Application of Multi-Criteria Decision Analysis (MCDA) to Prioritize Real-World Evidence Studies for Health Technology Management: Outcomes and Lessons Learned by the Canadian Real-World Evidence for Value of Cancer Drugs (CanREValue) Collaboration.

Multi-criteria decision analysis (MCDA) is a value assessment tool designed to help support complex decision-making by incorporating multiple factors and perspectives in a transparent, structured approach. We developed an MCDA rating tool, consisting of seven criteria evaluating the importance and feasibility of conducting potential real-world evidence (RWE) studies aimed at addressing uncertainties stemming from initial cancer drug funding recommendations. In collaboration with the Canadian Agency for Drugs and Technologies in Health's Provincial Advisory Group, a validation exercise was conducted to further evaluate the application of the rating tool using RWE proposals varying in complexity. Through this exercise, we aimed to gain insight into consensus building and deliberation processes and to identify efficiencies in the application of the rating tool. An experienced facilitator led a multidisciplinary committee, consisting of 11 Canadian experts, through consensus building, deliberation, and prioritization. A total of nine RWE proposals were evaluated and prioritized as low (n = 4), medium (n = 3), or high (n = 2) priority. Through an iterative process, efficiencies and recommendations to improve the rating tool and associated procedures were identified. The refined MCDA rating tool can help decision-makers prioritize important and feasible RWE studies for research and can enable the use of RWE for the life-cycle evaluation of cancer drugs.

The impact of marginalization on diffuse large B-cell lymphoma overall survival: a retrospective cohort study.

The aim of this study was to describe the impact of marginalization on DLBCL overall survival (OS) within the Canadian setting. We conducted a population-based retrospective cohort study of adult patients with newly diagnosed DLBCL in Ontario between 1 January 2005 and 31 December 2017 receiving a rituximab-containing chemotherapy regimen with curative intent followed until 1 March 2020. Our primary exposure of interest was the Ontario Marginalization Index (ON-Marg). The primary outcome was 2-year OS, accounting for patient age, sex, cancer characteristics, comorbidity burden, and rural dwelling status. While two-year overall survival was inferior for individuals in the most deprived marginalization quintile (70.4% Q5 vs. 76.0% Q1), after adjustment for relevant covariates neither the composite ON-Marg nor any of its dimensions had a significant effect. Within the Canadian context, among patients who receive chemotherapy, marginalization may not have a significant association with overall survival when accounting for key patient covariates, lending support for preserved outcomes.

Augmenting clinical trial economic analysis by linking cancer trial data to administrative data: current landscape and future opportunities.

BACKGROUND: Economic analyses based on clinical trial data are costly and time consuming, and alternative methods for performing economic analyses should be explored. OBJECTIVE AND METHODS: In this perspective, we examine the emerging role of administrative data for economic analyses in cancer. RESULTS: Compared with routinely collected clinical trial data, routinely collected administrative data have several strengths including high capture rates for healthcare encounters, less resource utilisation, low rates of misclassification, long follow-up periods and the opportunity to collect data points not traditionally captured in clinical trials. However, there are also limitations including the need for accurate data linkage across multiple databases and systems, the costs and time associated with data linkage, the potential time lag between trial data collection and the availability of administrative data, and limited data on quality of life, toxicity and indirect costs. In this perspective, we identify important barriers and potential solutions to performing economic analyses for oncology using administrative data, and outline strategies to increase research in this field. CONCLUSION: The use of routinely collected administrative data sets for economic analyses of clinical trials presents a unique opportunity that could complement and validate economic analyses based on trial-level data.

Power and sample size calculation for incremental net benefit in cost effectiveness analyses with applications to trials conducted by the Canadian Cancer Trials Group.

