He'd undergone surgery the following day. However, relations with his neurologist soon became strained due to his dogged insistence on researching his own treatments.
His doctor wanted him to stick to the standard regimen of surgery, radiotherapy and chemotherapy. But the rebellious Harvard alumnus insisted on adding to this a veritable cocktail of drugs - including acne, blood pressure and insomnia pills. All were cheap and had little or no toxicity, and for all of them Williams had gathered some credible evidence from scientific trials that they might reduce his tumour, boost his immune system and make chemotherapy more effective. But none had been approved in the United States for use in the management of brain tumours, so his specialist dismissed them.
"He said I would drive myself crazy researching all these things and that I might hurt myself," recalls Prof Williams, whose story is told in a new film, Surviving Terminal Cancer, released online today.
"I almost laughed. Hurt myself? I had the most aggressive kind of brain tumour. I was expected to die in a year. What did I have to lose?"
Even today the average life expectancy for glioblastoma multiforme patients is just 15 months, with survival rates highest among young people. Fewer than 10 per cent of people aged 50 and above survive for five years. So it is against all odds that Prof Williams has just celebrated his 70th birthday and 20 years of clean MRI scans.
"I'd been told that my chemotherapy wouldn't get rid of the tumour completely or indefinitely, so I focused on finding agents that might make chemo work better for me," he says. Sure enough, after his fourth round of chemotherapy in 1996, Prof Williams's tumour had vanished. It has never returned and thousands of people, including oncologists, have sought his advice since on beating a cancer known in medical circles as "the terminator".
In mainstream oncology Prof Williams is considered a freak case and his strategy of fighting cancer by "using every potentially efficacious agent I could lay my hands on" attracts suspicion. Yet a growing number of specialists and researchers say there is evidence that some of the common pills taken daily by millions for other ailments, could be "repurposed" to help fight cancer.
"We just need to look in our medicine cabinets," says Pan Pantziarka, scientist and UK spokesman for the Anticancer Fund, a Belgian non-profit organisation run by researchers and doctors.
"There is strong evidence that some of the medicines we use every day have anti-cancer properties."
At present, the pharmaceutical industry is using advances in our understanding of genetics to create "magic bullets", a new generation of ever smarter, ever more targeted therapies. These act like snipers, interfering with specific cell proteins or signalling pathways that have a role in cancer. Major successes include the chronic myeloid leukaemia drug imatinib (Glivec) which blocks a protein that makes cancer cells grow and divide.
But, according to Professor Angus Dalgleish, the Foundation Chair of Oncology at St George's Hospital, University of London, many targeted therapies are eventually doomed to failure.
"Cancer will do everything it can to survive and avoid being hit," he said.
"It's like a traveller who wants to cross London on the Tube. Yes, you could block him by taking out a major station like Oxford Circus, but he'll just switch to a different line. It's the same with cancer. After getting hammered by one agent, the tumour quickly reinvents itself through evolution. I always delay giving these targeted therapies as long as possible because I know they're not going to be working a few months down the line."
It costs more than US$1 billion to bring a new cancer drug to market and takes more than a decade. As drug companies face an increasingly uphill battle to invent new chemical entities that can be patented, they pass on the cost of a 90 per cent-plus failure rate, expensive trials and marketing to the NHS, insurance companies, healthcare providers and patients.
The good news is that early-stage laboratory experiments and clinical studies, as well as large-scale epidemiological research point to the potential cancer-fighting properties of dozens of medicines that millions of people take safely every day for other ailments. Aspirin is the most high-profile example. Research funded in part by Cancer Research UK shows that it can significantly cut the risk of bowel, throat and stomach cancer if taken daily by people aged 50-65 (although the CRC warns that aspirin can have side effects and should not be taken regularly without medical advice).
The Repurposing Drugs in Oncology (ReDo) Project, an international collaboration between the Anticancer Fund and US-based non-profit organisation Global Cures, has identified 70 potential agents where there is evidence of cancer-fighting properties. These include the diabetes tablet metformin, cholesterol-lowering statins, the antacid cimetidine, the de-worming tablet mebendazole, the anti-fungal itraconazole, and all the drugs Prof Williams took as he battled his brain tumour.
"Most modern cancer drugs are known as targeted therapies because they are aimed at very specific targets inside cancer cells," says Dr Pantziarka. "But these older medicines are known as 'dirty drugs' because they have multiple targets, interfering with more than one protein or signalling pathway at a time. Used in combination, they could be very effective.
"If these medicines were coming out today, some would be blockbuster cancer drugs. But most are no longer covered by patents so the pharmaceutical industry has no financial incentive to investigate them."
