Olivia Munn and the negative BRCA test: does a normal genetic test rule out breast cancer risk?

By Ramon Martins Andrade, attorney (OAB/RJ 188.374)

English translation of an article first published in Portuguese: read the original.

On 13 March 2024, the actress Olivia Munn said on Instagram that she had been diagnosed with breast cancer the previous year. Her account, carried by the Associated Press and by CBS News, has a sequence that matters. In February 2023 she had taken a genetic test for 90 cancer-linked genes, negative for all of them, and had a normal mammogram. Even so, her doctor calculated her breast cancer risk score, which took into account, in her words, “my age, familial breast cancer history, and the fact that I had my first child after the age of 30”, and arrived at a lifetime risk of 37%. Because of that number, she was referred for an MRI, which led to an ultrasound and then to a biopsy: luminal B cancer in both breasts, operated on with a double mastectomy thirty days after the biopsy; in all, there were four surgeries in ten months. The account is hers and ends there: nothing is known, and nothing is asserted here, about the doctors, the services or the clinical choices along that path.

Two things tend to get confused in this story. The mastectomy was treatment of a cancer already present, not preventive surgery. And what changed the course was not a gene, it was a calculation. The question in the title has a short answer, no; the long one runs through the science of risk, through what the SUS (Brazil’s public health system) and health plans guarantee to those at high risk without a mutation, through three countries and through liability.

What a negative genetic test answers, and what it leaves unanswered

The test answers a narrow question: whether, in the genes analysed, there is a variant capable of greatly raising the risk. The National Cancer Institute (INCA) records that “hereditary breast cancer accounts, in turn, for 5% to 10% of all cases”. Two studies published together in the New England Journal of Medicine in 2021 gave the measure. In the US CARRIERS consortium, Hu and colleagues sequenced 28 genes in 32,247 women with breast cancer and 32,544 without the disease: pathogenic variants in the 12 established predisposition genes appeared in 5.03% of the patients and in 1.63% of the others; BRCA1 and BRCA2 multiplied the odds of the disease by 7.62 and 5.23, and PALB2 by 3.83. In the Breast Cancer Association Consortium, Dorling and colleagues analysed 34 candidate genes in more than 113,000 women and found an association with risk for only nine. A larger panel is not a better panel.

If around 95 in every 100 women with breast cancer in the CARRIERS sample had no variant in the known genes, a negative report takes one explanation off the table, and not the biggest one. What remain are age, reproductive history, alcohol, weight, radiation and dense breasts, which INCA lists as risk factors, and family history itself. The difference between one negative and another is decisive. A woman who tests negative for the variant already found in a relative rules out that specific risk. A woman who tests negative with no known mutation in the family receives a result that does not explain the cases in the family, and family history remains the main piece of data. Not by chance, the British guideline requires family history to be verified “where no mutation has been identified” before any risk-reducing mastectomy. The genetic test is one part of the risk assessment, not the whole assessment.

The risk score: what it measures and how accurate it is

Risk scores combine family history, age and reproductive data to estimate the probability of breast cancer over ten years or over a lifetime. The Brazilian recommendations indicate the Gail model (BCRAT) and IBIS, or Tyrer-Cuzick, as the preferred ones; Olivia Munn’s Instagram profile today points to a calculator based on IBIS.

How accurate these models are was measured by Mary Beth Terry and colleagues in Lancet Oncology in 2019. They followed 15,732 women from Australia, Canada and the United States for a median of 11.1 years, with 619 cases of cancer. IBIS and BOADICEA, which use family history across several generations, were well calibrated: the ratio of expected to observed cases stood at 1.03 and 1.05, including among women who were BRCA-negative or untested. BRCAPRO and BCRAT underestimated the risk. On discrimination, the C-statistic stood at 0.71 for IBIS and 0.70 for BOADICEA. The best models rank reasonably well who is at more and less risk, but do not say who will fall ill; and I found no equivalent validation in Brazilian women.

