Sugar Crash on the PI3K Pathway: The First Documented Case of Inavolisib-Associated Diabetic Ketoacidosis

1. Introduction and Clinical Significance

The phosphatidylinositol-3-kinase (PI3K)/AKT/mTOR signaling pathway regulates cellular proliferation, survival, and metabolism. Activating PIK3CA mutations occur in approximately 40% of hormone receptor-positive (HR+), HER2-negative breast cancers, making PI3Kα-selective inhibitors a major therapeutic advance for this population [1].

Hyperglycemia is the most clinically significant metabolic toxicity of PI3Kα inhibitors, arising as a direct on-target effect of PI3Kα inhibition in insulin signaling. In clinical trials, all-grade hyperglycemia occurred in 63.7% of alpelisib-treated and 58.6% of inavolisib-treated patients [2,3,4]. While hyperglycemia is common and typically reversible with dose modification, progression to diabetic ketoacidosis (DKA) is rare and potentially fatal. A pharmacovigilance analysis of the FDA Adverse Event Reporting System identified 87 DKA cases associated with alpelisib, indicating a tenfold higher risk compared with other anticancer therapies [2]. Management remains an area of active investigation; current ADA 2026 guidelines recommend metformin as first-line therapy, with caution against SGLT2 inhibitors due to markedly elevated DKA risk in this setting [5].

Inavolisib is a next-generation, highly selective PI3Kα inhibitor recently approved in combination with palbociclib and fulvestrant for endocrine-resistant, PIK3CA-mutated, HR+/HER2− advanced breast cancer [6]. The FDA prescribing information warns that severe or fatal hyperglycemia, including ketoacidosis, can occur, and postmarketing fatal ketoacidosis has been reported [7]. However, one published case described inavolisib-induced hyperosmolar hyperglycemic state without ketoacidosis, and no detailed clinical case report of inavolisib-associated DKA has appeared in the peer-reviewed literature [8].

This case represents the first published peer-reviewed report of DKA associated with inavolisib. While hyperglycemia is a recognized on-target effect of PI3Kα inhibitors, progression to DKA is rare and potentially fatal. This report provides a practical clinical framework for early recognition and management of this life-threatening complication, reinforces the ADA recommendation for weekly glucose monitoring during the first two weeks of PI3Kα inhibitor therapy, and highlights the need for caution with SGLT2 inhibitors in this setting.

2. Case Presentation

2.1. Patient Demographics and Medical History

A 76-year-old woman presented to the emergency department with generalized weakness, inability to eat or drink, and a mechanical fall. Her oncologic history was notable for stage III hormone receptor-positive (HR+), human epidermal growth factor receptor 2-negative (HER2−) breast cancer diagnosed four years prior, for which she underwent left mastectomy and adjuvant therapy. She remained in remission until two months before this presentation, when surveillance imaging revealed new pulmonary and lymph node metastases consistent with disease recurrence. Molecular profiling of the recurrent disease identified a PIK3CA mutation. She was subsequently initiated on inavolisib in combination with palbociclib and fulvestrant two weeks prior to this admission.

Her medical history also included type 2 diabetes mellitus diagnosed ten years earlier, consistently well controlled on oral hypoglycemic therapy (pioglitazone) without ever requiring insulin. She also had hypertension, hyperlipidemia, and hypothyroidism. Her baseline glycosylated hemoglobin (HbA1c) two months prior to presentation was 5.8%, reflecting long-standing glycemic control. The patient reported full compliance with her medications, and there were no recent dietary changes, infections, or other identifiable precipitants. The only change in her medication regimen was the addition of inavolisib two weeks earlier.

2.2. Clinical Presentation and Diagnostic Assessment

On arrival, the patient was alert and oriented. Vital signs were as follows: temperature: 98.6°F, heart rate: 78 bpm, respiratory rate: 18 breaths/min, blood pressure: 139/71 mmHg, oxygen saturation: 98% on room air. Physical examination was notable for left mastectomy scars and vitiligo. The neurological examination was grossly normal. There were no signs of infection or acute decompensated heart failure. Laboratory evaluation revealed the following: serum glucose: >500 mg/dL, venous pH: 7.27, serum bicarbonate: 14 mmol/L, anion gap: 27, urine ketones: moderate, corrected sodium: 160 mmol/L, potassium: 4.4 mmol/L, creatinine: 2.0 mg/dL and HbA1c: 8.9%.