BACKGROUND: Historically, a priori power and sample size calculations have not been routinely performed cost-effectiveness analyses (CEA), partly because the absence of published cost and effectiveness correlation and variance data, which are essential for power and sample size calculations. Importantly, the empirical correlation between cost and effectiveness has not been examined with respect to the estimation of value-for-money in clinical literature. Therefore, it is not well established if cost-effectiveness studies embedded within randomized-controlled-trials (RCTs) are under- or over-powered to detect changes in value-for-money. However, recently guidelines (such as those from ISPOR) and funding agencies have suggested sample size and power calculations should be considered in CEAs embedded in clinical trials. METHODS: We examined all RCTs conducted by the Canadian Cancer Trials Group with an embedded cost-effectiveness analysis. Variance and correlation of effectiveness and costs were derived from original-trial data. The incremental net benefit method was used to calculate the power of the cost-effectiveness analysis, with exploration of alternative correlation and willingness-to-pay values. RESULTS: We identified four trials for inclusion. We observed that a hypothetical scenario of correlation coefficient of zero between cost and effectiveness led to a conservative estimate of sample size. The cost-effectiveness analysis was under-powered to detect changes in value-for-money in two trials, at willingness-to-pay of $100,000. Based on our observations, we present six considerations for future economic evaluations, and an online program to help analysts include a priori sample size and power calculations in future clinical trials. CONCLUSION: The correlation between cost and effectiveness had a potentially meaningful impact on the power and variance of value-for-money estimates in the examined cost-effectiveness analyses. Therefore, the six considerations and online program, may facilitate a priori power calculations in embedded cost-effectiveness analyses in future clinical trials.

Impact of Frailty on Health Care Resource Utilization and Costs of Care in Myelodysplastic Syndromes.

PURPOSE: The role of frailty in affecting survival in myelodysplastic syndromes (MDS) is increasingly recognized. Despite this, a paucity of data exists on the association between frailty and other clinically meaningful outcomes including health care resource utilization and costs of care. METHODS: We linked the Ontario subset of the prospective Canadian MDS registry (including baseline patient/disease characteristics) to population-based health system administrative databases. Baseline frailty was calculated from the 15-item MDS-specific frailty scale (FS-15). Primary outcomes were public health care utilization and 30-day standardized costs of care (2019 Canadian dollars) determined for each phase of disease (initial, continuation, and terminal phases). Negative binomial regression was used to assess the association between frailty and health care costs with Poisson regression to explore predictors of hospitalization. RESULTS: Among 461 patients with complete FS-15 scores, 374 (81.1%) had a hospitalization with a mean length of stay of 10.6 days. Controlling for age, comorbidities, Revised International Prognostic Scoring System, and transfusion dependence, the FS-15 was independently associated with hospitalization during the initial (P = .02) and continuation (P = .01) phases but not the terminal disease phase (P = .09). The mean 30-day standardized cost per patient was $8,499 (median, $6,295; interquartile range, $2,798-$11,996), largely driven by cancer clinic visits and hospitalization. On multivariable analysis, the FS-15 was independently associated with costs of care during the initial disease phase (P = .02). CONCLUSION: We demonstrate an association between frailty and clinically meaningful outcomes including hospitalization and costs of care in patients with MDS. Our results suggest that baseline frailty may help to inform patients and physicians of expected outcomes.

Examining the association between oncology drug clinical benefit and the time to public reimbursement.