The number only takes on meaning against a threshold. The American Cancer Society has recommended, since 2007, annual MRI as a complement to mammography for women with a lifetime risk of approximately 20 to 25% or more. In Brazil, the 2023 joint recommendations of the Brazilian College of Radiology, the Brazilian Society of Mastology and Febrasgo adopt the same cut-off: a woman with a risk of 20% or more, “calculated by one of the mathematical models based on family history”, should have annual MRI and mammography starting ten years before the age at diagnosis of the youngest relative, but not before 30 (category A). The British guideline calls a lifetime risk between 17% and 30% moderate, and one that reaches 30% high. The 37% in the account is above all of these cut-offs.

What the number changes: screening and prevention

For the general population, screening is done with mammography: the US Preventive Services Task Force (USPSTF) has recommended since 30 April 2024 the test every two years from age 40 to 74 (grade B), a rule that applies to women with a family history but not to carriers of a high-risk genetic marker. In Brazil, INCA has recommended since September 2025 mammography every two years from 50 to 74, and the Ministry of Health announced on 23 September 2025 access for women aged 40 to 49, on demand and with shared decision-making. For high risk, the medical societies add MRI, the test which, in the actress’s account, opened the path to the biopsy.

The calculation also opens the door to prevention with medication. In the IBIS-I trial, with 7,154 women at increased risk and five years of tamoxifen or placebo, breast cancer appeared, after a median of 16 years, in 7.0% and 9.8% of the groups (Cuzick and colleagues, Lancet Oncology, 2015). In IBIS-II, with 3,864 postmenopausal women, anastrozole reduced cases by 49% after a median of 131 months (Lancet, 2020); the effect on mortality still depends on follow-up. The USPSTF has recommended since 2019 offering these medications to women aged 35 or over at increased risk and at low risk of adverse effects (grade B). Risk-reducing mastectomy is the most radical measure and the most restricted: the British guideline considers it appropriate “only for a small proportion of women” from high-risk families, with a multidisciplinary team and prior genetic counselling.

There is an honest caveat. INCA states that, for women at high risk, “there are as yet no clinical trials that have identified differentiated and effective screening strategies for reducing mortality in this subgroup”, and recommends “individualized clinical follow-up”. MRI finds more tumours, and earlier; that this translates into fewer deaths has not yet been proven by a clinical trial. The risk number is not a sentence: it is what separates standard follow-up from individualized follow-up.

In Brazil: what the SUS guarantees to those at high risk without a mutation

The legal basis is Lei 11.664/2008, rewritten by Lei 14.335/2022. The current text requires the SUS to ensure mammographic examinations “for all women who have reached puberty, regardless of age” (art. 2, II); § 2-A, inserted by Lei 15.284/2025, guarantees mammography “for all women from the age of 40 (forty), in accordance with Ministry of Health guidelines”; item IV requires referral to higher-complexity services when the unit cannot complete the diagnosis or the follow-up; and § 1 provides that the examinations “may be complemented or replaced by others whenever requested by the responsible physician”. It is in that paragraph, together with the comprehensiveness principle of Lei 8.080/1990 (the Organic Health Law) (art. 7, II), that the MRI of a woman with a high calculated risk finds support. Where there is suspicion, Lei 12.732/2012 (art. 2, § 3, inserted by Lei 13.896/2019) requires the diagnostic work-up tests within 30 days, a deadline detailed in the article on suspicious mammograms in the SUS (in Portuguese).

What is missing in the SUS is the intermediate protocol: I found no Ministry of Health document that says which model to use, from what number to refer or when to offer MRI and chemoprevention.

And in the health plan: the list guarantees the test, not the gene

In supplementary health care, breast MRI is in Annex I of RN 465/2021 without a utilization guideline, with coverage according to the treating physician’s indication; digital mammography lost its age criterion on 1 October 2026, when RN 679/2026 removed DUT 52, as explained in the article on mammography without an age criterion (in Portuguese). The test that made the difference in the account is therefore covered when properly indicated.

