The great aim of modern cancer treatment is not necessarily to cure patients of their tumours but to allow people to die with their cancers and not because of them.

It is often said that this is already the case for prostate cancer, as about 70 per cent of cases respond well to treatment with chemical castration to block the male hormones on which these tumours thrive. It is the remaining 30 per cent, however, that has until now appeared a stubborn block to the cancer becoming a chronic condition that does not kill.

As prostate cancer is not the only type of cancer driven by sex hormones, the drug abiraterone could potentially also be applied to other forms of the disease.

Scientists from Cancer Research UK are soon to begin a trial that will evaluate abiraterone for the subset of breast cancers that are affected by male hormones, and it is also a candidate for some ovarian and endometrial cancers.

Even some tumours of nonreproductive tissues, such as the lung and bowel, are influenced by sex hormones, and laboratory work has suggested that these might respond to hormone-blocking drugs as well. In each case, however, abiraterone or similar therapies will be likely to work best in a subgroup of patients. This is because cancer is a disease of the genes, and to classify tumour types by the tissues in which they originate is too broad a system to allow doctors to select the most appropriate therapy.

It is well understood that there are at least five types of breast cancer, each driven by different patterns of genetic mutation, and these do not respond to the same drugs. Herceptin, for example, is effective only in the one in four patients whose tumours have a mutation in the Her2 receptor.

The same is true for other cancers, including prostate. The Institute of Cancer Research team has already identified one mutation, a rearrangement of a gene called ERG, whose presence predicts a good response to abiraterone, and others can be expected to be identified in the future.

As more of these genetic drivers become more fully understood, patients will start to be prescribed drugs not according to where cancers occur in the body, but according to the DNA profiles of their tumours.

Some drugs, indeed, may prove useful against a range of cancers that oncologists would not previously have thought might be linked, but which nevertheless are driven by the same mutated genes.

A gene called BRAF, for example, is often defective in malignant melanoma, the deadliest form of skin cancer, caused by exposure to ultraviolet light. It is also common, however, in colon cancer - in a part of the body that rarely sees sunlight. A BRAF antagonist might work well against both, in spite of their manifest differences.

This new era of bespoke cancer therapy, however, poses a significant challenge: cost. If certain drugs are designed to be appropriate for only a minority of patients, they are likely to become very expensive, as the overall market will be smaller than it is for one-size-fits-all blockbuster drugs.

Herceptin, for example, costs £20,000 per patient per year, and Avastin, for colorectal cancers with a particular genetic profile, has been rejected by the National Institute for Health and Clinical Excellence (NICE) as too expensive to offer value for money.

Those who pay for the NHS through their taxes have a right to medical care, but no right to demand particular types of care. Any system of healthcare has to draw the line somewhere, and it is the job of NICE to provide the evidence on which to base rationing. In recent years, dozens of cancer treatments - many of them offering groundbreaking clinical benefits - that have been licensed for use in Britain have been denied to NHS patients because of bureaucratic delays. Time and again, these have come down to the issue of cost.

If science really can deliver on the promise of turning cancer into a chronic illness, some of these economic issues are likely to be resolved. If drug companies are making products that people take for years, even decades, their costs can be expected to come down.

Even so, for the benefits of bespoke cancer therapy to be realised, creative thinking cannot be restricted to geneticists and oncologists. Health economists are also going to have to consider how the industry can build a market for its new products, and how health services will pay for them.

Times Recommends

• TV bidding war starts over Olympics
• Sofas were larger than life in TV adverts
• Driver pays fine with wheelbarrow of pennies