The target population for our research are potential BRCA positive parents (mother or father). Costs of IVF/PGT-M, with BRCA negative embryo selection and transfer versus natural conception with a 50% chance of BRCA positive newborn, were compared using a Markovian process decision analysis model (Fig. 1). The model assumed that all women in the IVF/ PGT-M arm would undergo ovarian stimulation and ovum pick. After intracytoplasmic sperm injection (ICSI), fertilization and embryo biopsy, a BRCA negative embryo would be selected for embryo transfer. In the natural conception arm, women would conceive naturally, without any manipulation, and assume a 50% chance of bearing a BRCA positive newborn, as BRCA is a dominant gene. BRCA negative newborns would assume to have the general populations' life-time-risk of breast and ovarian cancer. At age 40, BRCA positive females would be offered risk reduction salpingo-oophorectomy (RRSO) for ovarian cancer prevention (10). Management of breast cancer risk for BRCA positive females would include screening with yearly breast MRI/ultrasound or risk reduction mastectomy (RRM) for breast cancer prevention (10). Costs of these two strategies were compared using quality-adjusted life years (QALYs), which reflect both quality and quantity of life lived. Incremental cost effectiveness ratio (ICER) was used for cost-effectiveness analysis, compared to a willingness to pay threshold.
IVF/PGT-M arm
In this arm, all couples with either one BRCA positive parent were assumed to undergo ovarian stimulation, ovum pick-up and ICSI. After fertilization, embryos would be biopsied and only BRCA negative embryos would be transferred. All women were assumed to have IVF before age 35, where IVF success rates are optimal (11), as the knowledge of BRCA carrier state among afflicted families would to be known at an early age. The following steps were assumed in the IVF/ PGT-M arm, probabilities were taken from the ESHRE PGD consortium data collection regarding success of IVF/ PGT-M cycles preformed for single, autosomal dominant gene disorders (8): (1) ovarian stimulation and ovum pick-up, on average, 13 oocytes are retrieved per IVF cycle among patients < 35 years, undergoing IVF/PGD for dominant, single gene disorders, (2) insemination with ICSI, 82% of oocytes are successfully inseminated, (3) fertilization,76% of inseminated oocytes are fertilized, (4) embryo biopsy, 79% of embryos are successfully biopsied.
Accordingly, per cycle start, out of 13 oocytes retrieved, 6.4 embryos will be available for biopsy, half of which, 3.20 embryos, will be BRCA negative embryos, available for transfer. The first embryo will be used for fresh embryo transfer, the remaining embryos will be frozen for future frozen thawed embryo transfer cycles.
The overall live birth rate per embryo transfer for couples undergoing IVF/ PGT-M for genetic disorders reaches 45.8% (12). The model assumed that after the first fresh embryo transfer, 54.2% of couples who fail would have a second thawed embryo transfer, while 29% of couples who will fail the second transfer, will have a third, thawed embryo transfer. Therefore, 3 available healthy, non BRCA mutated embryos, with a 45.8% live birth rate per embryo transfer would result in an overall 84% chance of a live, BRCA negative baby, from one cycle of fresh embryo transfer and two more cycles of thawed embryo transfers. The remaining 16% of couples who failed the IVF/ PGT-M path would go back to the natural conception arm.
Natural conception arm
In this arm, couples are assumed to conceive naturally. As BRCA is a dominant gene, these couples would have a 50% chance of bearing a BRCA positive fetus, of those, 50% would be females, positive for BRCA gene mutations with respective increased risk of breast and ovarian cancer. As the chances of other BRCA related malignancies among male mutation carriers are much lower than among female mutation carriers these were not included in our model. BRCA negative newborns were assumed to have the natural populations' breast and ovarian cancer risk.