These findings were consistent with a diagnosis of diabetic ketoacidosis (DKA) with concurrent hypernatremia and acute kidney injury. The HbA1c of 8.9% represented a 3.1% increase from her baseline of 5.8% measured two months earlier, indicating rapid and severe glycemic deterioration temporally associated with inavolisib initiation.

2.3. Treatment and Hospital Course

The patient was admitted to the intensive care unit and managed with intravenous fluid resuscitation and a continuous insulin infusion per institutional DKA protocol. Once the anion gap closed and the patient tolerated oral intake, the insulin infusion was transitioned to subcutaneous insulin.

The endocrinology service was consulted and recommended ongoing glucose monitoring with insulin dose titration. The hematology-oncology service recommended withholding inavolisib. Following multidisciplinary discussion, inavolisib was permanently discontinued given the severity of the metabolic derangement.

The patient's DKA resolved, and she demonstrated rapid clinical and metabolic recovery. She was discharged in stable condition on her home medication, pioglitazone, with close outpatient follow-up.

2.4. Outcome and Follow-Up

At her most recent follow-up, approximately two months after discharge, her HbA1c had normalized to 4.7%, confirming the reversibility of the hyperglycemic episode following inavolisib discontinuation. Pioglitazone was subsequently stopped given the normalized glycemic status. Her metastatic breast cancer remained stable, and she was continued on single-agent fulvestrant.

2.5. Ethical and Methodological Disclosures

Informed consent was obtained from the patient for publication of this case report. Generative artificial intelligence tools were not used in the design, data collection, analysis, or interpretation of this case.

3. Discussion

This case reflects a broader trend in modern oncology: as targeted therapies improve cancer outcomes, their metabolic and endocrine toxicities are emerging as significant sources of morbidity requiring multidisciplinary management, a growing field recently termed oncoendocrinology [9].

Our patient had multiple recognized risk factors for severe PI3Kα inhibitor-induced hyperglycemia: preexisting type 2 diabetes, age >65 years, and likely elevated BMI. In the INAVO120 trial, hyperglycemia occurred in 65.5% of patients with BMI ≥30.0 versus 56.8% with BMI <30.0 [5]. In the phase I/Ib GO39374 study, patients with two or more risk factors had hyperglycemia rates of 87.9%, with grade 3–4 events in 39.4% [10]. A real-world cohort from Memorial Sloan Kettering reported any-grade hyperglycemia in 58% of 40 inavolisib-treated patients, with a median time to onset of 42 days [11].

The close temporal relationship, DKA developing approximately two weeks after starting inavolisib, in the absence of alternative precipitants such as infection, steroid use, or other pharmacologic triggers, strongly supports a causal association. Importantly, the other components of the patient's regimen are not known to cause clinically significant hyperglycemia. Fulvestrant does not list hyperglycemia among its adverse reactions; in the SERENA-2 trial, hyperglycemia occurred in only 1.4% at grade 1, and a FAERS analysis of 6,947 fulvestrant reports did not identify hyperglycemia as a safety signal [12]. Palbociclib is similarly not associated with hyperglycemia in adult humans; FAERS disproportionality analyses did not identify hyperglycemia as a safety signal [13]. The INAVO120 trial provides the most direct evidence: fasting glucose elevation occurred in 85% of patients receiving the inavolisib-containing triplet versus 43% receiving palbociclib–placebo–fulvestrant, with grade 3–4 hyperglycemia in 12% versus 0%, respectively [12]. Furthermore, the patient had well-controlled diabetes on pioglitazone alone for a decade with a baseline HbA1c of 5.8%, and the only pharmacologic change preceding DKA was the addition of inavolisib. After discontinuation, her glucose normalized and HbA1c declined to 4.7% while she continued fulvestrant without recurrence of hyperglycemia.

