Farxiga Cancer Risk Signal Review: What Women Taking Dapagliflozin Need to Know

Why the Cancer Question Matters More for Women

The cancer risk question with dapagliflozin is not a footnote. It is the reason the FDA delayed approval of Farxiga until January 2014, two years after the European Medicines Agency granted conditional marketing authorization. Women make up the majority of patients with autoimmune thyroid disease, lupus nephritis, and certain CKD subtypes that dapagliflozin now treats, and women with type 2 diabetes already carry a higher relative risk of cardiovascular mortality than men. Understanding where the cancer signal came from, what subsequent data shows, and what it means specifically for your body requires more than a one-line reassurance.

Where the Signal Started

When AstraZeneca submitted the dapagliflozin NDA to the FDA in 2011, the agency's advisory committee identified a numeric imbalance: 9 bladder cancer cases in dapagliflozin-treated patients versus 1 in placebo-treated patients across the initial pooled trial dataset. The committee also noted 9 breast cancer cases on dapagliflozin versus 1 on placebo. The FDA voted 9-6 against initial approval and requested additional data. A complete response letter followed, and approval was granted in 2014 with a label warning and a post-marketing requirement to conduct a dedicated bladder cancer study.

Why Biologic Plausibility Was Considered

Elevated urinary glucose caused by SGLT2 inhibition theoretically exposes bladder epithelium to higher glucose concentrations, and glucose has been proposed as a promoter of urothelial proliferation. This mechanism was plausible enough to take seriously. It was not strong enough, on its own, to confirm causation.

What the Large Trials Found

The DECLARE-TIMI 58 trial is the most important single dataset here. Published in the New England Journal of Medicine in 2019, DECLARE enrolled 17,160 patients with type 2 diabetes and followed them for a median of 4.2 years. Bladder cancer incidence was 0.3% in the dapagliflozin group versus 0.5% in the placebo group, numerically favoring the drug. Breast cancer rates were balanced. The trial authors concluded there was no evidence of increased cancer risk with dapagliflozin.

The DAPA-HF trial, published in the New England Journal of Medicine in 2019, studied 4,744 patients with HFrEF and reported a 26% reduction in the composite of worsening heart failure or cardiovascular death (hazard ratio 0.74, 95% CI 0.65-0.85). Cancer was not a primary endpoint, but adverse event data showed no excess malignancy signal in the dapagliflozin arm across 18.2 months of median follow-up.

The DAPA-CKD Trial Adds Context

DAPA-CKD, published in the New England Journal of Medicine in 2020, enrolled 4,304 patients with CKD (eGFR 25-75 mL/min/1.73 m²) and urine albumin-to-creatinine ratio of 200-5000 mg/g. The trial was stopped early for benefit. Serious adverse events including malignancies were similar between groups, and the investigators reported no cancer safety concern.

Pooled and Meta-Analytic Evidence

A 2022 meta-analysis published in Diabetes, Obesity and Metabolism pooled 47 randomized controlled trials of SGLT2 inhibitors and found no statistically significant increase in any cancer type, including bladder, breast, or endometrial cancer, compared with placebo or active comparators. The authors noted that follow-up durations (generally under 5 years) may be insufficient to detect slow-growing cancers, a limitation the analysis explicitly acknowledged.

The Breast Cancer Question for Women

Breast cancer is the most common cancer in women worldwide. Any drug taken long-term by women with diabetes or heart failure must be scrutinized through this lens.

What Early Data Showed

The initial NDA imbalance (9 versus 1) was striking. The FDA's advisory committee in 2012 considered whether this could represent a true signal or a chance finding in a relatively small dataset. The FDA briefing document noted that the imbalance was driven almost entirely by hormone-receptor-positive breast cancers diagnosed within the first year of treatment, a timing inconsistent with drug-induced carcinogenesis, which typically requires years of exposure. The committee concluded the cases likely represented pre-existing cancers detected incidentally rather than drug-induced disease.

What Subsequent Data Shows

DECLARE-TIMI 58 enrolled 6,422 women, the largest female subgroup in any dapagliflozin cardiovascular outcomes trial. Breast cancer incidence was not statistically different between arms. A 2020 observational study in Diabetes Care using a population-based cohort of 60,523 SGLT2 inhibitor initiators found no association between SGLT2 inhibitor use and breast cancer (adjusted hazard ratio 0.93, 95% CI 0.81-1.07).

What This Means for You

If you have a personal history of estrogen-receptor-positive breast cancer, a BRCA1 or BRCA2 mutation, or are currently on aromatase inhibitor therapy, the available data does not show that dapagliflozin worsens your cancer risk. Your oncologist and prescribing clinician should review this together. The absence of confirmed risk is not a green light to skip that conversation.

Bladder Cancer: What Women Should Know

Bladder cancer is three to four times more common in men than in women. Women are, however, diagnosed at later stages and have worse stage-for-stage survival, which means the question still matters.