The genetic test is another story. Utilization Guideline (DUT) 110.7, in Annex II of RN 465 as updated by RN 676/2026, covers BRCA1 and BRCA2 sequencing mainly for those who have already had cancer, within age and family criteria. For a healthy woman, item 7 guarantees the test only “where there is a deleterious mutation in BRCA1 or BRCA2 in a first-, second- or third-degree relative”. A woman without cancer, with a strong family history and no known mutation in the family, falls outside the guideline; what remains for her is § 13 of art. 10 of Lei 9.656/1998 (the Health Plans Act), which allows coverage outside the ANS list of mandatory coverage (the “rol”) on the basis of scientific evidence or a recommendation from a health-technology assessment body, a path the firm examined in the article on what DUT 110 requires to be covered (in Portuguese).

The same logic applies to prevention. Note 3 of DUT 110.7 and note 1 of DUT 110.26 place the asymptomatic carrier of a pathogenic variant on the same footing as a patient with cancer, making risk-reducing mastectomy and reconstruction mandatory, a regime addressed in the article on risk-reducing mastectomy (in Portuguese). For a woman without a mutation and with a high calculated risk, there is no written equivalence: mastectomy appears in the list without a guideline, but the health-plan operator tends to argue that coverage presupposes disease or a mutation, and the discussion proceeds case by case, with a medical report, a verified pedigree and the literature. Chemoprevention is the clearest gap. DUT 64, on oral antineoplastic therapy, lists tamoxifen only for neoadjuvant, adjuvant or metastatic use in hormone receptor-positive breast carcinoma, and Lei 9.656/1998 excludes from the reference plan the “supply of medication for home treatment” (art. 10, VI), with the exceptions it itself makes. The preventive pill falls outside mandatory coverage. If cancer appears, as in the case reported, reconstruction is guaranteed by art. 10-A of Lei 9.656/1998 and, in the SUS, by Lei 9.797/1999.

Comparative law: calculated risk as a criterion for access

Three countries show different paths for the risk that comes from the family, and not from a gene.

1. United States: the recommendation that becomes coverage

On 20 August 2019, the USPSTF recommended (grade B) that primary care clinicians assess women with a personal or family history of breast, ovarian, tubal or peritoneal cancer, or an ancestry associated with BRCA1 and BRCA2, “with an appropriate brief familial risk assessment tool”, and refer those with a positive result for counselling and, if indicated, genetic testing; Tyrer-Cuzick is among the tools evaluated. For women without that history, grade D. The grade has legal effect: federal law (42 U.S.C. § 300gg-13, inserted by the 2010 reform known as the Affordable Care Act) requires plans and insurers to cover, without cost-sharing, services graded A or B by the task force. Risk assessment, genetic testing for those who meet the criteria, biennial mammography and risk-reducing medication thus come within preventive coverage at no cost; MRI for high risk follows the guidelines of the medical societies.

2. United Kingdom: risk categories and a table of management

NICE guideline CG164, published on 25 June 2013 and updated on 14 November 2023, organizes care by categories of calculated risk: below 17% lifetime, near that of the population; from 17% to 30%, moderate; 30% or more, high. The surveillance table shows that the number changes management even without a gene: for a high-risk woman with a 30% or lower probability of carrying a BRCA or TP53 mutation, it directs that annual mammography be considered from 30 to 39 and offered annually from 40 to 59, but does not offer MRI, which is reserved for known or probable carriers. On the other hand, it directs that tamoxifen for five years be offered to high-risk premenopausal women and anastrozole to postmenopausal women, and that the same be considered at moderate risk. The British path is more conservative on imaging and more generous on prevention.

3. Australia: the algorithm written into the fee schedule

Australia took calculated risk into its payment rules. Item 63464 of the Medicare Benefits Schedule, in the wording updated on 1 July 2026, pays once a year for MRI of both breasts in an asymptomatic woman under 60 at high risk, defined in five scenarios. The first is a high-risk mutation identified in the patient or in a first-degree relative; the following ones deal with combinations of family history, breast cancer before 50 and mantle radiotherapy; the last is a lifetime risk “greater than 30%” or a ten-year risk greater than 5% “using a relevant clinical risk assessment algorithm”. The gene is one door; the calculation is another.