Our patient was already on pioglitazone, which is recommended as a second-line agent for PI3Kα inhibitor-induced hyperglycemia per ADA guidelines [14]. Yet pioglitazone was insufficient to prevent DKA, likely due to its slow onset of action [14]. This suggests that patients with preexisting diabetes may require additional prophylactic measures, such as metformin initiation, before starting PI3Kα inhibitors. During hospitalization, insulin was appropriately used for acute DKA management; however, for chronic glycemic management, insulin and sulfonylureas should be considered only as a last resort, as increased insulin levels may reactivate the PI3K pathway and counteract antitumor effects [8,14]. Unlike immune checkpoint inhibitor-induced diabetes, which involves autoimmune β-cell destruction and typically requires lifelong insulin replacement, PI3Kα inhibitor-associated hyperglycemia reflects on-target pharmacologic insulin resistance that is generally reversible with drug discontinuation, a distinction with important implications for long-term management [9]. This reversibility was clearly demonstrated in our patient.

Had weekly glucose monitoring been performed during the first two weeks, the hyperglycemia might have been detected before progressing to DKA. The FDA label recommends monitoring fasting glucose every 3 days for the first week, then weekly for weeks 2–4, then every 2 weeks for weeks 5–12, then every 4 weeks thereafter. As the field of oncoendocrinology continues to evolve, standardized, therapy-specific endocrine monitoring frameworks, as recently proposed by Shapoo et al.[9], will be essential to prevent, detect, and manage metabolic complications across the growing landscape of targeted cancer therapies.

This report has inherent limitations. As a single case, causality can be inferred but not definitively established. The patient's preexisting diabetes and age represent confounders that may have lowered her threshold for DKA. C-peptide and insulin levels were not measured during the acute episode, which would have helped characterize the mechanism of ketoacidosis. Despite these limitations, the strong temporal association, absence of alternative precipitants, and rapid metabolic recovery after drug discontinuation support a probable causal relationship. Prospective studies are needed to identify predictive biomarkers for severe hyperglycemic events, define optimal prophylactic antihyperglycemic strategies for high-risk patients, and clarify the safety of SGLT2 inhibitors in combination with PI3Kα inhibitors. As the indication for inavolisib expands, registry-based studies capturing real-world metabolic outcomes across diverse patient populations will be essential to refine risk stratification and monitoring protocols.

4. Conclusions

This case represents the first detailed peer-reviewed report of DKA associated with inavolisib. The close temporal relationship, absence of alternative precipitants, and rapid metabolic recovery after drug discontinuation strongly support a causal association. Well-controlled preexisting diabetes did not protect against severe metabolic decompensation, underscoring that all patients initiated on PI3Kα inhibitors require vigilant glucose monitoring, particularly during the first two weeks of therapy, consistent with current ADA recommendations. Clinicians should maintain a low threshold for evaluating ketoacidosis in patients presenting with nonspecific symptoms during PI3Kα inhibitor treatment. Multidisciplinary collaboration between oncology and endocrinology teams is essential to optimize metabolic surveillance and guide management decisions. Prospective registries are needed to better define the incidence, risk factors, and optimal management of PI3Kα inhibitor–associated DKA.

Figure 1. Dual role of the PI3K/AKT/mTOR pathway in tumor growth and glucose metabolism. PI3K inhibition reduces tumor proliferation but impairs insulin signaling, leading to hyperglycemia and diabetic ketoacidosis. PI3K – Phosphatidylinositol 3-Kinase, AKT – Protein Kinase B (PKB), commonly referred to as AKT, mTOR – Mechanistic Target of Rapamycin,GLUT4 - Glucose Transporter Type 4, DKA – Diabetic Ketoacidosis.

Figure 1. Dual role of the PI3K/AKT/mTOR pathway in tumor growth and glucose metabolism. PI3K inhibition reduces tumor proliferation but impairs insulin signaling, leading to hyperglycemia and diabetic ketoacidosis. PI3K – Phosphatidylinositol 3-Kinase, AKT – Protein Kinase B (PKB), commonly referred to as AKT, mTOR – Mechanistic Target of Rapamycin,GLUT4 - Glucose Transporter Type 4, DKA – Diabetic Ketoacidosis.

5. Patents

This section is not mandatory but may be added if there are patents resulting from the work reported in this manuscript.