Capivasertib Hyperglycemia Protocol Aims to Keep Patients On Drug

Fasting glucose rose from baseline in 37% of patients on capivasertib (Truqap) in CAPItello-291, while 0.6% stopped the drug for hyperglycemia. The median first rise came at 15 days.

An AstraZeneca-funded briefing for US clinicians now walks through fasting checks and metformin so that spike does not knock patients off an AKT inhibitor that delayed progression. Expert notes in the piece may differ from the approved label, and the sponsor paid for the publication.

A 37% Glucose Rise and a 0.6% Exit Rate

The US label reports that increased fasting glucose from baseline in 37% of 355 patients treated with capivasertib plus fulvestrant. That lab finding included grade 2 events (fasting glucose >160 to 250 mg/dL) in 11%, grade 3 (>250 to 500 mg/dL) in 2%, and grade 4 (>500 mg/dL) in 1.1%. Random glucose rose in 57%.

Dose reductions for hyperglycemia were required in 0.6% of patients, and permanent discontinuation for hyperglycemia was required in 0.6%. Diabetic metabolic decompensation occurred in 0.6%, including diabetic ketoacidosis in 0.3%. The safety of the drug has not been established in type 1 diabetes or in type 2 diabetes that is uncontrolled or already treated with insulin, because those patients were kept out of the studies.

CAPITELLO-291 GLUCOSE FIGURES

• Fasting glucose rise: 37% of 355 patients on capivasertib plus fulvestrant.
• First onset: Median 15 days, range 1 to 367 days.
• Drug changes: Dose cut in 0.6%; permanent stop in 0.6%.
• Added glucose drugs: 12% (43 of 355) had a glucose-lowering regimen started or changed, including insulin in 4.8% (17 of 355).

Those stop rates sit far below the skin and gut numbers on the same label. Cutaneous reactions led to permanent discontinuation in 7% of patients. Diarrhea, which occurred in 72%, led to a permanent stop in 2%. Clinics still build long glucose protocols because ketoacidosis can be fatal, and because an earlier PI3K drug made high blood sugar the reason people left targeted therapy.

In the phase 3 CAPItello-291 trial, 708 patients with HR-positive, HER2-negative advanced breast cancer were randomized to capivasertib 400 mg twice daily, 4 days on and 3 days off, plus fulvestrant, or to placebo plus fulvestrant (355 vs 353). About 70% had already received a CDK4/6 inhibitor. In the 289 patients with a PIK3CA, AKT1, or PTEN alteration (155 vs 134), investigator-assessed progression-free survival was 7.3 months vs 3.1 months (hazard ratio 0.50, P<0.001). The US Food and Drug Administration approved the combination on November 16, 2023. About 76% of the altered group had a PIK3CA alteration.

Alpelisib Taught Clinics to Treat Glucose as a Stop Sign

AKT sits downstream of PI3K in insulin signaling. Block the pathway in a tumor, and you also blunt glucose uptake in muscle and fat and let the liver push out more glucose. That on-target, off-tumor effect is why this class carries a diabetes-style monitoring schedule.

The scare that still shapes clinic behavior comes from alpelisib (Piqray). In SOLAR-1, any-grade hyperglycemia occurred in 63.7% of patients on alpelisib plus fulvestrant, and grade 3-4 events in 36.6%. Hyperglycemia led to discontinuation in 6.3% in published phase 3 comparisons. Diabetic ketoacidosis has been reported in 0.7% with alpelisib and in 0.3% with capivasertib. Real-world series at Memorial Sloan Kettering found even higher alpelisib rates than the trial: any-grade hyperglycemia in 80.3% of patients treated off protocol vs 34.0% on trial.

Capivasertib’s clinical adverse-event term “hyperglycemia” was 16.3% any grade and 2.3% grade 3-4 in those same cross-trial tables, a different count from the 37% fasting-glucose lab rise on the US label. Inavolisib (Itovebi) sat in between on all-grade events (58.6%) with grade 3-4 hyperglycemia in 5.6% and a 0.6% hyperglycemia stop rate in INAVO120. Trial entry bars were not the same. CAPItello-291 allowed HbA1c below 8.0%, which includes people with diabetes. SOLAR-1 used HbA1c of 6.4% or lower. INAVO120 used HbA1c below 6.0%.

