IELTS Academic Reading Practice Test 329

Shortly before Christmas in 1954, two surgical teams worked in adjoining operating rooms at the Peter Bent Brigham Hospital in Boston. In one, a healthy man was being prepared for an operation that would remove one of his kidneys; in the other, his twin brother, gravely ill with kidney failure, lay waiting to receive it. Nothing of the kind had succeeded before. Within weeks, newspapers on both sides of the Atlantic were reporting that a human organ had been transferred from one person to another and had begun to work.

Surgeons had been attempting to replace kidneys for half a century. In 1902 the Austrian surgeon Emerich Ullmann joined a dog's kidney to blood vessels in the animal's own neck and showed that it produced urine. Mathieu Jaboulay, working in Lyon four years later, attached kidneys taken from a pig and a goat to the blood vessels of two patients; neither lived for long. The first attempt to transfer a kidney between people was made in 1933 by Yuri Voronoy, using an organ from someone who had just died. Donor and recipient had incompatible blood groups, and the kidney never worked.

In the early 1950s, chronic kidney failure was effectively a death sentence. The artificial kidney, a machine that could clean a patient's blood, had been developed during the Second World War, but it was intended for short-term use in people whose kidneys might recover. Nobody had yet found a way of keeping a patient alive on such a machine for months or years, and the only prospect of long-term survival was a replacement organ. Repeated access to a patient's bloodstream was also difficult; a workable shunt did not appear until 1960.

The surgeon who took on the problem, Joseph Murray, had not set out to become a pioneer of transplantation. Trained in plastic surgery, he spent the war years at a military hospital in Pennsylvania, rebuilding the faces of soldiers who had been badly burned. Grafting skin onto these men taught him what no laboratory experiment could have shown so vividly: skin taken from another person was attacked and destroyed within a couple of weeks. He also knew of earlier reports that skin grafts exchanged between identical twins had survived. Back in Boston, he wondered whether a kidney might survive if donor and recipient were genetically identical.

The opportunity came in 1954, when a young man named Richard Herrick was admitted to the Brigham with kidneys that were failing. He had an identical twin, Ronald, who was willing to help. Identical twins develop from a single fertilised egg and therefore carry the same genes, so tissue from one is not recognised as foreign by the other. To be certain that the brothers really were identical, the team compared their blood groups and exchanged small skin grafts between them.

The proposal raised difficulties that had nothing to do with surgery. An operation of this kind — taking a healthy organ from someone who did not need it removed — had never been performed at the Brigham, and the hospital had to consider whether it could be justified at all. The question has returned to courts and ethics committees many times since, and consent from living donors is now covered by detailed rules.

On 23 December 1954, Hartwell Harrison removed Ronald's left kidney while Murray prepared the recipient in the next room. The organ was carried across and sewn into Richard's lower abdomen, its artery and vein joined to the large vessels supplying the leg and its ureter connected to the bladder. Richard's own diseased kidneys were left where they were; removing them would have added to the risk. Placing the new organ low in the pelvis has remained standard practice ever since.

The transplant worked almost at once. Richard's blood chemistry returned to normal, his appetite came back, and he was able to leave hospital and return to work. He married one of his nurses and lived eight more years, dying in 1963 after the disease that had destroyed his own kidneys returned in the new one. For the first time a patient had lived for years with an organ taken from someone else, and the operation had shown beyond doubt that the body would accept a kidney from an identical twin.

What the Boston case proved was that the technical problem of joining a kidney to a patient's circulation had been solved, and that rejection, not surgery, was the real obstacle. Murray's team went on to irradiate patients before operating in an attempt to suppress the immune response, and the arrival in the early 1960s of drugs that could hold rejection in check, notably azathioprine, made transplants between unrelated people possible. The first successful transfer of a kidney from a deceased donor followed in 1962. A further advance came in the 1980s, when cyclosporine, a drug obtained from a soil fungus, reduced rejection still further. Rejection has never been abolished, only held in check, and most transplant patients now take such drugs for the rest of their lives.