BACKGROUND: We examined if oncology drug indications with high clinical benefit, as measured by the American Society of Clinical Oncology Value Framework (ASCO-VF) and European Society for Medical Oncology Magnitude of Clinical Benefit Scale (ESMO-MCBS), received public reimbursement status faster than those with lower clinical benefit from the time of pan-Canadian Oncology Drug Review (pCODR) recommendation. METHODS: Oncology drug indications submitted to pCODR between July 2011 and October 2018 were examined. Included indications had a regulatory approval date, completed the pCODR review process, received a positive pCODR recommendation, and been funded by at least one province. Trials cited for clinical efficacy were used to determine the clinical benefit (per ASCO-VF and ESMO-MCBS) of drug indications. RESULTS: Eighty-four indications were identified, yielding 65 ASCO-VF and 50 ESMO-MCBS scores. The mean ASCO-VF and ESMO-MCBS scores were 44.9 (SD = 21.1) and 3.3 (SD = 1.0), respectively. The mean time to provincial reimbursement from pCODR recommendation was 13.2 months (SD = 9.3 months). Higher ASCO-VF and ESMO-MCBS scores had low correlation with shorter time to reimbursement, (ρ = -0.21) and (ρ = 0.24), respectively. In the multivariable analyses, ASCO-VF (p = 0.40) and ESMO-MCBS (p = 0.31) scores were not significantly associated with time to reimbursement. Province and year of pCODR recommendation were associated with time to reimbursement in both ASCO and ESMO models. CONCLUSIONS: Oncology drug indications with higher clinical benefit do not appear to be reimbursed faster than those with low clinical benefit. This suggests the need to prioritize oncology drug indications based on clinical benefit to ensure quicker access to oncology drugs with the greatest benefits.

The Past, Present, and Future of Economic Evaluations of Precision Medicine at the Committee for Economic Analyses of the Canadian Cancer Trials Group.

Precision medicine in oncology poses unique challenges to the generation of clinical and economic evidence used for cost-effectiveness analyses that can inform health technology assessment. The conduct of randomized controlled trials for biomarker-specific therapies targeted towards small populations has limitations in regard to feasibility, timeliness, and cost. These limitations result in associated challenges for groups involved in the generation of economic evidence to inform treatment-related decision making, including the Committee of Economic Analysis (CEA) at the Canadian Cancer Trials Group (CCTG). We provide a high-level description and vision about the new paradigm of clinical trial design, generation of economic evidence, and novel approaches to economic evaluations necessary in the space of precision medicine in oncology in Canada. The CEA's previous approach to precision medicine, including master protocol designs and single-arm studies, is reviewed. Methods and approaches currently under consideration by the CEA and national collaborators, such as the role of real-world and clinical trial evidence in enabling life-cycle assessment of therapies, are explored. Finally, future initiatives being planned in the space of precision medicine at CCTG, such as the incorporation of correlative studies to identify and test high-performing biomarkers in trials, are discussed.

A Prospective Economic Analysis of Early Outcome Data From the Alliance A041202/ CCTG CLC.2 Randomized Phase III Trial Of Bendamustine-Rituximab Compared With Ibrutinib-Based Regimens in Untreated Older Patients With Chronic Lymphocytic Leukemia.

INTRODUCTION: The Alliance A041202/CCTG CLC.2 trial demonstrated superior progression-free survival with ibrutinib-based therapy compared to chemoimmunotherapy with bendamustine-rituximab (BR) in previously untreated older patients with chronic lymphocytic leukemia. We completed a prospective trial-based economic analysis of Canadian patients to study the direct medical costs and quality-adjusted benefit associated with these therapies. METHODS: Mean survival was calculated using the restricted mean survival method from randomization to the study time-horizon of 24 months. Health state utilities were collected using the EuroQOL EQ-5D instrument with Canadian tariffs applied to calculate quality-adjusted life years (QALYs). Costs were applied to resource utilization data (expressed in 2019 US dollars). We examined costs and QALYs associated ibrutinib, ibrutinib with rituximab (IR), and BR therapy. RESULTS: A total of 55 patients were enrolled; two patients were excluded from the analysis. On-protocol costs (associated with protocol-specified resource use) were higher for patients receiving ibrutinib (mean $189,335; P < 0.0001) and IR (mean $219,908; P < 0.0001) compared to BR (mean $51,345), driven by higher acquisition costs for ibrutinib. Total mean costs (over 2-years) were $192,615 with ibrutinib, $223,761 with IR, and $55,413 with BR (P < 0.0001 for ibrutinib vs. BR and P < 0.0001 for IR vs. BR). QALYs were similar between the three treatment arms: 1.66 (0.16) for ibrutinib alone, 1.65 (0.24) for IR, and 1.66 (0.17) for BR; therefore, a formal cost-utility analysis was not conducted. CONCLUSIONS: Direct medical costs are higher for patients receiving ibrutinib-based therapies compared to chemoimmunotherapy in frontline chronic lymphocytic leukemia, with the cost of ibrutinib representing a key driver.