Model costs
In vitro fertilization, PGT-M and embryo transfer, as well as BRCA screening and other treatment related costs including, RRSO, RRM and breast and ovarian cancer treatment costs were received from the Israeli ministry of health 2020 pricing list according to specified codes with conservative assumptions of health resources utilities. Elaborate costs used in the model, from payer perspective, are shown in supplementary Tables 1, 2, 3, 4 and 5 for, ovarian cancer treatment, breast cancer treatment, IVF/PGT-M, BRCA positive patients follow-up and total costs, respectively. All costs were discounted at 3%.
Probabilities
Model probabilities are presented in Table 1. The probability of being at the end of each arm of the Markovian model was calculated by multiplying the probabilities of events along the arms' path. Stage distribution and Kaplan Meir survival curves for each stage were used to calculate mortality rates of subjects who developed breast cancer or ovarian cancer, based on the Surveillance, Epidemiology and End Results program data base (SEER) (13, 14). Survival curves were extrapolated by fitting Weibull distribution using the Nelder-Mead Algorithm.
Table 1
Probabilities used in model
Probability
|
Description
|
Probability
|
Range assumed
|
P1
|
Probability of male newborn
|
0.5
|
|
P2
|
Probability newborn is BRCA carrier
|
0.5
|
|
P3
|
Probability that a non-carrier will experience ovarian cancer 13
|
0.0128
|
0.0005–0.0989
|
P4
|
Probability that a non- carrier will experience breast cancer14
|
0.13
|
0.11–0.14
|
P5
|
Probability that carrier will undergo RRSO 20,21
|
0.65
|
0.3–0.75
|
P6
|
Probability that BRCA carrier will undergo RRM 25
|
0.16
|
0.13–0.3
|
P7
|
Reduction in risk of ovarian cancer from RRSO 26,27
|
0.8
|
0.8–0.96
|
P8
|
Probability that BRCA carrier without RRSO will get OC 28
|
0.2987
|
0.24–0.35
|
P9
|
Reduction in risk of breast cancer from RRM and RRSO27
|
0.91
|
0.78–0.99
|
P10
|
Probability that a BRCA carrier without RRM will experience breast cancer 28
|
0.53
|
0.44–0.62
|
P11
|
Reduction in risk of breast cancer from RRSO 22,24,29
|
0.0
|
0.37–0.65
|
P12
|
Reduction in breast cancer risk from RRM without RRSO 30
|
0.91
|
0.62–0.98
|
P13
|
Probability of live newborn with IVF/PGD (1 cycles of fresh ET and 2 cycles of thawed ET)
|
0.84
|
0.7–0.9
|
RRSO: risk reduction salpingo-oophorectomy, RRM: risk reduction mastectomy, IVF: in-vitro fertilization, PGD: pre-gestational diagnosis, ET: embryo transfer |
Quality adjusted life years and incremental cost effectiveness ratio
Value of health benefits for each strategy (IVF/ PGT-M versus natural conception) were calculated using quality adjusted life years (QALY's). QALY's are calculated by multiplying the utility value associated with a given state of health by the number of years lived in that state, where QALY of one reflects one year lived in perfect health and QALY of zero represents death state. The incremental cost effectiveness ratio was then calculated by using the formula: (average cost IVF/ PGT-M – average cost natural conception) / (average QALY IVF/ PGT-M – average QALY natural conception). The incremental cost effectiveness ration (ICER) calculated, enabled to determine whether offering IVF/ PGT-M to BRCA positive families is cost effective or not, compared to willingness to pay threshold. An intervention was defined as cost effective if the ICER per QALY is between 1–3 times per capita gross national product (GNP). GNP in Israel is estimated at 42,160 USD, equivalent to 142,500 new Israeli Shekels (NIS) (15). Interventions below 1 GNP per capita are considered very cost effective (16).
Sensitivity analysis
One-way sensitivity analysis was conducted with all variables to evaluate model uncertainties. In addition, a probabilistic sensitivity analysis (Monte Carlo simulation) was conducted with all variables, using 100 trials, each included 10,000 couples.