This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Pater, J. L. Articles by Cowan, D. H. Search for Related Content PubMed PubMed Citation Articles by Pater, J. L. Articles by Cowan, D. H.

Funding New Cancer Drugs in Ontario: Closing the Loop in the Practice Guidelines Development Cycle

From Queen’s University, Kingston; McMaster University, Hamilton; Cancer Care Ontario, Toronto; and University of Toronto, Toronto, Ontario, Canada.

Address reprint requests to Joseph L. Pater, MSc, MD, Queen’s University, 82-84 Barrie St, Kingston, Ontario, Canada K7L 3N6; email: jpater{at}ctg.queensu.ca

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
PURPOSE: The previously described practice guidelines development cycle follows an iterative model in which recommendations are reached by a process that incorporates practitioners at all phases. A key feature is the separation of the evidence-based systematic review and the generation of recommendations from policy decisions surrounding implementation. This article describes how this implementation phase has evolved in Ontario and how implementation has affected the guidelines process.

METHODS: The development of the New Drug Funding Program in Ontario and the appointment of a policy advisory committee (PAC) to make funding recommendations were reviewed. The decision-making framework of the PAC is described in this article.

RESULTS: The PAC has had to address a number of issues in making funding recommendations. These issues have included dealing with evidence arising solely from phase II versus phase III trials, using economic information, and involving community representatives in its deliberations. Its activities have had a substantial impact on the practice guidelines initiative.

CONCLUSION: It is possible to integrate an evidence-based, practitioner-driven approach to clinical guideline development with a funding program that takes policy considerations into account. However, even though these two roles are conceptually separate, the needs of the funding program have inevitably had an impact on the guidelines process.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
ONTARIO’S CANCER system, now called Cancer Care Ontario (CCO), established a province-wide practice guidelines initiative (CCOPGI)¹ in 1995. As indicated in the original publication,¹ this initiative was developed from a model, the practice guideline development cycle, which incorporates a formal iterative process of problem definition, evidence review, consensus development, guideline formulation, and eventual adoption and implementation of clinical recommendations. Two critical innovations of the initiative are the establishment of a formal feedback process from community-based and institutionally based practitioners² and the separation of responsibilities for generating clinical recommendations based on evidence from the creation of policies that would take into account additional nonclinical features, such as feasibility and affordability.

Since 1995, the CCOPGI has used the guideline development cycle to generate a number of evidence-based guidelines related to a variety of disease sites (eg, lung cancer³). During the period of initial guideline development, Ontario’s publicly funded cancer system faced increasing difficulty in supporting the steadily escalating costs of anticancer drugs. In response, the New Drug Funding Program was developed with the primary goal of ensuring that cancer patients in Ontario would have access to all systemic agents of proven value for their conditions. From the outset, the committee given responsibility for developing funding policies in this program was mandated to base its decisions on evidence-based recommendations from the CCOPGI. The purpose of this article is to describe how this was accomplished and to highlight some issues that arose in our efforts to develop a systematic approach to building policy based on a foundation of evidence.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Background
The system for funding cancer drugs in Ontario. In Ontario’s cancer system, intravenous (IV) and oral anticancer and supportive care drugs are funded differently. IV drugs, the focus of this article, have traditionally been given in the hospital and are an insured benefit under the Canada Health Act. Thus, public funding is provided for all intravenously administered cancer drugs, even when they are administered in an ambulatory setting. In contrast, oral drugs for outpatients are not included in the overall health insurance plan and are paid for by both public and private mechanisms.

Another important feature of Ontario’s system for funding IV drugs is that Ontario’s Ministry of Health and Long Term Care (MOHLTC) interprets the Canada Health Act to forbid the sale of publicly funded services to ensure that equity of access is preserved by removing the barrier related to "ability to pay." Thus, patients cannot buy IV drugs directly or through insurance; drugs not publicly funded are essentially not available to patients.

Before the development of the program described here, all intravenously administered anticancer and supportive care drugs were paid for out of the global budgets provided to hospitals and cancer centers as part of negotiated annual allocations from the MOHLTC. The decisions to provide these drugs and to pay for them were the responsibility of individual hospitals and the eight regional centers of the cancer system rather than a centralized provincial system.

Creation of the policy advisory committee. In the early 1990s, it became apparent that the original system described above could no longer provide equitable access to cancer agents, particularly new IV drugs such as paclitaxel that were much more expensive than previous treatments. In addition to the increasing expense for these drugs, it was clear that indications for treatment were expanding and that the number of cancer patients was growing as the population aged. Thus the cost of providing treatment would inevitably escalate at a time when governments and, by default, healthcare institutions were trying to contain health care costs. However, these provincial concerns were not a factor in the federal drug approval process that was licensing more and more new anticancer agents.