Assessing the benefit of cancer drugs approved by the European Medicines Agency using the European Society for Medical Oncology Magnitude of Clinical Benefit Scale over time.

BACKGROUND: Increasingly, cancer drugs are being approved based on surrogate measurements of efficacy. Clinically meaningful data, such as overall survival (OS) and quality of life, are often only presented in subsequent publications. We examined if the clinical benefit of cancer drugs, as measured by the European Society for Medical Oncology Magnitude of Clinical Benefit Scale (ESMO-MCBS), improves post-European Medicines Agency (EMA) approval as more data emerges. METHODS: Cancer drug indications approved by the EMA from January 2006 to December 2016 were reviewed and trials cited for efficacy were identified. Primary and subsequent publications (up to December 2019) of scorable trials were included. Changes in benefit over time were measured using ESMO-MCBS thresholds for non-curative (≥4 for substantial, =3 for intermediate and ≤2 for low benefit) and curative intent (A or B for major benefit) scoring. RESULTS: Fifty-five non-curative and two curative intent trials were included. At approval, 29.1% of non-curative trials were substantial, 45.5% intermediate and 25.5% low benefit. For curative intent trials, one displayed major benefit and one displayed no major benefit. We identified 82 subsequent publications for reassessment. A change in ESMO-MCBS classification was seen in 34.5% of non-curative trials (11 raised and 8 lowered). At 3-year reassessment, 36.4% of non-curative trials were substantial, 34.5% intermediate and 29.1% low benefit. Both curative trials showed no major benefit at reassessment. CONCLUSION: As over a third of trials changed classification, in either direction, reassessing the ESMO-MCBS score of approved cancer drugs may help to inform patients and ensure ongoing relevance of regulatory and reimbursement decisions.

A descriptive cost-analysis of MYX.1/MCRN003, a phase 2 clinical trial evaluating high-dose weekly carfilzomib, cyclophosphamide, and dexamethasone in relapsed and refractory multiple myeloma.

BACKGROUND: The prevalence of multiple myeloma is increasing and there is a need to evaluate escalating therapy costs (Canadian Cancer Statistics A, 2020). The MYX.1 phase II trial showed that high-dose weekly carfilzomib, cyclophosphamide, and dexamethasone (wKCD) is efficacious in relapsed and refractory disease. We conducted a descriptive cost analysis, from the perspective of the Canadian public healthcare system, using trial data. METHODS: The primary outcome was the mean total cost per patient. Resource utilization data were collected from all 75 trial patients over a trial time horizon. Costs are presented in Canadian dollars (2020). RESULTS: The cost of treatment was calculated from the time of patient (pt) enrollment until the second data lock. The mean total cost was $203 336.08/pt (range $17 891.27-$505 583.55) Canadian dollars (CAD, where 1 CAD = 0.67 Euro (EUR)) and $14 081.45/pt per cycle. The median number of cycles was 15. The predominant cost driver was the cost of chemotherapy accounting for an average of $179 332.78/pt or $12 419.17/pt per cycle. Carfilzomib acquisition accounted for the majority of chemotherapy costs - $162 471.65/pt or $11 251.50/pt per cycle. Fifty-six percent (56%) of patients had at least one hospitalization during the trial period with an average cost of $12 657.86 per hospitalization. Three patients developed thrombotic microangiopathy (TMA) with an average cost of $18 863.32/pt including the cost of hospitalizations and therapeutic plasma exchange. CONCLUSIONS: High-dose wKCD is an active triplet regimen for relapsed and refractory multiple myeloma (RRMM) associated with reduced total cost compared with twice-weekly carfilzomib-based regimens.

Minimization of resource utilization data collected within cost-effectiveness analyses conducted alongside Canadian Cancer Trials Group phase III trials.

