcome finding.”

JAMA  18 Aug 2015  Vol 314

677   I don’t keep up much with the gossip about medical journals, but I’m aware that some time ago, JAMA abandoned its policy of requiring independent statistical reviews of papers related to industry-funded trials. I was also perturbed to hear Howard Bauchner, the editor, give a lofty account two years ago of the reasons why his journal had doubts about signing up to AllTrials. I’m not sure if any of that has changed, but it’s amusing to find that most of the pharma-funded trials that JAMA publishes recently do not favour the tested product and are therefore unlikely to bring in much income from reprints. In this one, Leo Pharma went to the trouble of recruiting 900 patients with cancer and an episode of venous thromboembolism in 164 centres in Asia, Africa, Europe, and North, Central, and South America, hoping to prove that its product tinzaparin would prove more effective than warfarin in preventing recurrent VTE over the next 6 months. But it wasn’t.

687  The next trial might sell some reprints, though it hardly brings new news. “Novo Nordisk was involved in the study design and protocol development, provided logistical support, and obtained the data, which were evaluated jointly by the authors and the sponsor. All authors interpreted the data and wrote the manuscript together with the sponsor’s medical writing services.” Novo Nordisk’s newish blockbuster is liragutide, now licensed in the USA to reduce weight in non-diabetics. This trial shows that it reduces weight in people with type 2 diabetes over a period of 56 weeks. And that “further studies are needed to evaluate longer-term efficacy and safety.” Yes indeed.

767   The outlook for developing a better informed, more humane form of medicine could not be brighter. And I’m thrilled at the number of talented younger doctors who are taking up the challenge. It comes in two forms. Firstly there is the challenge of how best to share knowledge and decisions with patients. Secondly there is the challenge of how to generate uncontaminated evidence as quickly as possible in ways that can directly inform and improve clinical practice. Fortunately we’re getting past the stage of moaning that the knowledge architecture of medicine is broken, and looking at actually creating something new—not a static architecture but a living, adaptive organism where all the bits are joined up. Here’s an article about that. It’s called “Fusing Randomized Trials With Big Data:  The Key to Self-learning Health Care Systems?” Derek Angus surveys the possible sources of knowledge we now have and declares that “The grandest leap would be to fuse all these elements in a new kind of RCT that can be called a randomized, embedded, multifactorial, adaptive platform (REMAP) trial.”

781   Keeping active in old age is the subject of Muir Gray’s latest book, Sod Seventy. Although I meet Muir in the corridor most weeks, I haven’t yet read his book. I have another five years to go before I reach 70, if I ever do. I am by nature idle and sedentary. Grown a little older, I would have been a good subject for the LIFE trial which randomised people of a static disposition aged 70-89 to either a 24 month moderate-intensity physical activity program or a health education program and measured the effect on cognitive function. Contrary to expectation (and preaching) there was no cognitive difference between the groups at the end of the trial.

JAMA Intern Med  Aug 2015  Vol 175

OL Deprescribing is another plausible remedy for the ills of ageing, and one that I frequently advocate. The last of John Yudkin’s Ten Commandments runs “Honour thy elderly patient, for although this is where the greatest levels of risk reside, so do the greatest hazards of many treatments.” Here’s a Dutch trial called DANTE, in which they observed the effect of stopping blood pressure lowering medication in older people with mild cognitive deficits. Over 16 weeks, it made no difference to their cognitive and general function. But then would you expect it to? I’d be more interested in how many falls they had over two years.

Lancet  29 Aug 2015  Vol 386

857   The Ebola epidemic is over so we can all relax and forget about places like Sierra Leone and Guinea. Sadly, that is probably what will happen, though perhaps Bill & Melinda Gates will step in where the World Health Organisation is so pathetically ineffective, and take measures to prevent another outbreak. Vaccination may be vital in this. Moan as we may about the logistic failures of the Ebola containment process, the outbreak has illustrated how swiftly vaccine development can be achieved under pressure. Here’s a preliminary report of a trial of a recombinant, replication-competent vesicular stomatitis virus-based vaccine expressing a surface glycoprotein of Zaire Ebolavirus (rVSV-ZEBOV). The last data came in on July 20th and the report was published online on August 3rd. They indicate that rVSV-ZEBOV might be highly efficacious and safe in preventing Ebola virus disease, and is most likely to be effective at the population level when delivered during an Ebola virus disease outbreak via a ring vaccination strategy.

884   It’s odd that the British journals seem to have found a sudden interest in American outcomes research: something to do with US sales, I expect. It’s often hard to know if what happens to Medicare patients over there is any guide to what happens to NHS patients here. This study looked at data from 9.5 million Medicare patients who underwent 12 types of major surgery in the USA, and looked at the rates of readmission for each, which varied between 5.6% (knee replacement) and 21.9% (oesophagectomy). Overall about 75% of readmissions were to the original (“index”) hospital, and such patients had a lower mortality than those readmitted to a non-index hospital. I’m not sure what this really tells us. That sicker patients tend to be directed to other centres? Or that you’re safer with a surgeon who already knows you?

BMJ  22 Aug 2015  Vol 351

It’s fifty years since I learnt biology at the foot of a teacher called Mr Fordham who was towering in every sense. Just now I’m reading The Vital Question by Nick Lane: a brilliant and engrossing book, and one that leaves me even more awed and grateful for the grounding I got in 1965. There’s scarcely a question he raises that we Sheffield grammar school boys weren’t made familiar with back then: even the quantum nature of membrane energy transfer, and above all the fundamental importance of environmental interactions, especially at the level of micro-organisms. Now here’s a massively painstaking network meta-analysis of treatments aimed at eliminating Helicobacter pylori in adults. Half the analysis workforce is made up of doctoral students from China. It’s a really comprehensive and thoughtful analysis and they fully deserve the career boost which is bound to come from getting published in the BMJ. But H pylori, living at the very spout of the proton pumps in the stomach, is nothing if not a niche bacterium. Killing it is bound to depend on what it has had to contend with already. In an environment where every second person is already on omeprazole and gets regular amoxicillin, those agents aren’t likely to be very useful. This article is great at showcasing the range of treatments available, but predicting what will actually work for individual patients in particular localities is likely to be far more tricky.

I commented on the difference between assisted dying and voluntary euthanasia last week, and also touched on some figures from Belgium and the Netherlands. I won’t add anything here except to say that anyone who wants information about what is really happening around the world needs to read this very useful briefing paper.

Plant of the Week: Echinops bannaticus “Blue Globe”

These globe flowers are quite thuggish and very good at spreading themselves by pushing other plants out of the way and by seeding. Their flowers are like little hedgehogs (hence “echinops”) and despite breeders’ efforts to create clear blues, tend to be greyish. I got impatient with our main clump and dug it out a couple of years ago.

But then two seedling plants arose, one at the foot of a telegraph pole and the other at the foot of a wisteria. And these, for the present, seem to be doing quite a good job. They flower abundantly for many weeks in late summer and one of them is cheek by jowl with the wonderful rose called Absolutely Fabulous. This rose keeps producing an abundance of scented buff-yellow flowers, which are perfectly set off by those somewhat lugubrious spiky balls of blue-grey. So perhaps Echinops bannaticus is worth some garden space after all, but be sure to place it with something equally tall and a bit more cheerful.