This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Formative Research, is properly cited. The complete bibliographic information, a link to the original publication on https://formative.jmir.org, as well as this copyright and license information must be included.
Approximately 62% of patients with breast cancer with a pathogenic variant (
The study aims to develop a personalized risk management decision support tool for carriers of a pathogenic variant (
We developed a Bayesian network model of a hypothetical cohort of carriers of
Absence of adjuvant chemotherapy was the most powerful factor that was linked to a dramatic decline in survival. There was a negligible decline in the mortality in patients with triple-negative breast cancer, who received no chemotherapy and underwent any secondary risk–reducing strategy, compared with surveillance. The potential survival benefit from any risk-reducing strategy was more modest in patients with triple-negative breast cancer who received chemotherapy compared with patients with luminal breast cancer. However, most patients with triple-negative breast cancer in stage I benefited from bilateral risk-reducing mastectomy and risk-reducing salpingo-oophorectomy or just risk-reducing salpingo-oophorectomy. Most patients with luminal stage I/II unilateral breast cancer benefited from bilateral risk-reducing mastectomy and risk-reducing salpingo-oophorectomy. The impact of risk-reducing salpingo-oophorectomy in patients with luminal breast cancer in stage I/II increased with age. Most older patients with the
Our study showed that it is mandatory to consider the complex interplay between the types of
Breast cancer is the most common cancer and the leading cause of cancer mortality among women in economically developed and developing countries [
Patients with breast cancer with
Current guidelines describe different cancer risk management strategies: enhanced breast cancer screening, risk-reducing bilateral mastectomy (RRBM), risk-reducing salpingo-oophorectomy (RRSO), and chemoprevention [
Previous studies have revealed that different factors, such as the patient’s age at first breast cancer diagnosis, the type of pathogenic variant (ie,
The aim of this study is to develop a personalized risk management guideline for carriers of the pathogenic variants of
We have developed a temporal Bayesian network model to estimate the expected overall survival of a hypothetical cohort of
All risk estimates were converted into yearly estimates by conditional probabilities, depending on the original metric published in the literature with needed conversions (eg, risk for years between 5 and 10 used 10-year estimates converted to actual follow-up) [
If hazard ratios were given, survival for each group was computed as ref ^ hr, where “ref” is the expected survival for the reference group.
The simulation was run for a yearly follow-up of 40 years after diagnosis, yielding a temporal Bayesian network with 40 nodes per temporal variable (eg, ipsilateral recurrence). We predicted the overall survival following different prevention strategies for 144 cohorts (
Data on 1 million simulations were generated. Each subgroup combination had around 6900 patients simulated across the 9 different intervention policies: (1) surveillance; (2) contralateral risk–reducing mastectomy; (3) RRBM; (4) contralateral risk–reducing mastectomy and RRBSO; (5) RRBM-RRBSO; (6) 5-year tamoxifen therapy; (7) contralateral risk–reducing mastectomy and 5-year tamoxifen therapy; (8) RRBSO; and (9) RRBM and 5 years’ tamoxifen therapy. As a result, around 770 patients were distributed to each subgroup × policy combination. All these intervention policies were considered with or without adjuvant chemotherapy, totaling 18 different policies.
For each patient, the first temporal node to be activated was identified, and survival computed for each patient. The overall survival of patients assigned for each subgroup × policy combination was plotted as Kaplan-Meier curves for 40-year follow-up and compared by the log-rank test. Hazard ratios for each subgroup were computed according to the proportional hazard Cox regression. However, given the simulation nature of the data, it was not possible to analyze any
A temporal Bayesian network model was constructed using R (R Foundation for Statistical Computing) statistical software packages ‘bnlearn’ [
Key decision variables used in baseline and sensitivity analyses were obtained from peer-reviewed English language literature published in PubMed and from publicly available databases (
Both
Therefore, we did not include carriers of the
Therefore, we used stage-specific and breast cancer subtype–specific mortality rates adjusted for age, race/ethnicity, and socioeconomic status reported in the population-based study by
In the study published by Bayraktar et al [
In patients with
We used the distribution of stage at diagnosis of ovarian cancer and 10-year survival rates for ovarian cancer reported by Benedet et al [
A recently published study [
For contralateral breast cancer we assumed the same breast cancer distribution as for the first primary breast cancer [
Multiple studies showed that a younger age at the onset of first breast cancer is associated with a higher contralateral breast cancer risk [
According to the previously published studies, carriers of the
We considered an 80% risk reduction of ovarian cancer in carriers of the
In our hypothetical cohort, patients underwent an RRBSO within 5 years after the first primary breast cancer diagnosis. At the moment, there is no clear evidence suggesting that hormone replacement therapy does not offset the second primary breast cancer risk induced by RRBSO in
We assumed contralateral risk-reducing bilateral mastectomy–adjusted risk reduction of primary mastectomy [
We assumed the local failure cumulative incidence in patients who underwent a mastectomy, as reported by Pierce et al [
We used data from the combined International Retrospective-Prospective Carriers of the Pathogenic Variants of the
We assumed that tamoxifen reduces the age-specific ER status–adjusted risk of second primary contralateral breast cancer by 56% in carriers of the
Our simulation model of a hypothetical cohort was based on previously published data, and therefore did not require a submission to a research ethics committee.
