It is a day Dorothy Griffiths will never forget. Told that the spread of her breast cancer meant she had only six months to live, she tried to persuade her local director of public health that he should provide her with a cancer drug that might extend the pitifully short lifespan predicted for her.

'You do realise this would use up taxpayers' money,' he informed her. 'Anyway, how do you know it would prolong your life?'

But Griffiths is a determined sort of woman. A former NHS manager herself, she refused to give up. She finally won over the chief executive of her local hospital in Stoke-on-Trent, who agreed to buy her Herceptin, the drug she needed, on the NHS. The two others who were fighting with her for the therapy died before they could receive their first injection.

But two years on, 56-year-old Griffiths is still alive and making the most of every day. 'I am living with cancer, not dying of it,' she said ebulliently. 'I'm not really supposed to be here at all, am I? But this new treatment is worth millions to me. I've seen my grandson born, and that little boy, along with my two step-granddaughters, keeps me going.'

About 5,000 women a year are diagnosed with advanced breast cancer, but not all of them will be offered the same chance of prolonging their lives.

There still remains a lottery over cancer drugs in Britain, but not for the traditional reason of NHS underfunding. Instead, the biggest problem now facing the Government is the lack of facilities and staff to carry out the testing that it is needed in order to determine which patients need which drugs. As an increasing number of therapies, tailored to particular genetic peculiarities, come into being, the pressure will grow on the hospital backroom staff - the pathologists and lab technicians who help diagnose hundreds of different diseases.

For every woman facing the prospect of her cancer having spread, there are very difficult choices to be made, and the hardest choice of all is to decide how far to go with hospital treatment and when to stop. A growing number of women seem to eschew traditional chemotherapy altogether, in favour of a more holistic approach. One of those was Paul Merton's wife, Sarah Parkinson, who died last week after a long battle against the disease.

She had written about how she rejected traditional treatment in favour of better nutrition, yoga and positive thinking in an attempt to boost her immune system. The former TV producer felt that the fertility treatment she had received had contributed to the cancer, and that it was time to get her body back in balance.

Parkinson described earlier this year how she turned down traditional chemotherapy after a bruising encounter with an oncologist whom she described as 'appalling and callous'. It is not known whether she had been offered a test for Herceptin, but it seems unlikely she would have accepted it, believing it might do her more harm than good.

However, a new generation of drugs is coming on line that offers the prospect of treatment being better tailored towards the individual, and with fewer toxic side-effects. As our genetic knowledge grows, scientists can tease out the various strands of biochemistry that matter in cancer. Laboratories can think about targeting particular molecules, instead of the old approach of blasting a malignancy in all directions.

With the advances of science comes hope. But with every glimmer of hope come the age-old hurdles of too few resources and too little access to drugs that could ease patients' suffering.

Herceptin is unusual. It is the first monoclonal antibody to be licensed for its 'seek-and-destroy' properties against cancer. The Government's own rationing watchdog, the National Institute for Clinical Excellence (Nice), last year recommended that all women with advanced breast cancer who have a particular genetic make-up should receive the drug, which is known to prolong survival for those whose cancer has spread.

But it is now thought that hundreds of seriously ill patients who might be eligible for the drug are not receiving it. Britain has the lowest uptake of the therapy in all Europe.

One fifth of all breast cancer patients carry a specific genetic defect that makes them overexpress a protein, HER2 (human epidermal growth factor receptor 2). This protein influences the rate at which the cancer cells grow and divide, and makes it a more aggressive form of cancer to treat.

The results of the latest trial, published last week, show that Herceptin can double survival time, when it is given alongside another drug, docetaxel. With docetaxel alone, a woman whose cancer has spread beyond the breast has an average length of survival of 13 months.

In fact, Herceptin now seems so successful that it is the subject of a large-scale trial across Britain looking at whether it should be given to all breast cancer patients carrying the faulty gene as an early, first-line therapy, not just to those at an advanced stage of the disease.