A systemic therapy task force was therefore struck to recommend a solution to these problems. This task force, which included health professionals (physicians, nurses, and pharmacists), community representatives, and a health economist, met over several months in 1994. Its final report was accepted by the MOHLTC, and CCO was given the responsibility to implement its recommendations. Briefly, the key recommendation that has been accepted and implemented was the establishment of a centralized funding system for newly approved IV drugs (the New Drug Funding Program) so that effective drugs would be equally available to all cancer patients in Ontario. The responsibility for advising CCO regarding which drugs should be funded was given to a policy advisory committee (PAC). The PAC was developed with a composition similar to that of the systemic therapy task force, and because the recommendations of the PAC were to be based on guidelines developed by the CCOPGI, the director of that program was made a member of the PAC.

Therefore, from its inception, the PAC was envisioned as fulfilling a role complementary to the CCOPGI. The relationship between the PAC and CCOPGI was intended to be arms-length to preserve the separation, designed into the practice guideline development cycle,¹ of the responsibilities for reviewing and synthesizing evidence to create clinical recommendations from health policy decision making. The PAC process is, in effect, the final adoption step of the cycle. However, to date the PAC’s role has been applied only to clinical issues related to the funding of new and expensive anticancer drugs. The remainder of this article will describe how the PAC of the New Drug Funding Program has dealt with its mandate and how this new program has affected the evolution of the CCOPGI.

Evolution of the Committee’s Decision Framework
The PAC began meeting in 1997. In their initial deliberations, PAC members identified the multifaceted nature of the evaluation process for determining whether a new agent ought to be approved for public funding from the central funding envelope. Factors to be considered included the quality of the available evidence, the magnitude and relevance of the benefit achieved by the drug, the alternative therapies available, and the acquisition costs of the drug. An evaluation matrix was developed that displayed these different aspects. In addition, both a hierarchy of evidence and a hierarchy of benefit were formulated. Table 1 displays the matrix in its final form and includes examples of drug ratings. The decisions regarding these drugs are described in the next section to provide examples of the committee’s process.

Bisphosphonates in breast cancer. Both oral clodronate and IV pamidronate have been shown in randomized trials to improve symptoms and reduce skeletal complications in women with bony metastases caused by breast cancer.⁶ The PAC considered the following issues: the acquisition costs of the oral clodronate regimen are substantially less than that of the IV pamidronate regimen, the costs of the oral and IV forms are borne by different payers, and, for many patients, oral administration is more convenient. Therefore, the PAC decided to reimburse pamidronate only for patients who had tried and did not tolerate oral clodronate.

Rituximab for follicular lymphoma. The only data available for the PAC’s consideration regarding rituximab came from phase II studies.⁷ However, on the basis of an evidence summary produced by the CCOPGI’s Hematology Disease Site Group, the PAC decided to provide reimbursement for the use of rituximab in patients for whom anthracyclines had failed or who had limited marrow reserve. The basis for this decision was the fact that despite the limitations of the evidence, rituximab seemed to produce frequent and durable responses with limited toxicity.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
Issues Dealt With by the PAC
Evidence limited to phase II studies. A major challenge for the committee has been the evaluation of drugs for which the highest level of evidence comes from noncomparative trials (phase II) in which the only outcome measured is tumor response, an outcome that the committee ranked lowest on its hierarchy of benefits because its clinical relevance is often unknown. In Canada, as in the United States, the federal regulatory agency will approve drugs on the basis of data from phase II studies in certain circumstances, and several such drugs have become available recently. This situation was and remains problematic for the committee for two major reasons.

First, in general the CCOPGI does not attempt to develop firm recommendations in the absence of data from randomized trials. To address this problem, the CCOPGI developed a new document distinct from the practice guideline report called an evidence summary. The evidence summary is a synthesis of the research evidence, regardless of its quality, accompanied by an expert interpretation of the evidence together with an opinion-based statement that falls short of a formal clinical recommendation. Relevant trials are identified and reviewed by the appropriate disease site group, but the other time-consuming steps of the guideline cycle, including practitioner feedback, are omitted.

A second difficulty in using data from phase II studies as a basis for funding recommendations is that there is no obvious way to rank agents whose evaluation is restricted to such studies because the data supporting each agent are relatively weak. Initially an effort was made to devise a formal weighting system based on various aspects of the design and results from the phase II studies. In the end, this effort was abandoned mostly because no rational basis for setting a cutoff for funding decisions could be established.

As indicated in Table 2, two options at opposite extremes were initially considered: to fund all approved drugs, but solely for their approved indications, or to fund only those drugs for which the evidence for efficacy or effectiveness was from well-designed randomized trials. Both options were rejected because they represented an abandonment of the committee’s responsibility to ensure that truly effective agents were equitably available. Members of the PAC found it difficult to achieve consensus on denying agents that seemed to be promising in phase II studies when there were no, or few, available alternatives. In the end, the committee decided to review drugs in this category on a case-by-case basis, taking into account factors such as high complete response rates, prolonged response duration, or lack of toxicity.⁸ In addition, for drugs funded solely on the basis of reported response rates, approval is often conditional on the implementation of a system to monitor actual response rates in the Ontario setting. These data are fed back to practitioners and to the committee for future consideration.