Murray shared the Nobel Prize in Physiology or Medicine in 1990 with E. Donnall Thomas, who had pioneered the transplantation of bone marrow. Kidney transplantation is now routine: tens of thousands are carried out each year, and in many countries it is cheaper, as well as more effective, than long-term dialysis. Donation from a living relative, once extraordinary, is now commonplace. Yet a shortage of organs remains the limiting factor everywhere, and the ethical questions that the Boston team had to answer in 1954 have not gone away.

Every medicine on a pharmacy shelf has survived a long process designed to answer two questions: does the treatment work, and is it safe? Neither question can be settled by theory or by a doctor's impression. The history of medicine contains many treatments that seemed reasonable and were later shown to be useless, and a smaller number that were actively harmful. The clinical trial, in which a treatment is given to some patients and compared with another approach, is the instrument that makes such judgements possible. Its design is not a technicality but the thing that gives a result its meaning.

Testing a new treatment normally proceeds through four stages. In the first, a small number of volunteers, often fewer than a hundred, receive the drug so that researchers can establish how it is absorbed, how long it remains in the body and what dose can be tolerated. The second gives it to a few hundred patients to look for signs of benefit and to gather more information about side effects. The third, and usually largest, compares the new treatment directly with the best existing option, or with a placebo, in hundreds or thousands of people. The fourth continues after a medicine has been licensed, when it is used far more widely and rarer harms may at last become visible.

Why not simply give the new treatment to everyone and watch? Because many conditions improve on their own, because patients who expect to get better often report that they have, and because the course of a disease varies from one person to another. Without a group that receives a different treatment, or an inactive substitute, there is no way of knowing what would have happened otherwise. Where an effective remedy already exists, it is unethical to withhold it, so the comparison is made against that remedy rather than against a placebo. The choice of comparator determines what a trial can claim: showing that a drug is better than nothing is very different from showing that it beats the standard treatment.

The most reliable way to form the groups is by chance. Participants are allocated, usually by a computer program, to one group or another without the doctors or the patients choosing. Random allocation tends to balance out the characteristics that might otherwise distort the result: age, severity of illness, and unknown factors that nobody has thought to measure. Concealing the allocation until the moment of assignment matters as much as the randomness itself, because a doctor who can see the next envelope may be tempted to open it.

Blinding is a further safeguard. In a double-blind trial neither the participants nor the doctors treating them know who is receiving which treatment, and the codes are broken only when the analysis is complete. This matters most when the outcome is subjective, such as pain or quality of life, since expectation alone can change what people report. Some trials cannot be blinded at all: an operation, a course of physiotherapy or a change of diet is obvious to everyone involved. In such cases the person who assesses the results may be kept unaware of the treatment given.

Before a single patient is enrolled, the researchers must decide exactly what they will measure. The primary outcome is the one on which the trial's conclusion will rest, and altering it once the data have been seen is regarded as a serious flaw. The strongest outcomes are those a patient can feel: survival, a heart attack, a stroke, the ability to walk. Trials that rely on such outcomes must therefore recruit far more people, because events of this kind are uncommon among those well enough to take part.

Trials are often too short or too small to detect such events, and researchers then rely on a surrogate marker, a measurement such as blood pressure or the level of cholesterol in the blood that is believed to predict them. The difficulty is that the belief is not always correct, and a drug can improve the marker without helping the patient at all. Blood pressure is the classic example: it is cheap to record and can be measured in everyone who takes part, which is why trials so often depend on it.

Every trial is a sample, and its result is an estimate rather than a certainty. Statistical analysis attaches a range to that estimate, the confidence interval, which shows how much the finding might plausibly vary. A difference can be statistically significant, meaning that it is unlikely to be the product of chance, and still be too small to matter to a patient; conversely, a trial may be too small to detect a benefit that is genuinely there. Before starting, statisticians therefore carry out a power calculation to work out how many participants will be needed.

Because trials expose people to risk, they are closely regulated. Participants must give informed consent, and an independent ethics committee must be satisfied that the possible benefits justify the risks. Research is generally held to be acceptable only when there is genuine uncertainty about which treatment is better, a condition known as equipoise. Trials must also be registered in a public database before they begin, a rule introduced because too many studies with disappointing results were quietly left unpublished. Newer designs, including platform trials that test several treatments at once, are now changing how quickly such evidence can be gathered.