But that would mean offering the drug to twice the number who are currently eligible to receive it, and it is hard to see how the NHS could cope.

The country is short of some 300 trained pathologists, and that number is expected to increase to 400 within the next few years. Professor Chris Twelves, of St James' Hospital in Leeds, spelt it out: 'The pathology labs at NHS hospitals are under enormous pressure because of lack of resources and enough staff to cope with a mass of tests for all sorts of diseases.'

In other European countries, women who come into hospital with a diagnosis of breast cancer are routinely given the HER2 test, because it is a useful predictor of how the disease will progress and whether it is likely to spread to other organs.

But some women are waiting weeks, and sometimes months, before the test is carried out, according to Delyth Morgan, chief executive of the Breakthrough Breast Cancer charity. 'We know from our patients that many of them are having the test, but waiting a long time for the results. A significant number are not even getting the results back. Others seem not to be offered the test at all.'

Morgan added: 'Here we have a drug, with very few side-effects, that can give extra precious months of life to thousands of women.

'We see it as a litmus test for potentially similar drugs, but, without the right test, Herceptin can't be prescribed. We have to ask what is the point of developing these drugs if they are never going to reach the people they were developed for?'

Britain has seen enormous advances in the survival rates for breast cancer. More than 43,000 women each year develop the cancer, but for those caught at an early stage, their chances of surviving more than five years now stand at 90 per cent. The overall rate for breast cancer survival has leapt from 55 per cent in the 1970s to 74 per cent in 1993.

The hormone therapy Tamoxifen, which helps to prevent the cancer recurring, has been largely responsible for the falling mortality rate, but so has a nationwide screening programme for women over 45.

But more tests for particular forms of the disease now present the NHS with a new problem. It is not the cost, but the lack of diagnostic facilities that will cause the difficulties over the next decade.

With every new breakthrough comes the question of resources. Doctors announced last Friday that they are developing a test to see which women carry the faulty BRCA1 gene, which seems to predict how fast a tumour will grow.

If the gene is defective, the tumour is more likely to be resistant to some drugs, but less so to others, allowing doctors to tailor the chemotherapy accordingly. It has widespread implications, as one in 850 women in the UK is estimated to inherit a faulty copy of the gene.

Leading oncologist Professor Karol Sikora told The Observer: 'The future of cancer therapy is about personalised molecular therapy. With each of the future drugs will come a diagnostic to ascertain whether patients will benefit, which makes sense clinically as well as economically.

'The measurement of the Herceptin receptor will just be one of many tests needed, and we have to look at ways of meeting the shortfall we are currently facing on the testing front. If we don't, we'll fall further yet further behind the rest of Europe, even if the drugs are actually approved for use here.'

One possibility might be for the Government to start looking at recruiting more private pathology services in order to meet the demand. The use of private companies to fast-track patients who have waited months for surgery has already been a central plank of government policy, and many believe there is now a case for doing the same for diagnostics. A private laboratory was used by the drugs company Roche for its trial of Herceptin, but once it had been given Nice approval, it was switched to the health service.

Gill Alexander is one breast cancer patient who ended up relying on a private laboratory in order to be told her HER2 status. She first had cancer six years ago, but has suffered two relapses. Earlier this year, her oncologist in Norwich suggested she undergo a HER2 test, but she waited for months for the test results to appear. Even Addenbrooke's Hospital in Cambridge, well known for its cancer work, was unable to carry out the specialised test for her.

Frustrated, she ended up being referred as a Bupa patient to a private lab in Nottingham, which analysed her results and discovered that she was HER2-positive.

'I've been on the therapy since August and I feel great,' Alexander said from her home in Norfolk. 'I appreciate I've been really lucky, because it seems to me that this treatment is being kept under wraps. That's wrong; anyone with breast cancer deserves the best possible chance of surviving as long as they can.'