Even though these medicines are not officially labelled as cancer drugs, doctors are legally entitled to prescribe them "off-label" if there are solid grounds to believe they will be beneficial. Indeed, Lord Saatchi's Medical Innovation Bill aims to give oncologists (and other specialists) more confidence to be experimental in treating terminally ill patients for whom all existing therapies have failed and who are prepared to take risks. But in practice, most oncologists are unwilling to prescribe drugs that have not passed final-stage (Phase III) clinical trials for a particular type of cancer, and even more wary of combinations that may interfere with conventional treatment or have unforeseen side effects.
Prof Dalgleish is one of the few UK doctors willing to think differently.
"Say we have a patient who is fit and healthy in many ways but is going to be dead within months. If that patient asks me, 'Is there anything else I can do?' I will say, 'Yes. The data suggests you could consider metformin which appears to selectively reduce glucose uptake by tumour cells as opposed to normal cells. I suggest aspirin to tackle your inflammation and let's correct your vitamin D levels to boost your immune system.'
"I call it creative compassion because it's not in the rule book and it's what I would want for myself if I were in the same position. I wouldn't want to just be told to go and see the palliative care person."
Prof Justin Stebbing, of Imperial College London, the oncologist who treated actress Lynda Bellingham before her death from bowel cancer last year, agrees with the approach, though he tends not to prescribe off-label drugs himself.
"That's not because I'm ethically opposed to the idea, but it's not something I do every day so I'm not knowledgeable about dosages," he said. "However, I don't have a problem if patients get the drugs elsewhere and am open to referring them to trials that test these alternative approaches.
"As a profession, we can be cruel to cancer patients, giving them treatments that are horribly toxic with minimal benefits. I find it very frustrating when my colleagues are dismissive of patients who want to try other things that are non-toxic and may extend their lives."
Professors Dalgleish and Stebbing are the co-authors of a study into "cocktail cancer therapy" taking place at the Care Oncology Clinic in London. Over the next five years, the private clinic run by biotech firm SEEK is aiming to treat more than 10,000 cancer patients with a combination of four 'dirty drugs' - statins, metformin, the de-worming drug mebendazole and doxycycline, a common antibiotic.
Gregory Stoloff, founder of SEEK, says it's a non-profit trial funded by patients themselves, who pay for the initial consultation, then every three months to cover the cost of the drugs, consultations and the trial.
"Oncologists refer patients to us who have run out of treatment options and we put all of them on the treatment right away," he says. Because nobody is given a placebo, this is not a controlled clinical trial. But SEEK hopes that treating thousands of people will create enough data to enable an effective comparison with cancer survival rates in the rest of the population. "Controlled clinical trials are all very well but when you have people who are expected to die within months they want treatment now. And these are very safe medicines that people have been taking for decades."
Although Phase III clinical trials remain the gold standard for science, they don't necessarily serve cancer patients well, says US campaigner Dominic Hill, in whose film Prof Williams appears, along with oncologists, researchers and other cancer survivors.
"Today's life-saving treatment for HIV [called antiretrovirals] is given to patients despite it never having passed a randomised Phase III clinical trial," Mr Hill, whose brother-in-law Andreas died of a brain tumour in 2010, points out.
"How many cancer patients must die before the regulators recognise that this approach needs to be adapted to oncology?"
Dominic Hill's film Surviving Terminal Cancer is free to watch on survivingterminalcancer.com
Cancer and 'dirty drugs'
• Metformin
Several studies suggest that tumours grow more slowly in cancer patients who are on this anti-diabetic drug. Early-stage clinical trials are investigating its potential to prevent various cancers including prostate, breast, colorectal and endometrial.
• Statins
Pre-clinical studies suggest these cholesterol-lowering drugs may prevent various cancers and stop them spreading. One recent meta-analysis associated a daily statin with a significant risk reduction of liver cancer.
• Mebendazole
There is evidence this drug - usually prescribed to treat parasitical worm infections - may inhibit cancer cell growth and secondary tumours, though no clinical trials have been completed.
• Cimetidine
This over-the-counter antacid has direct anti-proliferative effects on cancer cells, inhibits cell adhesion, reduces tumour angiogenesis (growth of blood vessels essential to a tumour) and also boosts anti-cancer immunity.
• Itraconazole
The common anti-fungal treatment is also thought to be anti-angiogenic and has shown promise as an agent for prostate cancer, non-small cell lung cancer and basal cell carcinoma.
• Isotretinoin
This acne drug, marketed as Accutane, is occasionally used to treat certain skin cancers and nerve cancers, as well as to prevent the recurrence of some brain tumours; some studies suggest it is ineffective.