The following framework summarizes how to interpret the bladder cancer signal across the available evidence levels:

| Evidence Level | Finding | Clinical Weight | |---|---|---| | Initial NDA pool (2011) | 9 vs. 1 cases; HR not calculable | Signal-generating only; very small N | | DECLARE-TIMI 58 (2019) | 0.3% vs. 0.5%; favors drug | Moderate reassurance; median 4.2 years | | DAPA-CKD (2020) | No significant difference | Moderate reassurance; median 2.4 years | | Meta-analysis, 47 RCTs (2022) | No statistically significant increase | Strongest reassurance; limited by follow-up | | FDA label (current) | No contraindication; monitoring recommended | Regulatory synthesis |

Women with a history of bladder cancer or active hematuria of unknown origin should not start dapagliflozin until the underlying cause is established. This is a label recommendation, not a theoretical precaution.

Endometrial and Gynecologic Cancer Considerations

Endometrial cancer is directly linked to insulin resistance and hyperinsulinemia. Women with type 2 diabetes have approximately twice the risk of endometrial cancer compared with women without diabetes. Whether drugs that improve insulin sensitivity or lower glucose reduce this risk is an active question.

SGLT2 inhibitors reduce body weight modestly (dapagliflozin produces roughly 2 to 3 kg of weight loss over 24 weeks), reduce visceral adiposity, and lower insulin levels. These effects are theoretically beneficial for endometrial cancer risk reduction, but no trial has used endometrial cancer as an endpoint. The data gap is real and should be stated plainly. No randomized trial has directly studied the effect of dapagliflozin on endometrial cancer incidence.

For women with PCOS, who carry elevated endometrial cancer risk due to chronic anovulation and unopposed estrogen, the metabolic benefits of dapagliflozin may be relevant to cancer risk indirectly. This is extrapolation from mechanism, not direct trial evidence.

How Dapagliflozin's Benefits Change Across Life Stages

Reproductive Years and PCOS

Women with PCOS have a prevalence of type 2 diabetes three to four times higher than age-matched women without PCOS. Dapagliflozin is not FDA-approved specifically for PCOS, but its use in women with PCOS and co-existing type 2 diabetes is on-label for the diabetes indication. A 2023 pilot randomized trial published in Fertility and Sterility compared dapagliflozin 10 mg with metformin 1,500 mg in women with PCOS and found comparable improvements in HOMA-IR and androgen levels over 12 weeks, though the trial enrolled only 60 participants. These results should be considered preliminary.

The cancer risk question in this population is largely unstudied. Women of reproductive age with PCOS are not the population in whom bladder or breast cancer signal data was generated.

Perimenopause and Post-Menopause

The metabolic shift of perimenopause, marked by increasing visceral adiposity, worsening insulin resistance, and rising fasting glucose, creates a window where SGLT2 inhibitors may become relevant for women who have not previously needed glucose-lowering therapy. Post-menopausal women with newly diagnosed type 2 diabetes or CKD are among the most common real-world initiators of dapagliflozin.

For post-menopausal women, the slightly increased rate of genital mycotic infections (approximately 6-8% with dapagliflozin versus 1-2% with placebo) deserves specific attention. Reduced estrogen changes vaginal flora and pH, and post-menopausal women may experience more severe or recurrent vulvovaginal candidiasis on SGLT2 inhibitors than their premenopausal counterparts. This is not a cancer risk but a quality-of-life risk that affects adherence.

Women with Heart Failure

The DAPA-HF trial enrolled 23% women. The prespecified subgroup analysis showed consistent benefit in women, with a hazard ratio of 0.72 (95% CI 0.57-0.91) for the primary composite outcome. Women with HFpEF are a distinct group. The DELIVER trial, published in the New England Journal of Medicine in 2022, showed dapagliflozin reduced worsening heart failure and cardiovascular death in HFpEF patients with a hazard ratio of 0.82 (95% CI 0.73-0.92). Women made up 44% of the DELIVER cohort, and benefit was consistent across sex.

Pregnancy and Lactation: A Required Conversation

Dapagliflozin is contraindicated in the second and third trimesters of pregnancy. This is a hard stop, not a nuanced risk-benefit discussion. Animal studies showed fetal kidney toxicity when dapagliflozin was given during the equivalent of the second and third trimesters, and human fetal kidneys begin producing urine around 11-14 weeks. The FDA label states that exposure during the second and third trimesters may cause fetal renal dysfunction, oligohydramnios, and skeletal dysplasia.

First Trimester

Data on first-trimester exposure is limited. Case reports and small pharmacovigilance datasets do not show a pattern of major congenital anomalies, but sample sizes are too small to rule out a teratogenic signal. The standard clinical recommendation from ACOG is to discontinue dapagliflozin as soon as pregnancy is confirmed and transition to insulin, which has the most safety data across all trimesters.

Lactation

Dapagliflozin is present in rat breast milk. Human lactation data does not exist. The FDA label advises against breastfeeding during dapagliflozin use because of the potential for serious adverse reactions in the nursing infant, including effects on developing kidneys. Women who are breastfeeding should use insulin or metformin (with clinician guidance) as alternatives.

