We should change how we evaluate cancer drugs

Two in five Canadians will at some time be diagnosed with cancer and about one in four of us will die from the disease. Canadians need better access to effective new cancer drugs.

No one wants government drug plans to pay for expensive medicines if they have little clinical value, but Canadians with cancer want the opportunity to access new innovative drugs that may help them.

In this country, the Canadian Agency for Drugs and Technologies in Health (CADTH) evaluates the cost-effectiveness of new medicines, including cancer drugs, for all publicly funded drug plans outside Quebec. A recent examination two of us were involved in, 206 of CADTH’S evaluations of 95 cancer medicines found that it typically estimated higher costs per health gain achieved than the drugs’ manufacturers did.

That’s not so surprising. Manufacturers want to make a case for their medicines and naturally present bestcase expectations. CADTH’S assessments use less favourable assumptions. But the differences are frequently substantial, which raises questions as to why. It also raises concerns for cancer patients: medicines with unfavourable cost-effectiveness estimates may not be recommended for reimbursement by government drug plans. Or they may only be reimbursed if the company deeply discounts the price, which naturally makes companies reluctant to launch new drugs in Canada, which in turn results in delays in access or even no access at all.

A group of health professionals in oncology (including one of us) recently recommended a new approach to getting innovative cancer drugs to Canadians who need them. If a drug were made available under a special access program as soon as Health Canada approved its safety and effectiveness, patients for whom it might be appropriate could use it while data on the drug’s real-world benefit, safety and cost were collected to determine its practical cost-effectiveness in daily medical practice. Programs like this, which are subsidized by manufacturers, are common in several countries but not in Canada.

Canadian evaluations of cost-effectiveness currently use predictive data modelling to try to assess whether cancer drugs are likely to be cost-effective in normal medical practice. Predictive modelling necessarily relies on assumptions rather than widespread actual experience. Health benefits are measured through premarketing clinical trials in which, under strictly controlled monitoring, patients with a tightly defined and verified diagnosis are randomly selected to take either a new medicine or an existing one. Any benefit shown in the trial is assumed to carry over to the drug’s use in regular medical practice. The measure usually used to assess benefit in cost-effectiveness estimates is a “quality-adjusted life year” (QALY). It tries to combine both quality and quantity of life into a single assessment of a patient’s health. But “health” is a complex, multi-faceted physical, psychological and social state that is impossible to adequately quantify in a single value.

QALYS also don’t account for the severity of a disease, so can’t assess whether sicker individuals place more value on gains in health than less sick people do, which is frequently the case in cancer patients. In everyday practice, some patients treated with a new medication may be much sicker than those who were permitted in the clinical trial.

QALYS also fail to capture a medicine’s social benefits, such as reduced caregiving needs or increased economic productivity from less absenteeism from work.

Real-world data are also far from perfect, of course. Some real-world patients may not appear to do as well on a new medication as clinical trial patients did, yet the therapy may still be of benefit for many of them. The key is appropriately assessing the reliability of the data, the differences between real-world and clinical trial patients and the validity of the analysis.

In our view, getting new drugs into clinical practice more quickly would both benefit patients and allow for more comprehensive evaluations of whether these benefits are cost-effective and deserving of reimbursement.

Nigel Rawson is a senior fellow with the Macdonald-laurier Institute. Chak Balijepalli is founder and managing partner of Pharmalytics Group. David Stewart is a professor of medicine at the University of Ottawa, a medical oncologist at the Ottawa Hospital and author of “A Short Primer on Why Cancer Still Sucks.”

CANADIANS WITH CANCER WANT THE OPPORTUNITY TO ACCESS NEW INNOVATIVE DRUGS THAT MAY HELP THEM.