BACKGROUND: Cost-effectiveness analyses embedded within randomized trials allow for evaluation of value alongside conventional efficacy outcomes; however, collection of resource utilization data can require considerable trial resources. METHODS: We re-analyzed the results from four phase III Canadian Cancer Trials Group trials that embedded cost-effectiveness analyses to determine the impact of minimizing potential cost categories on the incremental cost-effectiveness ratios. For each trial, we disaggregated total costs into component incremental cost categories and recalculated incremental cost-effectiveness ratios using (1) only the top 3 cost categories, (2) the top 5 cost categories, and (3) all cost components. Using individual trial-level data, confidence intervals for each incremental cost-effectiveness ratio simulation were generated by bootstrapping and descriptively presented with the original confidence intervals (and incremental cost-effectiveness ratios) from the publications. RESULTS: Drug acquisition costs represented the highest incremental cost category in three trials, while hospitalization costs represented the other consistent cost driver and the top incremental cost category in the fourth trial. Recalculated incremental cost-effectiveness ratios based on fewer cost components (top 3 and top 5) did not differ meaningfully from the original published results. Based on conventional willingness-to-pay thresholds (US$50,000-US$100,000 per quality-adjusted life-year), none of the re-analyses would have changed the original perception of whether the experimental therapies were considered cost-effective. CONCLUSIONS: These results suggest that the collection of resource utilization data within cancer trials could be narrowed. Omission of certain cost categories that have minimal impact on incremental cost-effectiveness ratio, such as routine laboratory investigations, could reduce the costs and undue burden associated with the collection of data required for cancer trial cost-effectiveness analyses.

Evaluating the Indirect Costs of Care Associated with Salvage Chemotherapy for Relapsed and Refractory Aggressive-Histology Lymphoma: A Subset Analysis of the Canadian Cancer Trials Group (CCTG) LY.12 Clinical Trial.

We conducted an analysis of indirect costs alongside the LY.12 randomized trial in patients with relapsed or refractory (R/R) aggressive non-Hodgkin lymphoma (NHL). Lost productivity data for Canadian patients and caregivers in the trial were collected at baseline and with each chemotherapy cycle pre-transplant, using an adapted Lost Productivity questionnaire. Mean per patient indirect costs were CAD 2999 for patients in the GDP arm and CAD 3400 in the DHAP arm. A substantial majority was not working or had to reduce their workload during this treatment time. Salvage chemotherapy for R/R aggressive NHL is associated with significant indirect costs to patients and their caregivers.

Mammographic Surveillance in Older Women With Breast Cancer in Canada and the United States: Are We Choosing Wisely?

PURPOSE: Guidelines on mammographic surveillance after breast cancer treatment have been disseminated internationally and incorporated into Choosing Wisely recommendations to reduce low-value care. However, adherence within different countries before their publication is unknown. METHODS AND MATERIALS: Low-value mammography, defined as "short-interval" (within 6 months of radiation) or "high-frequency" (>1 within 12 months of radiation), was compared in Medicare fee-for-service in the United States and Ontario, Canada. Women ≥65 years diagnosed with breast cancer who underwent breast-conserving therapy with a minimum of 24 months of follow-up were included (n = 19,715 United States; 6479 Ontario). Secondary outcomes were patient and physician characteristics associated with discordance. RESULTS: Short-interval mammography was higher in the United States than in Ontario (55.9% vs 38.0%, P < .001), as was high-frequency (39.6% vs 7.9%, P < .001). In Ontario, younger age (42% ≥85 vs 58% <74 years, P < .001) and chemotherapy (69% vs 51%, P < .001) were associated with short-interval mammography; in the United States, age, earlier diagnosis year, stage, chemotherapy, rurality, and academic center treatment were associated with greater use. Chemotherapy was associated with high-frequency mammography in both countries (13% vs 7% in Ontario, P < .001; 69% vs 51% in United States, P = .02); younger age, earlier diagnosis year, stage, and nonacademic center treatment were associated in the United States. In both countries, radiation oncologists had the highest proportion of providers ordering low-value mammograms. CONCLUSIONS: Despite significant evidence guiding surveillance mammography recommendations, there are high rates of short-interval mammography in both the United States and Ontario, and high rates of high-frequency mammography in the United States. Further international efforts, such as Choosing Wisely, are needed to reduce low-value mammography.