• In general, the committee has used a threshold approach to funding decisions. That is, drugs are funded or not funded on the basis of benefit and quality of evidence before cost is considered. The cost impact of a new drug may, however, affect the specific circumstances in which the drug is funded or how carefully its use is monitored.
  
• Although the program has an annual budget, to date requests for additional funding for approved drugs have been supported by the MOHLTC.
  
• The New Drug Funding Program has responsibility only for one aspect of the cost of chemotherapy (ie, drug acquisition). Therefore, it cannot easily factor into its decisions savings and costs in other areas, for example, nursing and pharmacy.
  

Thus, whether the committee could or would make use of information on cost-effectiveness or cost utility has not yet been tested. Studies of the committee’s approach to decision making suggest⁹ that it is unlikely a single metric, for example, cost per quality-adjusted life years, would be satisfactory to the members.

The role of community members. CCO’s policy is to include community representatives on all committees. These committee members are generally people with a personal experience of cancer either directly or as a family member. They often have achieved stature in their own areas of interest and expertise. The PAC includes several community representatives to ensure that the voice from this group is strong and mutually supportive. For the deliberations of the PAC, this community and patient perspective is valuable in providing insights about the relative weight placed on different criteria and the relevance of various outcomes reported in trials. Two important contributions of the community representatives include insisting on an open process and raising debate about the role of the PAC as either a gatekeeper for new drugs or an advocate for the acquisition of new drugs for patients. The latter issue has had an important impact on how the PAC’s envelope of funding has changed over the course of its experience to date. It is interesting to note also how receptive the community representatives have been to the rigor of the process as they learn more about the scientific approach to cancer treatment.

Impact on the CCOPGI
As indicated in the introduction, the PAC is at arms length from but a natural outgrowth of the CCOPGI. The decision of the MOHLTC to create a New Drug Funding Program was premised on the use of evidence-based guidelines in the decision-making process. However, the existence of the program and its advisory committee has created some challenges for the CCOPGI.

The main problem has been one of setting priorities among all aspects of oncology patient management where guidelines are needed. Although the PAC is focused solely on new IV drugs, the CCOPGI has a responsibility to develop practice guidelines across all aspects of cancer care and all modalities of treatment. Because of the rapid emergence of many new drugs over the past several years and the need for a timely response for funding purposes, the CCOPGI has experienced strong pressures from the PAC that have detracted from guideline development obligations in other areas. This has caused frustration for some disease site groups. Further, adding the evidence summary reports to the portfolio of products developed by the CCOPGI (see "Evidence limited to phase II studies," above) has at times led to concern about dedicating resources to the development of reports where the evidence about efficacy is relatively weak at the expense of other high-priority issues in which the evidence is strong. Whether the existence of these various competing priorities has led to delays in guideline development to a point that has compromised care in the province is more difficult to evaluate. Attempts are being made to restructure the guideline development process to minimize the effects of these competing priorities.

Given the complexity of the guideline development process, a second challenge is that it is often not possible to create a fully developed document in the desired time frame. The solution used has been to base PAC decisions on draft guidelines that contain both a full summary of the available evidence and the initial draft recommendations by the relevant disease site group. However, these draft guidelines have not completed the remainder of the guideline cycle that requires a survey of relevant practitioners in Ontario, the practitioner feedback process.² Thus, the frustrations associated with perceptions of unrealistic time lines are at times compounded by the concerns of disease site group members that a less than optimal document is being produced. Failure to conduct the practitioner feedback process may serve to alienate practitioners in Ontario who have come to feel rightly entitled to have their input into such documents. However, these practitioners are often the main advocates for the availability of new drugs for their patients, and it is likely that they appreciate mechanisms that facilitate timely access to these drugs. Indeed, this approach has worked reasonably well from the perspective of the PAC in that it has regularly been possible to make evidence-based funding recommendations simultaneously with drug approval.

