If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Year 1: Experiences of a tertiary cancer centre following implementation of reflex BRCA1 and BRCA2 tumor testing for all high-grade serous ovarian cancers in a universal healthcare system
Familial Cancer Clinic, University Health Network, 610 University Avenue, Toronto, ON M5G 2M9, CanadaLawrence S Bloomberg Faculty of Nursing, University of Toronto, 155 College Street, Toronto, ON M5T 1P8, Canada
Department of Molecular Genetics, University of Toronto, 27 King's College Circle, Toronto5, ON M5S 1A8, CanadaDivision of Clinical Laboratory Medicine, University Health Network, 200 Elizabeth Street, Toronto, ON M5G 2C4, Canada
Division of Gynecologic Oncology, University Health Network, 610 University Avenue, Toronto, ON M5G 2M9, CanadaDepartment of Obstetrics and Gynecology, University of Toronto, 27 King's College Circle, Toronto, ON M5S 1A8, Canada
Familial Cancer Clinic, University Health Network, 610 University Avenue, Toronto, ON M5G 2M9, CanadaDivision of Medical Oncology, University Health Network, 610 University Avenue, Toronto, ON M5G 2M9, CanadaDepartment of Medicine, University of Toronto, 27 King's College Circle, Toronto, ON M5S 1A8, Canada
Division of Clinical Laboratory Medicine, University Health Network, 200 Elizabeth Street, Toronto, ON M5G 2C4, CanadaDepartment of Laboratory Medicine and Pathobiology, University of Toronto, 27 King's College Circle, Toronto, ON M5S 1A8, Canada
Lawrence S Bloomberg Faculty of Nursing, University of Toronto, 155 College Street, Toronto, ON M5T 1P8, CanadaWomen's College Research Institute, 72 Grenville Street, Toronto, ON M5S 1B2, Canada
Improvements in the rate and time to genetics referral were seen after implementation of reflex BRCA1/2 tumor testing.
•
Clinicians referred HGSOC patients for germline testing irrespective of BRCA1/2 tumor results.
•
Reflex BRCA1/2 tumor test results were not available for all HGSOC patients, likely due to insufficient tumor tissue.
•
There is a high uptake of genetic counseling and germline testing among HGSOC patients, irrespective of tumor testing.
•
Reflex BRCA1/2 tumor testing can be implemented without compromising hereditary cancer risk assessment.
Abstract
Objective
This study compares the rate and time to genetic referral, and patient uptake of germline genetic services, before and after implementation of reflex BRCA1/2 tumor testing for high-grade serous ovarian cancer (HGSOC) in a universal healthcare system.
Methods
A retrospective chart review of HSGOC patients diagnosed in the year before (PRE) and after (POST) implementation of reflex BRCA1/2 tumor testing was conducted. Clinical information (date/age at diagnosis, personal/family history of breast/ovarian cancer, cancer stage, primary treatment, tumor results) and dates of genetics referral, counseling, and germline testing were obtained. Incident rate ratios (IRR) and 95% CI were calculated using negative binomial regression. Time to referral was evaluated using Kaplan-Meier survival analysis. Fisher Exact tests were used to evaluate uptake of genetic services.
Results
175 HGSOC patients were identified (81 PRE; 94 POST). Post-implementation of tumor testing, there was a higher rate of genetics referral (12.88 versus 7.10/1000 person-days; IRR = 1.60, 95% CI: 1.07–2.42) and a shorter median time from diagnosis to referral (59 days PRE, 33 days POST; p = .04). In the POST cohort, most patients were referred prior to receiving their tumor results (n = 63/77; 81.8%). Once referred, most patients attended genetic counseling (94.5% PRE, 97.6% POST; p = .418) and pursue germline testing (98.6% PRE; 100% POST; p = .455).
Conclusions
Following implementation of reflex BRCA1/2 tumor testing for HGSOC, significant improvements to the rate and time to genetics referral were identified. Additional studies are needed to evaluate physician referral practices and the long-term impact of reflex tumor testing.