In Brazil, the 20% threshold exists only in the recommendations of medical societies, and the strongest coverage, from genetic testing of the healthy woman to preventive surgery, depends on an identified mutation. The woman with a negative test and a high risk is the one the Brazilian design leaves without a rule of her own.

When omission becomes a question of liability

The duties are written down. The Code of Medical Ethics forbids a physician to “fail to use all available means of health promotion and of prevention, diagnosis and treatment of disease, scientifically recognized and within their reach, in favour of the patient” (art. 32) and to fail to inform the patient of diagnosis, prognosis, risks and the aims of treatment (art. 34). The Estatuto dos Direitos do Paciente, Lei 15.378/2026 (the Patients’ Rights Statute), guarantees information that is “accessible, up to date and sufficient” for making a decision (art. 12, § 1) and access to the medical record without having to give reasons (art. 19). The hospital’s liability is strict as regards its own services (Código de Defesa do Consumidor, the Consumer Protection Code, art. 14); the physician’s depends on fault (art. 14, § 4; Civil Code, art. 951).

The line between failure and the limits of medicine runs through three questions. The first is whether the family history was asked about and recorded. Not running a calculator for a woman with no family history is not fault, and the USPSTF advises against routine assessment in those cases; not asking, or not writing it down, is a failure in history-taking. The second is what was done with a negative test in the face of a relevant family history: treating it as a definitive discharge, without explaining what it does not rule out and without a follow-up plan, runs counter to what Brazilian and foreign recommendations describe. The third arises when, after “all normal”, a sign appears. INCA treats a hard lump, bloody discharge and retraction of the skin or nipple as signs for urgent referral and warns that false negatives in screening “result in false reassurance for the woman”; a negative test and a normal mammogram do not authorize dismissing them. If the tumour was already visible on an earlier examination, the discussion is that of the limits of the imaging method (in Portuguese).

Even with a failure, there must be harm. The STJ (Superior Court of Justice) applied loss of a chance to physicians who did not follow Ministry of Health guidance, “taking from the patient a concrete and real chance of having a correct diagnosis” (REsp 1.985.977/DF, First Panel, decided on 18 June 2024), but rejected it when the expert was categorical in concluding that “it is not possible to state that the patient would have had a different course had the diagnosis been made earlier” (AgInt no AREsp 2.891.296/SP, Fourth Panel, decided on 1 June 2026). The leading precedent, REsp 1.254.141/PR (Third Panel, Justice Nancy Andrighi reporting, 4 December 2012), a case of inadequate treatment of breast cancer, set the limit: “The chance, however, can never reach the value of the asset lost”. In breast cancer, the chance is diagnosis at an earlier stage, and the proof is expert evidence. The general guideposts are in the guide on medical error (in Portuguese).

Before deciding

In my view, the lesson of the account is not that every woman should have a genetic panel, but that a negative panel is the beginning of a conversation. Fault, where it exists, will lie in the omission of history-taking and referral, not in the absence of a specific tool. For a woman with a family history, the practical path has three steps: take the family history to the consultation in writing, with the degree of kinship, the side of the family and the age at each diagnosis; ask for the risk to be calculated and for the number and the follow-up plan to be entered in the medical record; and, if the risk exceeds 20%, ask about the annual MRI that the Brazilian societies recommend, that the health plan covers on medical indication and that the SUS must offer. If the health plan denied the MRI, if the referral stalled in the queue or if the diagnosis came after a negative test without anyone having recorded the family history, get in touch with the firm: the analysis starts from the genetic report, the earlier images and what the medical record says about the family, and does not replace medical assessment.

About the author

Ramon Martins Andrade (OAB/RJ 188.374) is a lawyer who graduated from UFRJ in 2011, with master’s degrees from Université Sorbonne Nouvelle and from the Naval War College (EGN, Brazilian Navy). He is currently a researcher in Human Rights and Health at ENSP/Fiocruz.

This content is for information and education only. It does not constitute legal advice, advertising of results or any guarantee of outcome, under Rule 205/2021 of the Brazilian Bar Association (OAB).