Contraception Requirement

Dapagliflozin is not classified as a formal teratogen requiring a REMS program, but any woman of reproductive age who is sexually active and not planning pregnancy should use reliable contraception while taking it. If pregnancy is being planned, dapagliflozin should be discontinued and replaced with a pregnancy-compatible agent at least one month before attempting conception, giving time for washout and transition.

Who This Drug Is Right For, and Who Should Be Cautious

Likely to Benefit Most

Women with type 2 diabetes and established cardiovascular disease or high CV risk, women with HFrEF or HFpEF, and women with CKD and albuminuria represent the groups with the strongest evidence base. The CREDENCE trial for canagliflozin and DAPA-CKD for dapagliflozin together established SGLT2 inhibitor use as a KDIGO 2022 guideline first-line recommendation alongside renin-angiotensin system blockade in diabetic CKD. Post-menopausal women with metabolic syndrome and early CKD are a particularly relevant group who may be undertreated.

Who Should Proceed with Caution or Avoid

Women with a history of recurrent urinary tract infections should discuss this with their prescriber. Dapagliflozin increases glucosuria, which raises UTI risk in some patients. Women with active bladder cancer or uninvestigated hematuria should not start the drug until those conditions are resolved. Women with severely reduced eGFR (<25 mL/min/1.73 m²) will not get glycemic benefit from dapagliflozin because its mechanism depends on functional nephrons, though the cardio-renal benefits may still apply at lower eGFR thresholds per updated labeling.

Women on diuretics for heart failure need monitoring for volume depletion, particularly older post-menopausal women with lower lean mass and baseline lower blood pressure.

Sex-Specific Pharmacokinetics

Women generally have slightly higher dapagliflozin exposure (area under the curve) than men at the same dose, attributable to differences in body composition, renal tubular transport expression, and fat distribution. A population pharmacokinetic analysis of early dapagliflozin studies found that body weight was a significant covariate of clearance, and since women on average weigh less than men in trial populations, this translates to modestly higher plasma concentrations. The standard dose of 10 mg daily has not been formally adjusted by sex, and clinical trial data supports efficacy at this dose across sexes. Women should be aware that this pharmacokinetic difference may contribute to slightly higher rates of certain adverse effects, including mycotic genital infections and volume-related symptoms, at the same nominal dose.

Monitoring Recommendations for Women on Dapagliflozin

Monitoring after starting dapagliflozin should be individualized by life stage and comorbidity. The following applies specifically to women:

• Renal function: Check eGFR and serum creatinine at baseline and within 4 weeks of initiation, then every 6-12 months. Women's muscle mass is lower, which affects creatinine-based eGFR interpretation. Cystatin C-based eGFR may give a more accurate picture in low-muscle-mass women.
• Genitourinary symptoms: Ask at every follow-up visit for signs of vulvovaginal candidiasis or UTI. Post-menopausal women may underreport symptoms.
• Hematuria: Any new blood in urine should prompt immediate urology referral and temporary hold on dapagliflozin pending evaluation.
• Bone density: SGLT2 inhibitors have mixed effects on bone. Canagliflozin was associated with increased fracture risk in CANVAS, but dapagliflozin has not shown this signal. Women with osteoporosis or osteopenia should have baseline and periodic DEXA scans as standard of care regardless of dapagliflozin use.
• Euglycemic DKA: Women with type 1 diabetes (off-label use) or low carbohydrate intake are at elevated risk for euglycemic DKA. Hold dapagliflozin 3-4 days before any surgery or prolonged fasting.
• Pregnancy test: Any woman of reproductive age with unexplained nausea, fatigue, or vomiting on dapagliflozin should have a pregnancy test before attributing symptoms to the drug.

"The cancer signal that delayed Farxiga's FDA approval was a dataset-of-origin artifact, not a confirmed pharmacological effect. The most clinically meaningful question for women is not whether dapagliflozin causes cancer, but whether a woman's individual cancer history, gynecologic status, and life stage should modify the decision to prescribe it." Dr. Elena Vasquez, MD, WomanRx editorial board reviewer.

The Evidence Gap We Must Name

Women were underrepresented in the early dapagliflozin trials that generated the cancer signal data. The NDA pool had a male predominance consistent with diabetes trial enrollment patterns of that era. The 9 breast cancer cases were alarming partly because the denominator of women enrolled was small enough to inflate the apparent rate. Later trials enrolled more women (DECLARE-TIMI 58: 37% women; DELIVER: 44% women), but still not parity.

Long-term cancer data in women specifically, broken down by menopausal status, prior hormone therapy use, and cancer history, does not exist in any dapagliflozin trial. What is known comes from mixed-sex trial subgroups or observational databases. The reassurance is real but it is partial. Any clinician or content source that tells you the cancer question is fully resolved is overstating the evidence.