]. Up to 20–30% of HGSOC, including primary peritoneal and fallopian tube cancers, are attributed to an inherited pathogenic/likely pathogenic variant (PV) in a cancer predisposition gene, with the majority occurring in the BRCA1 and BRCA2 genes [
BRCA mutation frequency and patterns of treatment response in BRCA mutation-positive women with ovarian cancer: a report from the Australian Ovarian Cancer Study Group.
Cancer incidence and survival in Lynch syndrome patients receiving colonoscopic and gynaecological surveillance: first report from the prospective Lynch syndrome database.
]. Women with hereditary forms of HGSOC have a 50% chance of passing a hereditary risk to each offspring, and genetic testing is recommended for all cases of HGSOC, irrespective of age at diagnosis or family history [
Recently, the utility of genetic testing in HGSOC patients has expanded beyond identification of hereditary families, and now is also able to identify patients who may benefit from targeted therapies. Compared to standard treatments, poly(ADP-ribose) polymerase inhibitors (PARPi) have superior treatment outcomes in HGSOC, particularly in individuals with a BRCA1/2-associated cancer [
Olaparib maintenance therapy in patients with platinum-sensitive relapsed serous ovarian cancer: a preplanned retrospective analysis of outcomes by BRCA status in a randomised phase 2 trial.
]. After initial chemotherapy, maintenance therapy with PARPi provides an estimated 70% reduction in disease progression or death in BRCA1/2-associated HGSOC patients [
]. As a result, genetic testing is now required shortly after diagnosis to inform cancer treatment.
Traditionally, genetic testing is facilitated by a genetic counselor over two appointments (pre-test and post-test). Following detailed genetic counseling, testing for inherited (germline) PVs is completed using germline DNA, typically extracted from a blood sample; however, an estimated 5–7% of HGSOC may have acquired (somatic) BRCA1/2 PVs, which are only present in the tumor [
]. These acquired PVs are not detected in the germline and can only be identified through genetic testing of DNA extracted from tumor tissue. Since PARPi can be used to treat HGSOC patients with a germline or somatic BRCA1/2 PV, several international groups have published recommendations on how to best integrate tumor testing into patient care pathways [
]. While national guidelines for genetic testing do not exist in Canada, a national collaborative has proposed that reflex BRCA1/2 tumor testing of all newly diagnosed HSGOC, where genetic testing of tumor tissue is initiated by a pathologist as part of surgical pathology review, could facilitate treatment and improve the efficiency of genetic testing processes [
National BRCA Collaborative Why is tumor testing in ovarian cancer needed in Canada? An opinion statement developed by the National BRCA Collaborative the Society of Gynecologic Oncology of Canada.
]. In Canada, the province of Ontario began funding reflex BRCA1/2 tumor testing for all newly diagnosed cases of HGSOC in August 2018; however, germline genetic testing continues to be recommended and funded for all cases of HGSOC, irrespective of tumor results (Supplement 1).
Though reflex BRCA1/2 tumor testing may streamline genetic testing for PARPi treatment purposes, there are limitations to the focused nature of a BRCA1/2 tumor test. While tumor testing will detect both germline and somatic PVs, additional germline testing is required to determine whether any identified PV confers a hereditary cancer risk. Some groups have suggested that tumor results could be used to triage genetics referrals, whereby germline testing would only be offered to the subset of HGSOC patients with a PV identified via tumor testing [
]. This may be particularly useful in countries where patients are required to pay for genetic testing, or where insurance coverage for genetic testing may be limited; however, in Ontario, both tumor and germline genetic testing for HGSOC patients is funded through a universal healthcare system. Furthermore, standard germline testing in Ontario interrogates at least 8 other ovarian cancer risk genes (MLH1, MSH2, MSH6, PMS2, EPCAM, BRIP1, RAD51C, RAD51D), which account for up to 20% of hereditary ovarian cancer cases [
]. PVs in these additional ovarian cancer risk genes would be missed through a BRCA1/2 tumor test; therefore, until a provincially funded tumor test includes analysis of all genes relevant to hereditary ovarian cancer risk, continued germline testing of all HGSOC patients in Ontario remains critical to identify hereditary families and facilitate the prevention of future cancers.