HYPERGLYCEMIA IN PHASE 3 PATHWAY TRIALS

The lower capivasertib AE rate arrived even though the trial enrolled a wider glucose range than the PI3K-alpha studies. Authors of a 2026 pathway-toxicity review also pointed to the 4-days-on, 3-days-off schedule as one reason peaks may ease. PI3K/AKT pathway changes are common in this breast cancer subtype, on the order of 50% of HR-positive, HER2-negative cases, with PIK3CA alterations in 30% to 40%, AKT1 in about 5%, and PTEN loss in 5% to 10%.

Front-Loaded Fasting Checks Through Week 8

The labeled schedule is bunched in the first two months, matching that 15-day median. Before the first dose, fasting glucose and HbA1c are drawn and fasting glucose is optimized. After start, fasting glucose is checked on day 3 or 4 of the dosing week during weeks 1, 2, 4, 6, and 8, then monthly, with HbA1c every 3 months.

THE LABELED FASTING GLUCOSE SCHEDULE

• Before day 1: Fasting glucose and HbA1c, then optimize fasting glucose.
• Weeks 1, 2, 4, 6, and 8: Fasting glucose on day 3 or 4 of the on-drug stretch.
• Month 3 onward: Monthly fasting glucose, and as needed.
• Every 3 months: HbA1c, plus extra fasting checks if glucose is already high.

If hyperglycemia appears, fasting glucose is monitored at least twice weekly, on days on and off the drug, until it returns to baseline. While a patient is on glucose-lowering medicine, the label calls for weekly fasting checks for 2 months, then every 2 weeks. Home meters are encouraged for people with risk factors. Hold the drug during suspected serious infection or acute illness, because those states raise the risk of severe hyperglycemia and ketoacidosis.

Primary prevention with metformin was not allowed in CAPItello-291. Even so, 12% needed a change in glucose-lowering drugs during the study. That gap between a reactive trial and a preventive clinic routine is what later expert papers try to close.

When Should Capivasertib Be Held for High Glucose?

Hold rules kick in once fasting glucose moves past 160 mg/dL. Below that band, the first move is a glucose-lowering pill, not a cancer-dose cut. Above it, the drug pauses until fasting glucose is 160 mg/dL or lower, and the clock to restart is 28 days.

FASTING GLUCOSE AND DOSE CHANGES

Dose steps are 320 mg twice daily, then 200 mg twice daily, still 4 days on and 3 days off. If the second cut is not tolerated, the drug is stopped. If ketoacidosis is suspected, hold at once. If it is confirmed, stop for good. The 160 mg/dL hold line is stricter than alpelisib’s labeled pause above 250 mg/dL, which means capivasertib clinics interrupt earlier even though severe events are less common.

Start Metformin Before the First Capivasertib Dose

The label is reactive. A December 4, 2025, expert opinion in npj Breast Cancer, led by breast medical oncologist Erika Hamilton, goes further for people already at risk: HbA1c of 5.7% or higher, prediabetes, body mass index of 30 or higher, or a high baseline fasting glucose.

we recommend initiating prophylactic extended-release metformin 500 mg/day one week before initiating capivasertib, escalating to 2000 mg/day, as needed and as tolerated for patients with an elevated risk of developing hyperglycemia

Erika Hamilton, breast medical oncologist, npj Breast Cancer expert opinion

That metformin a week before starting capivasertib is expert consensus, not a boxed FDA instruction, and the trial never tested it. The same group would add an SGLT2 inhibitor or a thiazolidinedione if metformin is not enough, and would keep insulin secretagogues off the list because they can cause low blood sugar. Insulin, in this pathway, is a short bridge. Some oncologists also worry that extra insulin can feed back into PI3K/AKT signaling, which is one reason SGLT2 drugs, which lower glucose without raising insulin, keep coming up in clinic notes.

Diet advice in the same paper is blunt: daily activity, a lower-carbohydrate pattern, small meals, and, if needed, a ketogenic cap under 50 g of carbohydrate a day, tailored so a cachectic patient is not treated like a patient with obesity. Steroids given for rash can lift glucose 1 to 2 days later, so a rash protocol can undo a glucose protocol if the two teams do not talk.

Claims Data Put Capivasertib Ahead of Alpelisib

A Komodo Research Dataset analysis led by Neil M. Iyengar, presented at the 43rd Annual Miami Breast Cancer Conference, compared adults with metastatic breast cancer who started capivasertib (411 patients) or alpelisib (144) on or after November 16, 2023, with no prior PI3K or AKT inhibitor. Mean ages were 61.7 and 61.8 years. Baseline diabetes was 18.7% vs 21.5%.

