Breast cancer calculus

When Cecilie Parsons got cancer in her right breast in 2001, the removal of the tumour was nothing compared to the chemotherapy that followed.

“I'd feel like crap, really sick,” she recalls. “None of the anti-nausea drugs worked, so I was basically throwing up everything. I lost all my hair by the second week.”

What Parsons endured is all too typical for breast cancer patients today since doctors typically play it safe, recommending chemotherapy after tumours are removed.  But new research from the National Research Council promises to spare many women what she went through by ruling out the need for chemotherapy.  

An algorithm created by the National Research Council could eliminate the need for chemotherapy with many breast cancer patients.

Most of the time, chemotherapy is harmful and unnecessary, experts say. Mark Basik, a breast cancer surgeon at the Jewish General hospital in Montreal, estimates that about 80 per cent of breast cancer patients undergo chemotherapy and only about 40 per cent actually need it.

Doctors often have no way of predicting whether chemo is called for or when it's no more than “liquid Drano,” to use Parsons's term.

The NRC research could help them make an informed choice based on the DNA of the cancerous cells.

Watching genes

Cancer cells are similar to normal cells but their DNA is mutated. Incorrect genes in the DNA make cancer cells multiply too fast or misbehave.

In Parsons's case, the cancer was related to a mutation in a single gene, called BRCA1.

But in general it's difficult to decide how to treat a breast cancer patient because her DNA contains tens of thousands of genes. There usually isn't a simple link between a single gene and a type of cancer.

Not all women need to get chemotherapy.  This could be a guide for those who don't really need the full treatment. —Maureen O'Connor

Maureen O'Connor of the National Research Council co-authored a recent study on predicting cancer recurrence. Her team examined public data on more than 1,000 patients' breast cancer histories. They knew which genes were being expressed abnormally and which patients got cancer again years later.

“The problem up until now is a statistical problem,” says O'Connor. “It becomes a statistical nightmare to go and find the most important genes in all of that data.”

O'Connor and her team created a computer program to filter the data and find the most telling genes. They made an algorithm — a set of computations — to predict for any new patient whether her cancer is likely to recur after surgery by looking at fewer than 50 relevant genes.

“Not all women need to get chemotherapy,” says O'Connor. “This could be a guide for those who don't really need the full treatment.”

O'Connor's team tested the algorithm with hundreds of patient histories and found that when the algorithm recommends against treatment, it's correct between 87 and 100 per cent of the time, depending on the sub-type of cancer. The Breast International Group, a non-profit research consortium, suggests anything above 88 per cent is an accuracy level doctors can trust.

Algorithm for success

For now, the algorithm lies in a lab and a research paper. To get doctors using it, someone needs to develop a “kit,” a set of materials and steps any doctor can use.

“We're talking to companies right now who maybe would look into doing that,” says O'Connor.

O'Connor sees two ways the algorithm could work for clinics. Either doctors could analyze their patients' cancer genes for themselves, or they could mail their patients' samples to a laboratory.

Doctors can send tumour samples to a U.S. company that uses its own algorithm to predict whether one sub-type of breast cancer will recur.

Something similar to the latter option already exists. Mark Basik of the Jewish General hospital says Oncotype DX is a $3,800 USD test developed earlier this decade that helps doctors and patients decide whether chemotherapy is needed after breast cancer surgery. Doctors send sample tissue to Genomic Health, a Californian company. The company runs its own algorithm, which focuses on 21 genes, and sends the doctors a number between 0 and 100 representing how likely cancer is to recur.

Even at $3,800, it's worth it: a 2005 study published in the American Journal of Managed Care suggests the cost of treating a breast cancer patient is $2,000 lower on average when Oncotype DX is considered, because chemotherapy and further testing are sometimes reduced.

Oncotype DX isn't standard in Canada yet. Doctors are still waiting for the results of a clinical trial — titled TAILORx — to confirm its accuracy rate of greater than 90 per cent.

The TAILORx trial launched in 2006 and finally reached its 10,000-patient quota in October, mostly from North America. Only 10 years from now will researchers know how accurate Oncotype DX's 10-year patient forecasts are.

In the waiting room

A clinical trial is the hurdle facing O'Connor's team, too, once its kit gets developed, according to Basik.

“It might take 10 years,” he says.

Then there's competition.

“Oncotype DX, being the first, basically had to prove its usefulness against nothing,” says Basik. “Anything that comes afterwards in a sense will have to prove its usefulness against Oncotype DX.”

Chemo is hard, and if you don't have to go through it, you shouldn't. —Cecilie Parsons

The Canadian algorithm has a competitive advantage: it's 87 to 100 per cent accurate for nine sub-types of breast cancer, while Oncotype DX is over 90 per cent accurate for only one sub-type, representing about half of all breast cancer cases. O'Connor and her team have recently applied their algorithm to prostate cancer, too, with which they hope to reach 99 per cent accuracy.

If the kit develops smoothly and a clinical trial succeeds, the algorithm should be ready to enter Canada's cancer clinics.

Parsons, who is supporting friends through lymphoma and colon cancer, hopes to see the algorithm help future breast cancer patients avoid painful treatment.

“Chemo is hard, and if you don't have to go through it, you shouldn't,” she says. “The more information the patient has at the beginning, the better.”