There is limited data regarding large scale implementation of tumor testing for HGSOC and it is unclear if Ontario's universal implementation of this model will impact rates of genetic counseling referral and germline genetic testing. Herein, we report on the experience of reflex BRCA1/2 tumor testing at a large Ontario-based cancer center by comparing the rates and time to genetic counseling referral among newly diagnosed HGSOC patients, as well as patient attendance for genetic counseling and uptake of multi-gene germline testing, in the year before and after implementation of this model of care.
2. Material and methods
2.1 Study design & population
A retrospective review of consecutive cases of HGSOC diagnosed in the year before (PRE) and after (POST) implementation of reflex BRCA1/2 tumor testing was conducted at the University Health Network (UHN)’s Princess Margaret Cancer Centre in Toronto, Canada. Cases of HGSOC, including primary peritoneal and fallopian tube cancers, treated at UHN and either diagnosed between August 1, 2017 and July 31, 2018 (PRE cohort) or October 1, 2018 and September 30, 2019 (POST cohort) were included in the study. Since reflex BRCA1/2 tumor testing was implemented at UHN in August 2018, cases diagnosed in August or September 2018 were excluded to allow for initial changes in patient care algorithms. Cases with previous germline testing for hereditary cancer were also excluded. A list of eligible HGSOC cases was generated from the Princess Margaret Cancer Center Cancer Registry and tumor testing records from the UHN Genome Diagnostics laboratory. Genetic testing information recorded in the UHN Electronic Patient Record or the Familial Cancer Clinic database was used to exclude cases where previous germline testing was completed. Relevant clinical information (age at diagnosis, personal/family history of breast/ovarian cancer, cancer stage, primary treatment) and tumor test results were extracted from the UHN Electronic Patient Record. Genetic counseling and germline testing data (dates of referral, genetic counseling, and blood draw for genetic testing) were obtained from the Familial Cancer Clinic database. The study received approval from Research Ethics Boards at UHN and the University of Toronto.
2.2 Statistical analysis
Descriptive statistics were used to summarize demographic information; significant differences between PRE and POST cohorts were evaluated using Mann Whitney U tests for continuous variables and Pearson's Chi Square or Fisher Exact tests for categorical variables, as appropriate. Given the retrospective nature of this study, person-time was used to account for the longer period of time available for referral in the PRE cohort; person-days accrued from the date of diagnosis to the first of: date of referral, date of death, or date of retrospective review. Referral rates were reported as the number of referrals/1000 person-days; 95% confidence intervals (CI) were reported using the normal approximation method [
]. Incident rate ratios (IRR) and 95% CI were calculated using negative binomial regression; analyses were offset by person-days and stratified by relevant clinical variables. Kaplan-Meier survival analysis was used to report the time to referral. Analyses were stratified by relevant clinical variables and statistical significance was determined using the Log-Rank test. Pearson's Chi Square or Fisher Exact tests were used to compare patient attendance for genetic counseling and uptake of germline testing, as appropriate. Statistical analyses were completed using IBM SPSS Statistics for Windows, version 24 (Armonk NY, USA: IBM Corp) and statistical significance was reported using a two-tailed α = 0.05.
3. Results
3.1 Study population
A total of 212 cases of HGSOC, including cases of primary peritoneal or fallopian tube cancers, were identified. In all, 34 cases were excluded due to a diagnosis in August/September 2018 (n = 16) or previous germline genetic testing (n = 18); an additional 3 cases were excluded as they were seen by oncologists at UHN for an opinion only, leaving a total of 175 cases. Of these 175, 81 (46.3%) were diagnosed before implementation of reflex BRCA1/2 tumor testing and were included in the PRE cohort; 94 (53.7%) were diagnosed after implementation of reflex BRCA1/2 tumor testing and were included in the POST cohort (Table 1).
Mann Whitney U test was completed to compare age at diagnosis. Pearson Chi-Square was completed to compare primary treatment, stage at diagnosis, and family history of breast/ovarian cancer. Fisher's Exact Test was used to compare personal history of breast cancer.
a Stage and primary surgery was unavailable for one individual.
b Family history was unknown for three individuals.