During treatment, any glucose-lowering agent was used in 28.2% of capivasertib patients vs 48.6% on alpelisib. Insulin use was 6.8% vs 16.0%. Relative to baseline, any-agent use rose 8.0 percentage points vs 23.6, and insulin rose 1.7 points vs 12.5. Among people without baseline diabetes, 19.0% vs 39.3% received a glucose-lowering agent. Among people not already on those drugs, 13.7% vs 35.2% started one, and 2.8% vs 13.7% started insulin.

Those gaps look clean until follow-up is on the table. Only 3.5% of capivasertib patients were still on treatment past 90 days, vs 39.6% of alpelisib patients, with median follow-up of 6 months vs 7 months. A shorter time on drug can hide glucose problems that appear later. The randomized 0.6% stop rate in CAPItello-291 does not have that problem. The claims file is a signal that community prescribing is using fewer diabetes drugs with capivasertib, not proof that the metabolic load stays low for a full year.

Why the Dosing Week Moves Blood Sugar

Continuous glucose monitoring in published case series shows the same sawtooth: readings climb on the 4 on-drug days and fall on the 3 off days, often within about 12 hours of the last dose. A fasting draw on an off day can look reassuring while post-meal spikes on an on day are already in the 200s. HbA1c, averaged over months, can miss that weekly swing. That is why the label pins early labs to day 3 or 4 of the dosing week, and why some teams now hand over a home meter in week 1 rather than wait for a missed clinic draw.

Hope S. Rugo, division chief of breast medical oncology at City of Hope, has been telling practices to treat a sudden illness as a reason to skip the next dose, then call in the morning, instead of taking the tablet, waiting on a callback, and landing in an emergency department with glucose in the 300s. One of her patients reached 491 mg/dL. Metformin plus a hold brought glucose down, and the drug restarted at a lower dose.

If you don’t feel good, don’t take the dose that evening. Call in in the morning.

Hope S. Rugo, division chief of breast medical oncology, City of Hope

The education packet’s last instruction is to keep patients on treatment. In CAPItello-291 that meant a 4.2-month gain in progression-free survival for the biomarker-altered group, with hyperglycemia as a lab event in 37% and a reason to stop in 0.6%. The protocol is the work that sits between those two numbers.

Frequently Asked Questions

What breast cancer is capivasertib approved to treat?

The US indication is HR-positive, HER2-negative locally advanced or metastatic breast cancer with one or more PIK3CA, AKT1, or PTEN alterations found by an FDA-authorized test, used with fulvestrant after progression on at least one endocrine-based regimen in the metastatic setting, or after recurrence on or within 12 months of finishing adjuvant therapy. Premenopausal and perimenopausal women also receive an LHRH agonist; men may receive one as well.

How is the capivasertib dose lowered if side effects persist?

The starting dose is 400 mg twice daily (two 200 mg tablets morning and evening) for 4 days, then 3 days off. The first reduction is 320 mg twice daily on the same schedule, and the second is 200 mg twice daily. If a patient cannot tolerate the second reduction, the drug is stopped for good. Tablets are swallowed whole and are not split, crushed, or chewed.

Is capivasertib used in people who already take insulin?

The US label states that safety has not been established in type 1 diabetes or in type 2 diabetes that is uncontrolled or requires insulin at baseline, because those patients were excluded from the clinical studies. People with well-controlled type 2 diabetes may need closer fasting checks and a stronger glucose-lowering plan if they start the drug.

What should a patient do after a missed or vomited dose?

If a dose is missed by 4 hours or less, it is taken. If more than 4 hours have passed, that dose is skipped and the next dose is taken at the usual time. If a patient vomits a dose, an extra tablet is not taken; the next dose stays on the regular clock.

What happens if diabetic ketoacidosis is suspected on capivasertib?

The drug is held immediately when ketoacidosis is suspected. If ketoacidosis is confirmed, capivasertib is permanently discontinued. Patients are told to contact a clinician for excessive thirst, dry mouth, more urination than usual, blurred vision, increased appetite with weight loss, abdominal pain, or trouble breathing, and to report fever, infection, or a new illness before the next dose.

Disclaimer: This article is news reporting and analysis of published trial results, US prescribing information, expert consensus, and claims data on capivasertib (Truqap) and hyperglycemia. It is for information only and is not medical advice, a treatment plan, or a substitute for the FDA-approved label. Readers should not start, stop, or change capivasertib, metformin, insulin, or any other medicine based on this article; those decisions belong to the treating oncologist, and to an endocrinologist or diabetes clinician when glucose is unstable. Figures, indications, and monitoring rules reflect the cited sources as of the dates on those documents and can change with new labeling or new evidence.