Real-World Costs of Azacitidine Treatment in Patients With Higher-Risk Myelodysplastic Syndromes/Low Blast-Count Acute Myeloid Leukemia.

PURPOSE: Azacitidine (AZA) is a standard of care for higher-risk myelodysplastic syndrome (MDS)/low blast-count acute myeloid leukemia (AML). Despite this, there is a paucity of data on the real-world health care resource utilization costs of AZA in this population. METHODS: We linked the Ontario AZA MDS registry-higher-risk MDS/low blast-count AML-to population-based health system administrative databases. Patients were observed for 24 months after first AZA and censored at the earliest of 90 days after last AZA, date of death, time of AML induction/stem-cell transplantation, or March 31, 2016. Costs (2015 Canadian dollars) were expressed as standardized mean and median 28-day costs. Univariable quantile regression was used to explore the association of baseline patient and disease characteristics and median cost. Multivariable quantile regression was used to explore predictors of median costs. RESULTS: Among 877 patients in the registry, mean standardized 28-day cost per patient was $17,638 (median, $15,272; interquartile range [IQR], $11,869-$19,580) and $13,450 (median, $11,043; IQR, $7,981-$14,882) excluding the cost of AZA. Major nondrug drivers of cost were cancer clinic visits and inpatient care (mean standardized 28-day cost, $4,631; median, $1,558; IQR, $238-$4,961). Transfusion dependence at AZA initiation (P = .001) and greater comorbid disease burden (P = .009) were independently associated with increased cost. CONCLUSION: Our cohort of patients with uniformly higher-risk MDS/low blast-count AML treated with AZA demonstrates substantial costs of care above and beyond the cost of AZA alone. These results provide insight into the costs of AZA in the real world with implications for resource allocation.

Laboratory Workup of Lymphoma in Adults.

OBJECTIVES: The diagnostic workup of lymphoma continues to evolve rapidly as experience and discovery lead to the addition of new clinicopathologic entities and techniques to differentiate them. The optimal clinically effective, efficient, and cost-effective approach to diagnosis that is safe for patients can be elusive, in both community-based and academic practice. Studies suggest that there is variation in practice in both settings. THE AIM OF THIS REVIEW IS TO: develop an evidence-based guideline for the preanalytic phase of testing, focusing on specimen requirements for the diagnostic evaluation of lymphoma. METHODS: The American Society for Clinical Pathology, the College of American Pathologists, and the American Society of Hematology convened a panel of experts in the laboratory workup of lymphoma to develop evidence-based recommendations. The panel conducted a systematic review of the literature to address key questions. Using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach, recommendations were derived based on the available evidence, the strength of that evidence, and key judgments as defined in the GRADE Evidence to Decision framework. RESULTS: Thirteen guideline statements were established to optimize specimen selection, ancillary diagnostic testing, and appropriate follow-up for safe and accurate diagnosis of indolent and aggressive lymphoma. CONCLUSIONS: Primary diagnosis and classification of lymphoma can be achieved with a variety of specimens. Application of the recommendations can guide decisions about specimen suitability, diagnostic capabilities, and correct utilization of ancillary testing. Disease prevalence in patient populations, availability of ancillary testing, and diagnostic goals should be incorporated into algorithms tailored to each practice environment.

Health system costs for cancer medications and radiation treatment in Ontario for the 4 most common cancers: a retrospective cohort study.