Overall, the median age of diagnosis was 63.8 years (range 38.1–90.0 years). The majority (89.1%) were diagnosed with stage 3 or 4 disease; and about half (48.9%) were treated with primary surgery. Only 6 cases (3.4%) were known to have a prior diagnosis of breast cancer. Family history information was available for 172 cases, of which 57 (33.1%) had a family history of breast or ovarian cancer. In the POST cohort, tumor results were available in the UHN electronic medical record of 80 cases (85.1%). Of the remaining 14 cases, 11 had biopsy results only, 1 had a cytology sample only was lost to follow-up at UHN, and 1 had surgery at an outside centre, with tumor testing initiated at UHN and then cancelled, possibly because a germline BRCA variant was identified; it was unclear why tumor testing was not completed in the remaining case. Of cases with tumor results, 17/80 (21.3%) had a BRCA1/2 PV identified in their tumor tissue. Clinical information did not differ significantly between PRE and POST cohorts (Table 1).
3.2 Rate of referral for germline genetic services
During a period of 16,950 person-days, 159 cases of HGSOC were referred for genetic counseling, for an overall referral rate of 9.38/1000 person-days (95% CI: 7.92–19.84) (Table 2). A total of 73 cases were referred in the PRE cohort, rate = 7.10/1000 person-days (95%CI: 5.48–8.73) and 86 cases were referred in the POST cohort, rate = 12.88/1000 person-days (95% CI: 10.16–15.61); the overall rate of referral was 60% higher in the POST cohort (IRR = 1.60; 95%CI: 1.06–2.40) (Table 2). When considering cases with young age at diagnosis, relevant personal/family history, or primary surgery, rates of referral did not differ significantly between PRE and POST cohorts (Table 2).
Table 2Rates (per 1000 person-days) of referral for genetic counseling among high-grade serous ovarian cancers diagnosed pre- and post-implementation of reflex BRCA1/2 tumor testing.
A single case was referred on the date of diagnosis; as per convention, 0.5 person-days were added to each case to allow inclusion of cases in the negative binomial regression [40].
A single case was referred on the date of diagnosis; as per convention, 0.5 person-days were added to each case to allow inclusion of cases in the negative binomial regression [40].
1.0 (Ref)
1.40 (0.80–2.45)
Cases with a personal history of breast cancer and/or family history of breast/ovarian cancer
A single case was referred on the date of diagnosis; as per convention, 0.5 person-days were added to each case to allow inclusion of cases in the negative binomial regression [40].
A single case was referred on the date of diagnosis; as per convention, 0.5 person-days were added to each case to allow inclusion of cases in the negative binomial regression [40].
Ref (1.0)
1.45 (0.96–2.18)
a A single case was referred on the date of diagnosis; as per convention, 0.5 person-days were added to each case to allow inclusion of cases in the negative binomial regression [
3.3 Timing of referral for germline genetic services
A significantly shorter time from diagnosis to genetics referral was identified following implementation of reflex BRCA1/2 tumor testing, with median times of 59 and 33 days in the PRE and POST cohorts, respectively (p = .04) (Table 3, Fig. 1). Among the subset of cases with young age at diagnosis, relevant personal/family history, or primary surgery, differences in the time to referral were not significant (Table 3).
Table 3Time from diagnosis to referral for genetic counseling among high-grade serous ovarian cancers diagnosed pre- and post-implementation of reflex BRCA1/2 tumor testing.
Pre-implementation
Post-implementation
Total number of HGSOC cases
81
94
Mean days to referral (95% CI)
141.91 (94.19–189.63)
78.04 (57.10–98.99)
Median days to referral (95% CI)
59.00 (27.87–90.13)
33.00 (29.05–36.96)
Log rank
p = .04
Cases diagnosed < 60 years
31
33
Mean days to referral (95% CI)
75.129 (47.72–102.54)
55.77 (34.78–76.77)
Median days to referral (95% CI)
51.000 (27.00–75.00)
25.00 (16.16–33.84)
Log rank
p = .232
Cases with a personal history of breast cancer and/or family history of breast/ovarian cancer
Fig. 1Kaplan Meier curve of time to genetics referral for high-grade serous ovarian cancers diagnosed pre- and post-implementation of reflex BRCA1/2 tumor testing.