Few scientific practices attract as much public unease as the use of animals in medical research, and few are defended so firmly by the scientists who rely on them. The disagreement is not really about facts. Both sides agree that animals are used in large numbers and that much of the resulting knowledge has been useful. What divides them is a question of ethics, and behind it a question of evidence: whether the benefits are as great as they are claimed to be, and whether they could be obtained in other ways.

Many millions of animals are used in laboratories around the world each year, most of them rodents but also fish, birds, rabbits, dogs, pigs and primates. In most wealthy countries the practice is regulated in detail. Britain's Animals (Scientific Procedures) Act of 1986 requires a separate licence for each project and each person who carries out procedures, and a European directive of 2010 sets similar standards across the European Union. Beneath these rules lies a framework proposed more than sixty years ago by William Russell and Rex Burch, whose three principles — replacement, reduction and refinement — remain the vocabulary of the debate.

The case for the practice rests on a history that is difficult to dismiss. Insulin, which turned a fatal disease into a manageable one, was developed in the early 1920s through experiments on dogs. The polio vaccines of the 1950s were produced using cells cultured from monkey kidneys. Transplant surgery was worked out in animals before it was attempted in people, and the safety of new medicines is still tested in animals before the first human trial. No responsible regulator, defenders argue, would allow a drug to be given to a healthy volunteer without evidence that it is unlikely to cause serious harm.

The counter-argument begins with the differences between species. A mouse and a human being share a great deal of basic biology, but they differ in the fine detail of their immune systems, their metabolism and the way their brains respond to injury. Over the past three decades a long list of drugs that protected the brains of rodents from stroke have failed in patients, and several treatments that cured sepsis in mice have done nothing for people in intensive care. The great majority of candidate medicines that look promising in animal tests are abandoned during human trials.

Then there is the suffering itself. An animal in a laboratory cannot agree to take part, cannot understand why it is in pain and cannot be compensated for what it endures. Some philosophers argue that the inability to consent is beside the point: what gives a being moral standing is its capacity to suffer, and a mammal in a cage has as much of that capacity as a human infant. Others reply that human beings have obligations to their own species that they do not have to others, and that a world without animal research would be a world with more human disease. Both positions are defensible; neither can be proved.

The alternative is not nothing. Cells from human tissue can be grown in dishes, and in recent years researchers have learned to build organoids, tiny clusters of cells that reproduce some of the structure of a kidney or a brain. Mechanical devices lined with living cells, often called organs-on-chips, mimic the way a drug passes through the wall of the gut or the capillaries of the lung. Computer models can predict how a molecule will bind to a target. What none of them yet provides is a whole living system, in which a drug is absorbed, processed by the liver, excreted by the kidneys and observed by an immune system.

Where animal work continues, the question of how it is done matters almost as much as whether it is done. Systematic reviews of published animal studies have repeatedly found that the experiments are poorly designed, too small to give reliable answers and reported in ways that make them hard to compare. Housing animals in more natural conditions, with space to explore and material to chew, is not merely a kindness: stressed animals produce noisier data. Refinement, in other words, improves the science as well as the ethics.

My own view is that neither abolition nor business as usual can be defended. To suppose that a being capable of fear and pain has no claim on us at all is a position I find hard to argue for. But to suppose that a mouse and a child have equal claims, and that a century of medical progress should be set aside, is a position few parents would accept if their own child were ill. What follows is not a compromise for its own sake but a demand for a much heavier burden of proof: an experiment should be permitted only when no adequate alternative exists and when independent reviewers have agreed that the likely benefit is substantial.

That standard is not the one applied today. Animals are still used for purposes that could be served by other means, and the funding devoted to developing those means is small compared with the money spent on research that depends on animals.

The argument, then, is not about whether animal research has done good; it plainly has. It is about how much of it we are entitled to go on doing, now that we know how imperfect the method is and how much more we could learn from other kinds of evidence.