Interprovincial Programs
The program described in this article is specific to a single province in Canada—Ontario. However, all Canadian provinces face similar issues around the evaluation and funding of new cancer drugs. Further, although the Canada Health Act does not require all provinces to fund the same "medically necessary" services, substantial differences in availability of cancer treatments in different regions of the country are, conceptually at least, no more tolerable than the emerging practice variations in Ontario that gave rise to the New Drug Funding Program in the first place. Thus, there is a strong rationale for attempting to achieve some consistency in decision making across Canada. This process has, in fact, begun through an association of provincial cancer agencies. Through this mechanism, provinces have begun to share information on drug-funding policies and will begin to share the work of guideline development. This will have the added value of avoiding duplication of the time-consuming and expensive process of the systematic review and synthesis of the evidence.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
As originally described, the CCOPGI developed an approach to formulating evidence-based guidelines that successfully involved oncologists in the process, in part, by divorcing the assessment of evidence from a clinical perspective from the broader policy considerations of the consequences of implementing recommendations. However, in a single-payer, publicly funded cancer system like that of Ontario, the rapid expansion of the role (and cost) of drugs in cancer therapy has made essential the development of a formal process for deciding which drugs to fund. In this article, we have described the function of a separate policy committee and its impact on an ongoing practice guideline development initiative. In our view, this separation of mandate has been successful in isolating policy considerations from evidence-based clinical recommendations. However, there has been an impact on the ability of the CCOPGI to function efficiently in meeting the broader needs of the cancer care community. Furthermore, it is not certain that the ability to keep these considerations separated can be maintained in the long-term. The clinicians who develop evidence-based guidelines are now well aware of the financial implications of funding the drugs they are considering. This, in turn, may eventually affect how carefully they weigh evidence and how narrowly they frame their recommendations. In the meantime, we believe the system that has been developed in Ontario is at least one successful model for integrating the careful assessment of evidence into both clinical practice and health policy. This process further demonstrates the utility of the guideline development cycle for these purposes.

	ACKNOWLEDGMENTS

 
Supported in part by a grant from Cancer Care Ontario and the Ontario Ministry of Health and Long Term Care, Toronto, Ontario, Canada.

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

 
1. Browman GP, Levine MN, Mohide EA, et al: The practice guidelines development cycle: A conceptual tool for practice guidelines development and implementation. J Clin Oncol 13: 502-512, 1995[Abstract/Free Full Text]

2. Browman GP, Newman TE, Mohide EA, et al: Progress of clinical oncology guidelines development using the practice guidelines development cycle: The role of practitioner feedback. J Clin Oncol 16: 1226-1231, 1998[Abstract]

3. Evans WK, Newman TE, Graham I, et al: Lung cancer practice guidelines: Lessons learned and issues addressed by the Ontario Lung Cancer Disease Site Group. J Clin Oncol 15: 3049-3059, 1997[Abstract]

4. Le Chevalier T, Brisgard D, Douillard JY, et al: Randomized study of vinorelbine and cisplatin versus vindesine and cisplatin versus vinorelbine alone in advanced non-small cell lung cancer: Results of a European multi-center trial including 612 patients. J Clin Oncol 12: 360-367, 1994[Abstract]

5. Goss GD, Logan DM, Newman TE, et al: Use of vinorelbine in non-small cell lung cancer. Cancer Prev Control 1: 28-38, 1997[Medline]

6. Bloomfield D, Warr D, Whelan T, et al: Use of bisphosphonates in patients with bone metastases from breast cancer. Curr Oncol 6: 144-154, 1999

7. Imrie K, Esmail R, Buckstein R, et al: Use of rituximab in the treatment of lymphoma: An evidence summary. Curr Oncol 6: 228-235, 1999

8. Klein R: Puzzling out priorities. BMJ 317: 959-960, 1998[Free Full Text]

9. Singer PA, Martin DK, Giacomini M, et al: Priority setting for new technologies in medicine: Qualitative case study. BMJ 321: 1316-1318, 2000[Abstract/Free Full Text]

This article has been cited by other articles:

				R. De Palma, A. Liberati, G. Ciccone, E. Bandieri, M. Belfiglio, M. Ceccarelli, M. Leoni, G. Longo, N. Magrini, M. Marangolo, et al. Developing Clinical Recommendations for Breast, Colorectal, and Lung Cancer Adjuvant Treatments Using the GRADE System: A Study From the Programma Ricerca e Innovazione Emilia Romagna Oncology Research Group J. Clin. Oncol., March 1, 2008; 26(7): 1033 - 1039. [Abstract] [Full Text] [PDF]

				G. P. Browman, B. Manns, N. Hagen, C. R. Chambers, A. Simon, and S. Sinclair 6-STEPPPs: A Modular Tool to Facilitate Clinician Participation in Fair Decisions for Funding New Cancer Drugs J. Oncol. Pract, January 1, 2008; 4(1): 2 - 7. [Abstract] [Full Text] [PDF]

				G. P. Browman, J. Makarski, P. Robinson, and M. Brouwers Practitioners As Experts: The Influence of Practicing Oncologists "in-the-Field" on Evidence-Based Guideline Development J. Clin. Oncol., January 1, 2005; 23(1): 113 - 119. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Pater, J. L. Articles by Cowan, D. H. Search for Related Content PubMed PubMed Citation Articles by Pater, J. L. Articles by Cowan, D. H.