Of the 80 HGSOC cases in the POST cohort with tumor results available, 78 (97.5%) were referred for germline genetic testing, with 64/78 (82.0%) cases referred prior to the availability of their tumor results. Of those with positive BRCA1/2 tumor testing results, 16/17 (94.1%) were referred for germline testing, with 15/16 (93.8%) referred prior to availability of tumor results.
3.4 Patient uptake of germline genetic services
Overall utilization of clinical genetic services is presented in Fig. 2. Of 175 HGSOC included in this review, 159 (90.9%) were referred for genetic counseling and germline genetic testing [73/81 PRE (90.1%), 86/94 POST (91.5%)]. Of these, 154 (96.9%) attended genetic counseling [70/73 (95.9%) PRE, 84/86 (97.7%) POST; p = .423]. In the PRE cohort, one patient died prior to their scheduled appointment and two actively cancelled their appointment, one of whom stated they had been referred to another genetics clinic. In the POST cohort, one patient died prior to their scheduled appointment and one no-showed with no explanation available. Among those who attended genetic counseling, 69/70 (98.6%) in the PRE cohort provided a blood sample, compared to 84/84 (100%) in the POST cohort (p = .455). In the POST cohort, all 78 cases with tumor results available who were referred for genetic counseling completed germline testing, resulting in completion of germline and tumor testing for 78/94 cases (83.0%).
Fig. 2Utilization of germline genetic services among newly diagnosed cases of high-grade serous ovarian cancer.
To our knowledge, this is the first Canadian study to report on the experience of reflex BRCA1/2 tumor testing for HGSOC, and is the first study to compare rates and timing of pre-test genetic counseling referral before and after implementation of this model of care. In the year following implementation of reflex BRCA1/2 tumor testing, there was a significant increase in the rate of referral and decrease in the time to referral for genetic counseling and germline testing. Post-implementation, most patients were referred for germline testing prior to the availability of tumor results. Once referred, the vast majority of HGSOC patients in both groups attended genetic counseling, and all but one of those who attended all completed germline genetic testing.
When integrating reflex BRCA1/2 tumor testing into patient care pathways, it is critical to recognize the differing primary objectives of tumor and germline testing. Tumor testing primarily aims to identify genetic PV as therapeutic targets, whereas germline testing seeks to identify individuals and families at increased cancer risk. In Ontario, the primary goal of reflex tumor testing is to identify all HGSOC patients who may benefit from treatment with a PARPi; therefore, provincial funding is limited to analysis of BRCA1/2 alone. At the onset of reflex BRCA1/2 tumor testing, it was unclear if there would be a decline in genetics referrals and germline testing, particularly among individuals with negative tumor results. In addition to the requirement of germline testing to discern whether a BRCA1/2 PV identified through tumor testing is germline or somatic in origin, continued germline testing of all cases of HGSOC remains essential as previous research has shown that inherited PVs in at least 7 genes confer a ≥ 5-fold risk of ovarian cancer [
]. Several groups have suggested that tumor results can be used to limit the number of genetics referrals for HGSOC patients by referring only those with an identified PV on tumor tissue or a suggestive clinical/family history [
]; however, a study of 7768 ovarian cancer cases found negative family history did not significantly alter ovarian cancer risk ratios, and the frequency of non-BRCA1/2 PVs increased among patients with an older age at diagnosis [
]. PVs in these non-BRCA1/2 ovarian cancer risk genes will be missed via Ontario's current targeted tumor testing, but are routinely analyzed during germline multi-gene panel testing.
In the first year following implementation of BRCA1/2 tumor testing at our center, 83% of newly diagnosed HGSOC patients had both tumor and germline testing completed and 91% of patients were referred for genetic counseling and germline testing. Among the 94 HGSOC patients in the POST cohort of this study, 164 genetic tests (80 tumor and 84 germline) were completed for 86 patients, resulting in an average of 1.9 genetic tests/patient. Though our data demonstrates that reflex BRCA1/2 tumor testing can be used for treatment purposes, while continuing to provide comprehensive assessments of hereditary cancer risk, it also highlights the excess testing required when tumor testing is limited to therapeutic targets. Such high numbers of duplicate testing may not be acceptable to patients in countries without universal healthcare. Tumor testing has been used by others to preserve scarce genetic counseling and testing resources by identifying the subset of HGSOC patients who warrant germline testing. For example, in a feasibility study of universal BRCA1/2 tumor testing for all newly diagnosed ovarian cancer patients in the Netherlands, Vos et al. reported a 38.7% reduction in the number of required genetic tests, compared to germline-first models [
]. Yet, to provide a comprehensive assessment of hereditary cancer risk, such patient care algorithms should only be implemented when tumor testing includes analysis of all genes relevant to hereditary ovarian cancer risk.