The predicted 40-year overall survival rate for carriers of
The impact of secondary prevention strategies on the survival of carriers of the
The most effective secondary prophylactic strategies in
Absence of adjuvant chemotherapy was the most powerful factor that was linked to a dramatic decline in survival for patients with breast cancer with
Most carriers of
In patients with breast cancer with the
Interestingly, we noted that the impact of RRBSO in patients with the
The impact on the survival of secondary risk–reducing strategies among carriers of
RRBM-RRBSO was the most effective risk-reducing strategy in patients with luminal breast cancer who received adjuvant chemotherapy. The protective role of RRBSO in patients with luminal breast cancer increased with their age at diagnosis, stage of the disease, and was impacted by the type of pathogenic variant (
The impact on survival of secondary risk–reducing strategies in carriers of the pathogenic variants of
The most effective secondary prophylactic strategies in
The most effective secondary prophylactic strategies in
The most effective secondary prophylactic strategies in
The most effective secondary prophylactic strategies in
The potential survival benefit from any risk-reducing strategy was modest in patients with TN breast cancer when compared with patients with luminal breast cancer. However, most carriers of
RRBM-RRBSO was only the most effective risk-reducing strategy in patients with TN breast cancer under 40 years in stage I with the
There was a negligible increase in survival in almost all carriers of the pathogenic variants of
However, RRBM-RRBSO or just RRBSO was the most effective risk-reducing strategy in patients with TN breast cancer aged over 40 years with the
There were difficulties in validating our results because of the lack of previously published studies with similar subgroups of patients that match all the detailed clinical and treatment variables.
Validation of our method was performed using TN subgroup cohort definitions from the largest published prospective study (POSH) [
Kaplan-Meier survival plots for simulation for 15 years, performed using TN subgroup cohort definitions from the largest published prospective POSH study.
In our simulation, the overall survival was similar to the results in the POSH study. Similarly, in our simulation model patients with TN breast cancer aged under 40 years with the
To our knowledge, this is the first study to simulate the expected overall survival and determine the most effective personalized management strategies for carriers
To date, only Schrag et al [
Our study showed that most
In our model, we assumed breast cancer subtype–specific, population-based mortality and the general breast cancer population–based impact of adjuvant chemotherapy on outcomes. According to the prospective study by Clifton et al [
However, a growing body of evidence indicates that
In the systematic review published by Davey et al [
In a study published by Wan et al [
The main limitation of our study is that it is a computer simulation and it can misrepresent reality. Nevertheless, our model could prove to be a valuable decision support tool and we plan to validate our model on the target patient population. Risk modifiers assume adjusted risk estimates, and therefore they are additive, and some evidence is thin and dated in some of the included estimates. We assumed independence of risk factors where it was not possible to model any interaction, and this is a limitation. Nonetheless, the final validation shows concurrent results, supporting our model.
At present, no personalized guidelines are available for the prophylactic management of second primary breast cancer in patients with the
Overall survival following different prevention strategies for 144 cohorts.
Key decision variables used in baseline and sensitivity analyses.
Secondary cancer prevention strategies in
breast-conserving treatment
estrogen receptor
International Federation of Gynaecology and Obstetrics
human epidermal growth factor receptor 2
pathologic complete response
progesterone receptor
risk-reducing bilateral mastectomy
risk-reducing bilateral salpingo-oophorectomy
triple negative
The authors confirm that the data supporting the findings of this study are available within the article or its Multimedia Appendices.
JM, DP, and MJC conceptualized this study. JM curated the data. PRP performed formal analysis of the study data. JG and FC were responsible for funding acquisition and project administration. JM and PRP proposed the study methodology and wrote the manuscript. MJC, EM, and GT were responsible for acquisition of resources and study supervision. PRP performed software/statistical analysis. JM, PRP, MN-M, MJC, and FC validated the study. JM, MN-M, MJC, EM, and GT performed data visualization. JM, MN-M, MJC, EM, and FC reviewed and edited the manuscript.
FC has received consultancy fees from Amgen, Astellas/Medivation, AstraZeneca, Celgene, Daiichi-Sankyo, Eisai, GE Oncology, Genentech, Gilead, GlaxoSmithKline, Iqvia, Macrogenics, Medscape, Merck-Sharp, Merus BV, Mylan, Mundipharma, Novartis, Pfizer, Pierre-Fabre, prIME Oncology, Roche, Sanofi, Samsung Bioepis, Seagen, Teva, and Touchime.