BACKGROUND: Previous costing and resource estimates for cancer have not been complete owing to lack of comprehensive data on cancer-related medication and radiation treatment. Our objective was to calculate the mean overall costs per patient of cancer-related medications and radiation, as well as by disease subtype and stage, in the first year after diagnosis for the 4 most prevalent cancers in Ontario. METHODS: We conducted a retrospective cohort study using provincial health administrative databases to identify population health system resources and costs for all patients diagnosed with breast, colorectal, lung or prostate cancer between Jan. 1, 2010, and Dec. 31, 2015 in Ontario. The primary outcome measure was the overall average cost per patient in the 365 days after diagnosis for cancer-related medications and radiation treatment, calculated with the use of 2 novel costing algorithms. We determined the cost by disease, disease subtype and stage as secondary outcomes. RESULTS: There were 168 316 Ontarians diagnosed with cancer during the study period, 50 141 with breast cancer, 38 108 with colorectal cancer, 34 809 with lung cancer and 45 258 with prostate cancer. The mean per-patient cost for cancer-related medications was $8167 (95% confidence interval [CI] $8023-$8311), $6568 (95% CI $6446-$6691), $2900 (95% CI $2816-$2984) and $1211 (95% CI $1175-$1247) for breast, colorectal, lung and prostate cancer, respectively. The corresponding mean radiation treatment costs were $18 529 (95% CI $18 415-$18 643), $15 177 (95% CI $14 899-$15 456), $10 818 (95% CI $10 669-$10 966) and $16 887 (95% CI $16 648-$17 125). In general, stage III and IV cancers were the most expensive stages for both medications and radiation across all 4 disease sites. INTERPRETATION: Our work updates previous costing estimates to help understand costs and resources critical to health care system planning in a single-payer system. More refined costing estimates are useful as inputs to allow for more robust health economic modelling and health care system planning.

Cost-effectiveness of first-line treatment options for patients with advanced-stage Hodgkin lymphoma: a modelling study.

BACKGROUND: Several strategies are available for the initial treatment of advanced-stage Hodgkin lymphoma, but the optimal strategy in terms of cost-effectiveness is unclear. The aim of this study was to compare the quality-adjusted effectiveness and costs of five modern treatment options for transplantation-eligible patients with newly diagnosed advanced-stage Hodgkin lymphoma. METHODS: A Markov decision-analytic model was developed using a 20-year time horizon. Five of the most common treatment approaches were selected based on clinical experience and expert opinion: (1) six cycles of doxorubicin, bleomycin, vinblastine, dacarbazine (ABVD), including data from the HD2000 trial, Viviani and colleagues, and EORTC trial; (2) six cycles of bleomycin, etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine, and prednisone (BEACOPP; from the HD15 trial or PET-adapted as in the HD18 trial, two initial cycles of BEACOPP followed by four additional cycles for patients with a positive PET and either two or four additional cycles of BEACOPP for patients with a negative PET); (3) PET-adapted escalation (as in the RATHL trial, two cycles of standard ABVD chemotherapy followed by an additional four cycles of ABVD or AVD in PET-negative patients and four cycles of BEACOPP in PET-positive patients); (4) six cycles of brentuximab vedotin, doxorubicin, vinblastine, dacarbazine (A-AVD) or ABVD as in the Echelon-1 trial; and (5) PET-adapted de-escalation (as in the AHL2011 trial, two cycles of BEACOPP followed by PET2 scan; PET-positive patients received two additional BEACOPP cycles and PET-negative patients received two cycles of ABVD; at PET4, PET-negative patients completed two further cycles of either ABVD or BEACOPP depending on what they received after PET2, and PET-positive patients received salvage therapy). Note that all uses of BEACOPP in these strategies were BEACOPPescalated. The randomised groups of interest from these studies comprised 4255 patients enrolled between April, 2000, and January, 2016. Baseline probability estimates and utilities were derived from the included trials in addition to a systematic review of published studies. A Canadian public health payer's perspective was considered (CAN$1=US$0·74) and adjusted for inflation for 2018. All costs and benefits were discounted by 1·5% per year because life-years now are more valuable than future potential life-years. FINDINGS: Probabilistic analyses (10 000 simulations) showed that, for a willingness-to-pay threshold of CAN$50 000, a PET-adapted de-escalation strategy based on AHL2011 was more cost-effective 87% of the time. This strategy had the highest number of life-years (14·6 years [95% CI 13·7-15·1]) and quality-adjusted life years (13·2 years [95% CI 10·2-14·4]), and the lowest direct costs ($53 129 [95% CI 31 914-94 446]) compared with the other treatment regimens. Sensitivity analyses showed that the model was robust to key variables, including probability of treatment-related mortality, relapse, frequency of secondary malignancy, death from secondary malignancy, and probability of infertility after BEACOPP. INTERPRETATION: Our results suggest that, when considering cost, effectiveness, and short and long-term toxicities, the preferred treatment strategy for patients with newly diagnosed advanced-stage Hodgkin lymphoma is the PET-adapted de-escalation regimen starting with BEACOPP and de-escalating to ABVD as appropriate. Although our findings do not provide an absolute best treatment approach for clinicians to follow for all patients, they can contribute to shared decision making between patients and treating physicians. FUNDING: None.

