'Buying a Ford but Paying for a Ferrari': Value in Cancer Care

Liam Davenport

October 12, 2017

MADRID, Spain — The relationship between oncologists and pharmaceutical companies, as well as the way in which anticancer drugs are assessed by approval agencies, will have to change fundamentally if true value in drug pricing is to be achieved, says an expert.

Speaking during a Special Session at the European Society for Medical Oncology (ESMO) 2017 Congress, Ian F. Tannock, MD, PhD, emeritus professor of medical oncology and medical biophysics, Princess Margaret Cancer Centre and the University of Toronto, Ontario, Canada, said that the "incredible" rise in cancer costs is both unrelated to their clinical performance and unsustainable.

Consequently, more emphasis should be placed on assessing the value of a drug, as determined by the ESMO Magnitude of Clinical Benefit Scale (MCBS) or the updated American Society of Clinical Oncology (ASCO) Value Framework, rather than approving a drug if it simply meets a prespecified outcome, he said.

The value of a drug should also be used to set maximum prices, Dr Tannock believes. All of this will necessitate changes within approval agencies to shift the balance away from overreliance on P values and a renegotiation of the relationship between oncologists and pharmaceutical companies to allow more open criticism.

Dr Tannock began his presentation by noting that the cost of cancer treatments is rising extremely rapidly and that this "clearly brings into question" the notion of ongoing sustainability.

He said, "If this goes on, we're going to make choices between treating patients with cancer and sending our children to school and university."

He pointed out that even before the introduction of the ESMO and ASCO value scales, he and his colleagues had looked at the cost of anticancer agents and their impact on median overall and progression-free survival, finding no relationship between the two.

Dr Tannock and colleagues then conducted a retrospective analysis of 109 randomized controlled trials (RCTs) of systemic therapies in non-small cell lung, breast, colorectal, and pancreatic cancer.

All the trials were positive, in that they demonstrated significant differences on prespecified primary or secondary outcomes in favor the experimental group.

However, analysis revealed that only 38% of the RCTs met the ESMO criteria for clinical benefit on the MCBS. "This shows the stark difference between a P value for the primary outcome and whether or not there is any benefit, as judged by the scale," Dr Tannock said.

Next, the researchers focused on 100 of the RCTs that had cost data, finding a negative correlation between the cost of a treatment and its benefit score on the ASCO value scale. Moreover, treatments that met the ESMO threshold for benefit had a lower median incremental monthly cost than those that did not, at $2980 vs $8620.

What we're doing as oncologists is we're buying a Ford but we're paying for a Ferrari. Dr Ian Tannock

 "Clearly, when we are in this cost and benefit area, there is absolutely no relationship between cost and benefit," Dr Tannock observed. "What we're doing as oncologists is we're buying a Ford but we're paying for a Ferrari."

Importance of Looking at Value

Dr Tannock described the efforts by ESMO and ASCO, among others, to quantify the value of novel anticancer treatments as "laudatory." However, he said he personally finds ESMO's MCBS to be "the better scale, and that's because it sets the threshold for value, which makes it very easy to use."

He continued: "In my view, value should become the criterion for inclusion of treatment in clinical guidelines."

He also believes that it should be used by the US Food and Drug Administration (FDA) and the European Medicines Agency (EMA) in deciding whether or not to approve new drugs, "but obviously that will need a change in the statutes of those organizations, and it should ultimately set the price of new treatment."

Dr Tannock then pointed to research that he and his colleagues had conducted to determine whether contemporary RCTs meet the ESMO threshold for clinically meaningful benefit.

They looked at 277 trials of patients with breast, non-small cell lung, colorectal, and pancreatic cancer, of which 142 (51%) had reported positive results. Of those, just 43 (30%) met the ESMO threshold for demonstrating a clinical benefit.

Moreover, only 31% of the trials had been designed to detect or exclude differences in outcome that would meet the ESMO threshold, ranging from 65% of trials of treatments with curative intent and 18% of those looking at therapies with palliative intent.

"I think that needs to be taken into account when one assesses the real value of those treatments, even though they may have been approved by the EMA," Dr Tannock said.

Asking why there is such a disparity between positive results on an RCT and a treatment's inherent value, he went on to highlight that both the FDA and EMA "register drugs on the basis of a P value of less than 0.05."

Dr Tannock added that, "because they are in business to make a profit," pharmaceutical companies will consequently do large trials to detect small differences and meet that P value–based criterion.

