Value in Health Supplementary Materials: Economic Implications of 21-Gene Breast Cancer Risk Assay from the Perspective of an Israeli-Managed Health-Care Organization

ARTICLE SUPPLEMENTARY INFORMATION
(Supporting information is provided by the author to support the content
of the parent article)

Economic Implications of 21-Gene Breast Cancer Risk Assay from the Perspective of an Israeli-Managed Health-Care Organization

Shmuel H. Klang, Ariel Hammerman, Nicky Liebermann, Noa Efrat, Julie Doberne, John Hornberger

– Value in Health Supplemental Information

Appendix I. Grading evidence

Source: Hornberger et al. Leuk Lymph 2008.

Appendix II. Validity of economic analyses

Attributes	Checklist	Section in report	Page
Structure
Statement of decision problem/objective	To assess the validity of the cost-effectiveness of Oncotype DX compared with traditional prognostic pathways in the diagnosis and treatment of ER positive, node-negative early stage breast cancer from an Israeli healthcare provider perspective. A secondary aim is to assess the factors that influence the cost-effectiveness of Oncotype DX use.	Introduction	4
Justification of modeling approach	For most models in oncology, a Markov modeling framework often is the preferred framework.	Methods	4
Statement of scope/perspective	Third-party payer perspective.	Introduction	4
Thorough description of all assumptions & strategies/comparators	All assumptions are described in the Methods section.	Methods	4-7
Definition of relevant health states	Utility scores for the following health states were obtained from the literature: breast cancer during chemotherapy and breast cancer recurrence. We omit computing utilities and QALY differences for second primary cancer caused by chemotherapy, but comment on this in the manuscript.	Methods: Health utilities; Table 1	5; 18
The appropriateness of Markov cycle length	Breast cancer recurrence can occur beyond 10 years after initial diagnosis and treatment. Transition rates were interpolated from 10-year recurrence rates published in the literature. Whether the cycle length is 1 month or 1 year does not affect the estimates of survival associated with recurrence.	Methods	5
Data
All relevant data sources identified and appropriately used	We report literature sources to determine (1) annual risk of death by age for women, (2) costs for chemotherapy and supportive care, (3) health utilities, and (4) rates of adjuvant chemotherapy recommended before compared with actual use after knowledge of Oncotype DX risk status.	Methods: Risk of recurrence and death; Costs; Health utilities	5
Follows well-established guidelines on literature retrieval and synthesis	Yes.	-	-
Proper grading of evidence	We assessed the grade of the evidence using a published system for cancer developed by one of the co-authors (JH).	Appendix 1	22
Proper analysis and use of primary data	Assumptions are evidenced based and have been chosen to reflect no bias in interpretation. Summary of parameters used in the model are transparently displayed.	Methods: Health utilities; Table 1	5; 18
Discount both benefits and costs	We discounted cost and benefits at a fixed annual rate of 3%.	Methods: Costs	5
Examine appropriate patient subgroups	The target population was restricted to those reported in pivotal validation studies – women with node-negative, estrogen-receptor-positive early-stage breast cancer. We examined subgroups based on recurrence score (low, intermediate, and high), and extent of cancer (pN0, pN1mi). Data on other subgroups have been reported in congresses (e.g., node-positive disease, and women receiving aromatase inhibitors from the ATAC trial). Because these data have not been published in peer-reviewed journal nor studied at CHS, we did not report the implications to these populations.		-
Include half-cycle correction	Half-cycle correction included in the model.	-	-
Extrapolation of data beyond duration of the available data (e.g., in a clinical trial) may be appropriate depending on whether the interventions under consideration have implications beyond the trial duration	We extrapolated the implications on recurrence, and subsequent mortality, to women tested at the CHS. This is common approach for cost-effectiveness of interventions for breast cancer given the long-term implications of this disease.	Methods	5
Uncertainty
Instability of the model and its findings under conditions different than the base reference case are assessed	Variation ranges from 75% - 125% of base case value unless there are special circumstances for a particular parameter.	Methods: Sensitivity analyses	6
Examine variations in model structure and input parameters	1-way sensitivity analyses conducted. Examined cycle lengths of 1 month versus 1 year.	Methods: Sensitivity analyses	6
Parameters that most influence the findings of the analyses are highlighted	(1)Patient mean age, (2) risk of recurrence in low-risk group, (3) risk of recurrence in intermediate-risk group	Results	8
Indicate areas of future research	Assess whether utility findings are maintained in other settings	Discussion	9-10
Consistency
Internal consistency –mathematical programs used for the analyses should be devoid of errors	Devoid of errors.	-	-
Internal consistency – changes in model parameters should provide results that are consistent with theory	Devoid of errors.	-	-
Face validity – amenable to intuitive explanation	Face validity high.	-	-
Calibration – to the extent that data is available that was not also used to develop the model, the analyses should be assessed for their ability to predict the results of the new dataset, called predictive validity	Not applicable as we used all available data.	-	-
Peer review	Analysis currently undergoing peer review by Value in Health. [To be revised if and upon receiving acceptance of the manuscript for publication.]	-	-

Appendix III. The Quality of Health Economic Studies (QHES) Instrument

	Questions	Points	Yes	No
1	Was the study objective presented in a clear, specific, and measurable manner?	7	X
2	Were the perspective of the analysis (societal, third-payer, etc.) and reasons for its selection stated?	4	X
3	Were variable estimates used in the analysis from the best available source (i.e., randomized control trial – best, expert opinion – worst)?	8	X
4	If estimates came from a subgroup analysis, were the groups prespecified at the beginning of the study?	1	X
5	Was uncertainty handled by (1) statistical analysis to address random events, (2) sensitivity analysis to cover a range of assumptions?	9	X
6	Was incremental analysis performed between the alternatives for resources and costs?	6	X
7	Was the methodology for data abstraction (including the value of health states and other benefits) stated?	5	X
8	Did the analytic horizon allow time for all relevant and important outcomes? Were benefits and costs that went beyond 1 year discounted (3% to 5%) and justification given for the discount rate?	7	X
9	Was the measurement of costs appropriate and the methodology for the estimation of quantities and unit costs clearly described?	8	X
10	Were primary outcome measures(s) for the economic evaluation clearly stated and did they include the major short-term was justification given for the measure/scales used?	6	X
11	Were the health outcomes measures/scales valid and reliable? If previously tested valid and reliable measures were not available, was justification given for the measures/scales used?	7	X
12	Were the economic model (including structure), study methods and analysis, and the components of the numerator and denominator displayed in a clear, transparent manner?	8	X
13	Were the choices of economic model, main assumptions, and limitations of the study stated and justified?	7	X
14	Did the author(s) explicitly discuss the direction and magnitude of potential biases?	6	X
15	Were the conclusions/recommendations of the study justified and based on the study results?	8	X
16	Was there a statement disclosing the source of funding for the study?	3	X
	Total Points	100	100	0

Value in Health Article Supplementary Information