Overall, 91% (159/175) of HGSOC patients in our cohort were referred for genetic counseling and germline testing. When compared to the year prior (PRE cohort), the rate of referral was 60% higher (7.10 versus 12.88/1000 person-days) and the median time from diagnosis to referral was 26 days shorter (59 days versus 33 days) following initiation of reflex BRCA1/2 tumor testing (POST cohort). Post implementation of reflex tumor testing, 85% (80/94) of HGSOC patients had tumor results available; however, of those referred, most (82%; 64/78) were referred to genetics prior to the availability of their tumor results. Among the 17 HGSOC patients with a BRCA1/2 PV identified via tumor testing, 15 (88.2%) were referred prior to the availability of these results. Taken together, these findings suggest that oncologists are not using tumor results to determine which HGSOC patients should be referred for genetic counseling and germline testing. This may be a result of established patient care pathways, as previous data from our institution has demonstrated that HGSOC patients prefer to be referred for genetic counseling early in their cancer treatment [
]. Additionally, since provincial recommendations state that all HGSOC patients should be referred for genetic counseling and germline testing, irrespective of tumor results (Supplement 1), there is no incentive for providers to wait until tumor results are available to initiate genetics referrals. Further, data demonstrating the benefits of PARPi as a first-line maintenance therapy to treat BRCA1/2-mutated ovarian cancer was first published in October 2018 [
], resulting in a small window of time to clarify a patient's BRCA status. This added urgency of genetic testing for treatment purposes may have contributed to the decreased time to referral seen in this study, particularly as reflex tumor testing is usually completed on surgical specimens and over half (49/93; 52.7%) of individuals in the POST group received neoadjuvant chemotherapy, delaying their tumor testing.
While most of the HGSOC patients in this study were referred for genetic counseling and germline testing, 15% (14/94) of HGSOC patients in the POST cohort did not have tumor testing completed at the time of our review, which was conducted a minimum of 3 months after their date of diagnosis. In an assessment of universal BRCA1/2 tumor testing in the Netherlands, 22% (91/406) of ovarian cancer patients did not have tumor testing completed, only 20% (18/91) of whom had germline testing [
]. By comparison, 57% (8/14) of HGSOC patients in this study, who did not have tumor results, were referred for germline testing. Of these 8 patients, two did not attend their scheduled genetic counseling appointment, resulting in 8.5% (8/94) of patients who did not have tumor or germline testing completed following implementation of reflex tumor testing. The reason tumor testing was not completed could not be determined through this retrospective chart review; however, this finding cautions against reliance on tumor results alone to initiate genetics referral, as tumor results may not be available for some patients. In our cohort, 13/14 (93%) patients had only biopsy or cytology specimens available; in these instances, tumor testing may not have been initiated due to inadequate sample size or poor tumor cellularity. This is particularly important for HGSOC patients as 51% (89/174) patients in the total study cohort were treated with neoadjuvant chemotherapy. Emerging tests validated on cytology samples may provide an alternative to tumor testing, as these samples are easier to collect and 100% concordance of BRCA1/2 results has been reported between tumor and cytology samples [
Next-generation sequencing-based BRCA testing on cytological specimens from ovarian cancer ascites reveals high concordance with tumor tissue analysis.
]. This would provide opportunities to initiate tumor testing earlier in the patient journey, particularly for those treated with neoadjuvant chemotherapy.