The impact of pricing strategy on the costs of oral anti-cancer drugs.

BACKGROUND: The soaring costs of anti-cancer drugs pose a threat to the sustainability of cancer care. The pricing strategy chosen by manufacturers can impact the costs of oral anti-cancer drugs during dose modifications, but this issue remains under-recognized in the literature. In general, with the flat pricing strategy, there is a single fixed price for each tablet regardless of dosage strength, whereas with linear pricing, the price of each tablet increases with its dose. We hypothesize that flat pricing will have increased drug costs compared to linear pricing during dose reductions since the cost remains fixed despite decreased dose requirements. This practice may have significant financial implications considering the high costs, extensive utilization, and frequent occurence of dose reductions with anti-cancer drugs.  METHODS: Oral anti-cancer drugs reviewed by the pan-Canadian Oncology Drug Review program between 2011 and 2018 were identified. The cost per mg and cost per 28-day cycle were calculated for dose levels -2 to +2. The percent change in cost per mg and cost per cycle during dose modifications from the standard dose were calculated. We conducted Mann-Whitney U and Fisher-exact tests to compare the association between drug costs during dose reductions and pricing strategy. RESULTS: In this study, 30 oral anti-cancer drugs for use in 41 indications were analyzed; 44% of drugs used linear pricing and 56% used flat pricing. Dose reductions increased the mean cost per mg for drugs with linear pricing by 14.7% (range: 0%-50%) at dose level -1 and 17.2% (range: 0%-50%) at dose level -2 and flat pricing by 60.8% (range: 19%-100%) at dose level -1 and 99.1% (range: 0%-300%) at dose level -2. The cost per mg was significantly increased in drugs using flat pricing compared to linear pricing when dose reduction to either level -1 (P = 0.010) or level -2 (P = 0.006) occurred. The mean cost per cycle was decreased for drugs using linear pricing by 20.9% (range: -40% to 0%) at dose level -1 and 48.7% (range: -60% to -25%) at dose level -2 and flat pricing by 0.8% (range -6% to 0%) at dose level -1 and 11.0% (range: -50% to 100%) at dose level -2. The cost per cycle was significantly decreased in drugs with linear pricing compared to flat pricing when the standard dose is reduced to either dose level -1 (P = 0.005) or dose level -2 (P = 0.026). CONCLUSIONS: Overall, flat pricing had significantly greater costs compared to linear pricing during dose reductions of anti-cancer drugs. While there is a general expectation that the cost of drugs should decrease with dose reduction, drugs with flat pricing were generally found to have increased cost per mg and no change in the cost per cycle. The resultant increased spending on drug acquisition (despite purchasing lower doses) lead to financial wastage, which has significant implications on cost-effectiveness considerations and budgets. Future economic evaluations should take into consideration the hidden costs associated with dose reductions of flat priced drugs.