"There is always a better way": Managing uncertainty in decision making about new cancer drugs in Canada

Determinants of the Cancer Drug Funding Process in Canada

Background: Canada has a publicly funded healthcare system with a complex drug funding process. After Health Canada approval to market a drug, the pan-Canadian Oncology Drug Review (pCODR) (now renamed the CADTH reimbursement review) makes a non-binding funding recommendation to the Canadian provinces (except Quebec), which each then decide whether the drug will be publicly funded. We identified the determinants of funding in this process. Methods: We analyzed drugs for advanced lung (n = 15), breast (n = 8), colorectal (CRC) (n = 7), melanoma (n = 10), and neuroendocrine (NET) (n = 3) cancers undergoing the funding decision process from 2011 to 2019. Determinants of funding assessed in the model included list price, cancer type, drug class, and pCODR recommendation. The primary outcome was the correlation between list price and time to funding (TTF: Health Canada approval to first provincial funding). Secondary outcomes included an exploratory analysis of predictors of drug funding. Results: We analyzed 43 drugs: targeted agents 72%, immunotherapy 20%, chemotherapy 7%. A total of 72% were funded in at least one province. Median TTF was 379 days (IQR 203−601). Median list price (28-day course) was CAD 8213 (IQR CAD 5391−9445). Higher list price was not correlated with TTF (correlation coefficient −0.20, p = 0.28). There was no association between list price and pCODR recommendation or the decision to fund in at least one province. A positive pCODR recommendation correlated with the provinces’ funding decisions (p < 0.001), where 89% of drugs with a positive recommendation were funded and 100% of drugs with a negative recommendation were not funded. Tumor type was predictive of TTF (p < 0.001): CRC drugs were the slowest at a median of 2541 days (IQR 702−4379), and NETs were the quickest at a median of 0 days (IQR 0−502). Cancer type predicted decision to fund in at least one province (p = 0.005), with funding for 100% of NET drugs at the high end and 29% of CRC drugs at the low end. Drug class was predictive of TTF (p = 0.01): 465 days (IQR 245−702) for targeted agents, 443 days (IQR 298−587) for chemotherapy, and 339 days (IQR 164−446) for immunotherapy. Conclusions: Determinants of drug funding included cancer type, drug class, and pCODR recommendation but not list price. Factors other than cost were more heavily weighted in the funding decisions of cancer drugs in Canadian provinces.

Purchasing medicines for the public sector: Evaluation of the performance of centralised procurement in Portugal

OBJECTIVE

The study aimed to evaluate centralised procurement of medicines (CPM) in Portugal.

METHODS

Data were collected through different methods, including a review of the literature and (procurement) documents and an analysis of selected bids. Thirty-seven face-to-face interviews with representatives of public authorities, users (hospitals and regional health administrations), patient associations and pharmaceutical industry were held in Portugal in Q1/2020.

RESULTS

CPM has contributed to improved transparency in processes and governance, to increased equity in access to medicines across the country and to lower workload for some users. The findings of the impact on medicine prices and availability are mixed. The benefits of CPM are undermined by some gaps: Lengthy, bureaucratic processes have resulted in delayed availability of medicines at the beginning of a year and in coping strategies of hospitals such as parallel individual procurements. The list of active ingredients under CPM has not been updated since 2016. The procurement agency does not routinely perform market consultations. Key performance indicators for CPM are lacking.

CONCLUSIONS

Portuguese policy-makers are urged to develop an updated procurement strategy to provide guidance and clarity on the objectives of CPM, the role of the procurement agency and further authorities and key performance indicators.

Describing Sources of Uncertainty in Cancer Drug Formulary Priority Setting across Canada

Over the years, there have been significant advances in oncology. However, the rate that therapeutics come to market has increased, while the strength of evidence has decreased. Currently, there is limited understanding about how these uncertainties are managed in provincial funding decisions for cancer therapeutics. We conducted qualitative interviews with six senior officials from four different Canadian provinces (British Columbia, Alberta, Quebec, and Ontario) and a document review of the uncertainties found in submissions to the pan-Canadian Oncology Drug Review (pCODR). Participants reported considerable uncertainty related to a lack of solid clinical evidence (early-phase clinical trials: generalizability, immature data, and the use of unvalidated surrogate outcomes). Proposed strategies to deal with the uncertainty included risk-sharing agreements, collection of real-world evidence (RWE), and ongoing collaboration between federal groups and provinces. The document review added to the reported uncertainties by classifying them into five main categories: trial validity, population, comparators, outcomes, and intervention. This study highlights how decision makers must deal with significant amounts of uncertainty in funding decisions for cancer drugs, most of which stems from methodological limitations in clinical trials. There is a critical need for transparent priority-setting processes and mechanisms to reevaluate drugs to ensure benefit given the high level of uncertainty of novel therapeutics.