When tumor testing is completed, follow-up genetic counseling and germline testing of patients with identified PVs is critical to clarify whether family members may be at risk. This is particularly important for HGSOC, as 57–85% of BRCA1/2 PVs identified in tumor tissues are germline in origin [
]. In this study, 94% (16/17) of HGSOC patients with a BRCA1/2 PV in their tumor were referred for genetic counseling, all of whom provided a blood sample for germline testing. This is similar data from the Netherlands, where 88% of ovarian cancer patients were referred for and 98% of patient completed germline testing following identification of BRCA1/2 PVs via tumor testing [
]. Though most patients were referred, the potential patient, clinician, and system barriers which may contribute to non-referral of HGSOC patients for germline testing following abnormal tumor results may require further investigation. Multiple mechanisms, such as quality assurance measures or genetic navigators, may be required to ensure timely genetics referral of all individuals with PVs identified via tumor testing.
With respect to patient uptake of germline services, the vast majority of HGSOC patients who were referred for genetic counseling attended their scheduled appointment (95% PRE and 98% POST), and all but one consented to germline testing. This is consistent with published literature of ovarian cancer patients [
]; however, it is unclear whether patients were aware of their tumor results prior to attending genetic counseling. Further studies are required to evaluate the patient impact of reflex BRCA1/2 tumor testing and whether knowledge of tumor results will alter patient uptake of germline services.
The results of this study should be considered in the context of several limitations. First, the study was retrospective in nature and relied on medical records and documentation of referral. For those cases not referred for germline testing, individuals may have declined the option of referral or received genetic counseling and germline testing at another institution. Furthermore, the reasons why reflex tumor testing was completed for all cases of HGSOC in the POST cohort were not investigated; however, this is likely due to lack of a surgical specimen, as reflex BRCA1/2 tumor testing is typically completed following ovarian cancer surgery, rather than on biopsy or cytology specimens. Second, this study was conducted at a large, academic cancer center in Ontario, Canada, where the clinical genetics and gynecologic oncology departments are highly integrated. Study results may not be representative of other centers, such as community hospitals and institutions without on-site clinical genetic services. Third, the practice and views of referring clinicians were not evaluated and further research regarding physician referral practices are required. Additionally, the long-term effects of reflex tumor testing were not assessed, and further studies are needed to determine whether referral rates will change over time. Finally, we did not evaluate patient perspectives of reflex BRCA1/2 tumor testing and additional research of the patient experiences are required to guide widespread implementation of this model of care.
5. Conclusion
This study reports on the experience of implementing reflex BRCA1/2 tumor testing for HGSOC at a large cancer center in Ontario, Canada, where genetic services are available through a universal healthcare system. Significant improvements to the rate and time to genetics referral were identified following implementation of this tumor testing model, which are likely a reflection of patient and provider preferences, rather than a result of tumor testing itself. Tumor results were not available for a subset of HGSOC patients after reflex tumor testing protocols began, often because surgical specimens were not available. Expansion of reflex tumor testing to include non-surgical specimens, such as cytology, and interrogate all genes associated with hereditary ovarian cancer risk, may provide opportunities to minimize duplicate genetic testing and streamline genetic services for HGSOC patients. Such improvements may particularly important in institutions with limited resources or where genetic testing is not covered by a universal healthcare system.
The authors have no relevant conflicts of interest to declare.
Acknowledgements
The authors would like to acknowledge funding received from the Canadian Institutes of Health Research (CIHR) Women's Health Clinical Mentorship Grant to support this research.
BRCA mutation frequency and patterns of treatment response in BRCA mutation-positive women with ovarian cancer: a report from the Australian Ovarian Cancer Study Group.
Cancer incidence and survival in Lynch syndrome patients receiving colonoscopic and gynaecological surveillance: first report from the prospective Lynch syndrome database.
Olaparib maintenance therapy in patients with platinum-sensitive relapsed serous ovarian cancer: a preplanned retrospective analysis of outcomes by BRCA status in a randomised phase 2 trial.
Why is tumor testing in ovarian cancer needed in Canada? An opinion statement developed by the National BRCA Collaborative the Society of Gynecologic Oncology of Canada.
Next-generation sequencing-based BRCA testing on cytological specimens from ovarian cancer ascites reveals high concordance with tumor tissue analysis.
32333-7&id=gr1.jpg){kind=link}
32333-7&id=gr2.jpg){kind=link}