He cited the case of a trial comparing denosumab with zoledronic acid for the treatment of bone metastases in men with castration-resistant prostate cancer, noting that its large size of more than 1900 patients meant that even small differences in clinical outcomes would become significant.

He said that it was "not even a difference that registers highly on the value scale, and yet this type of trial, at the moment, can make huge profits for the pharmaceutical industry."

Dr Tannock noted that the American Statistical Association recently published a statement on the uses and limitations of P values, highlighting that, at base, they merely state the compatibility of a result with a statistical model.

Among other things, the statement emphasizes  the following:

•  P values do not measure the probability that the studied hypothesis is true or the probability that the data were produced by random chance alone.

•  Scientific conclusions and business or policy decisions should not be based only on whether a P value passes a specific threshold.

•  A P value, or statistical significance, does not measure the size of an effect or the importance of a result.

•  By itself, a P value does not provide a good measure of evidence regarding a model or hypothesis.

Dr Tannock said that despite this statement, the decisions made by major journals and the registration agencies "are pretty much entirely based on P values."

He underlined that repetition of results in further studies and determining the magnitude of effect "are much more important than P values," noting that the estimate of clinical value derived from RCTs involving highly selected patients "may not apply when the same treatment is used in daily practice."

He then quoted Daniel D. Von Hoff, who observed that "to get into many of current trials you need to be an Olympic athlete with cancer."

Dr Tannock said, "We exclude many patients who have other comorbidities, and yet when we take those trials into the clinic, we tend to apply those results, regardless of the comorbidity that's there."

He also contended that RCTs overestimate the clinical benefit and underestimate the toxicity of treatment, saying that approximately 60% of potentially fatal adverse events that occur during phase 3 clinical trials are not included in the initial FDA drug label, and that around 40% are not reported in any published RCT.

Referring to his initial observation over the "incredible" increase in the cost of anticancer drugs, Dr Tannock said that "a system where the inflation of drug costs is far exceeding all other commodities is simply not sustainable."

"That drug development system is broken and has eventually got to change."

He believes that could happen "in several ways," including governments' recognizing that a lot of the funding for the research that leads to drug discoveries comes from the public purse and from charities.

For example, Dr Tannock pointed out that "a lot of the science" underlying the development of programme cell death (PD)-1 and PD ligand-1 inhibitors was supported by funding from the National Institutes of Health or charities, "and yet when these agents come on the market as a benefit of that research, governments are not influencing that pricing."

He said that, ultimately, the FDA and the EMA "should have their statutes changed so that, instead of approving new agents on the basis of a P value in a very large trial, that they base that on an estimate of value."

Furthermore, there needs to be a move to value-based pricing. "Let's face it, everything else we buy is pretty much based on value. If you want a nicer refrigerator, you pay more for it. That isn't the case of anticancer drugs."

Dr Tannock pointed to a paper that he and his colleagues published in 2016 on value-based pricing, saying that drug registration should be based on the demonstration of not just efficacy but of value in appropriate clinical trials, alongside agreement over the maximum price that could be charged per life-year gained.

"So if you went that route, you could allow pharmaceutical companies to try and register new drugs with small effects…but they'd have to market that drug at a very low price to fit within a limit on a cost per life-year gained," he said.

"I think that would then discourage mega-trials using borderline drugs to detect small differences in outcome."

Dr Tannock said that oncologists and societies should support publicizing the cost of drugs, alongside the lack of relationship with their effectiveness and the cost of producing them.

He added that the relationship between oncologists and pharmaceutical companies should be "critically reviewed."

"Almost every presentation here has a list of conflicts of interest. Well, I'm not suggesting that people aren't presenting their data properly, but, you know, the companies are buying our silence," he said.

"Those honoraria that we receive, that silver that we take, makes it harder for us as physicians to criticize the companies and their prices."

Dr Tannock continued: "Clearly we have to avoid using expensive drugs that are marginally effective, and I think we need to advocate for value-based approval and/or pricing."

"That will involve us becoming somewhat adversarial initially with companies, but companies will eventually see that there are other ways to retain a reasonable, but not excessive, profit."

No funding was declared. Dr Tannock has advised multiple companies about design of clinical trials, for which he has received contributions to his research fund and has chaired international company-sponsored trials for hormone-refractory prostate cancer (TAX-327, VENICE).

European Society for Medical Oncology (ESMO) 2017 Congress. Presented September 10, 2017.

For more from Medscape Oncology, follow us on Twitter:  @MedscapeOnc


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: