(S)-STX-478

• Molecular Formula: C16H12F5N5O2
• Molecular Weight: 401.29 g/mol
• InChI: InChI=1S/C16H12F5N5O2/c1-6-9-2-7(17)3-10(18)12(9)28-11(6)13(16(19,20)21)26-15(27)25-8-4-23-14(22)24-5-8/h2-5,13H,1H3,(H2,22,23,24)(H2,25,26,27)/t13-/m0/s1
• InChIKey: LGPNQALKGDDVBD-ZDUSSCGKSA-N

(S)-STX-478 is the only commercially available PI3Kα inhibitor combining allosteric mutant selectivity (15-fold K_D preference for H1047R over WT) with documented CNS penetration. Unlike alpelisib, it spares WT PI3Kα-mediated glucose homeostasis—validated by 0% Grade ≥3 hyperglycemia in Phase 1 trials where 48% of patients were pre-diabetic or diabetic. Essential for protocols requiring selective interrogation of oncogenic PI3Kα signaling without confounding metabolic toxicity. Choose (S)-STX-478 for brain metastasis models, combination studies with CDK4/6 inhibitors, and translational programs where metabolic safety is non-negotiable.

(S)-STX-478 (S-Tersolisib): A Mutant-Selective Allosteric PI3Kα Inhibitor for Precision Oncology Research and Procurement

(S)-STX-478 (CAS 2883540-91-6) is the S-enantiomer of STX-478 (tersolisib), an orally bioavailable, CNS-penetrant, allosteric inhibitor of phosphatidylinositol 3-kinase alpha (PI3Kα) [1]. Unlike ATP-competitive PI3Kα inhibitors that bind the orthosteric active site and inhibit both wild-type (WT) and mutant enzyme with equal potency, STX-478 occupies a cryptic allosteric pocket that forms upon a major conformational rearrangement of residues 936–940 in the PI3Kα kinase domain, enabling selective inhibition of oncogenic kinase-domain mutants (e.g., H1047R) while largely sparing WT PI3Kα in healthy metabolic tissues [1]. The compound is currently in Phase 1/2 clinical evaluation (NCT05768139) as monotherapy and in combination with fulvestrant and CDK4/6 inhibitors for HR+/HER2− breast cancer and other PI3Kα-mutant advanced solid tumors [2].

• [1] Buckbinder L, St Jean DJ Jr, Tieu T, Ladd B, Hilbert B, Wang W, et al. STX-478, a mutant-selective, allosteric PI3Kα inhibitor spares metabolic dysfunction and improves therapeutic response in PI3Kα-mutant xenografts. Cancer Discov. 2023;13(11):2432–2447. View Source
• [2] Juric D, Giordano A, Jhaveri K, Munster P, Rodón Ahnert J, LoRusso P, et al. Abstract PS7-02: First-in-human results of STX-478, a mutant-selective PI3K alpha inhibitor, in HR+ breast cancer and advanced solid tumor patients. Clin Cancer Res. 2025;31(12_Supplement):PS7-02. View Source

Why (S)-STX-478 Cannot Be Substituted by Generic PI3Kα Inhibitors in Mutant-Selective Research Applications

The PI3Kα inhibitor class encompasses compounds with fundamentally divergent mechanisms of target engagement. FDA-approved agents such as alpelisib are ATP-competitive orthosteric inhibitors that inhibit WT PI3Kα and mutant enzyme equipotently, resulting in on-target WT toxicities—most notably severe hyperglycemia and rash—that necessitate dose reductions, antidiabetic co-medication, and in some cases discontinuation [1]. In contrast, (S)-STX-478 binds a distinct allosteric site and achieves mutant selectivity through differential binding kinetics: its equilibrium dissociation constant (K_D) is 4.7 nM for H1047R mutant PI3Kα versus 70 nM for WT, translating to a 15-fold affinity difference [1]. This mechanistic distinction is not shared by orthosteric inhibitors or even by other mutant-selective PI3Kα inhibitors (e.g., inavolisib, which is an ATP-competitive inhibitor/degrader), meaning that substituting one PI3Kα inhibitor for another in research protocols will produce non-equivalent pharmacological profiles with respect to target engagement, metabolic perturbation, and therapeutic index [2].

• [1] Buckbinder L, St Jean DJ Jr, Tieu T, Ladd B, Hilbert B, Wang W, et al. STX-478, a mutant-selective, allosteric PI3Kα inhibitor spares metabolic dysfunction and improves therapeutic response in PI3Kα-mutant xenografts. Cancer Discov. 2023;13(11):2432–2447. View Source
• [2] Kearney AL, Vasan N. A new wave of PI3Kα inhibitors. Cancer Discov. 2023;13(11):2313–2315. View Source

(S)-STX-478 Quantitative Differentiation Evidence: Head-to-Head Data Against Key PI3Kα Comparators

Biochemical Mutant Selectivity: 14-Fold Preferential Inhibition of H1047R Mutant PI3Kα Over Wild-Type vs. Alpelisib's Complete Lack of Selectivity

In direct head-to-head biochemical assays comparing STX-478 with alpelisib, STX-478 inhibited the most prevalent oncogenic PI3Kα variant H1047R with an IC50 of 9.4 nM and showed 14-fold selectivity over WT PI3Kα (IC50 = 131 nM). Under identical assay conditions, alpelisib exhibited equipotent inhibition of mutant and WT PI3Kα with no measurable mutant selectivity [1]. Surface plasmon resonance confirmed this differential binding: STX-478 had a K_D of 4.7 nM for H1047R versus 70 nM for WT PI3Kα, a 15-fold affinity difference driven by kinetic discrimination (slower off-rate from mutant) rather than binding conformation differences [1].

Evidence Dimension	Biochemical IC50 for PI3Kα H1047R mutant vs. WT enzyme
Target Compound Data	H1047R IC50 = 9.4 nM; WT PI3Kα IC50 = 131 nM; fold-selectivity = 14×; K_D H1047R = 4.7 nM; K_D WT = 70 nM
Comparator Or Baseline	Alpelisib: equipotent against H1047R and WT PI3Kα (no mutant selectivity; IC50 values overlapping)
Quantified Difference	STX-478: 14-fold selectivity for H1047R over WT; Alpelisib: ~1-fold (non-selective)
Conditions	In vitro biochemical enzyme activity assay; PI3Kα H1047R and WT proteins; surface plasmon resonance for K_D determination

Why This Matters

Mutant selectivity at the biochemical level is the mechanistic basis for STX-478's ability to spare WT PI3Kα in healthy tissues, a property absent in alpelisib and most orthosteric PI3Kα inhibitors, making (S)-STX-478 the appropriate choice for experiments requiring selective interrogation of mutant PI3Kα signaling without confounding WT pathway inhibition.

• [1] Buckbinder L, St Jean DJ Jr, Tieu T, Ladd B, Hilbert B, Wang W, et al. STX-478, a mutant-selective, allosteric PI3Kα inhibitor spares metabolic dysfunction and improves therapeutic response in PI3Kα-mutant xenografts. Cancer Discov. 2023;13(11):2432–2447. View Source

Metabolic Sparing in Primary Human Adipocytes: ~100-Fold Higher EC50 for Glucose Uptake Inhibition vs. Alpelisib

In primary human subcutaneous adipocytes—a physiologically relevant model of insulin-dependent glucose disposal—alpelisib inhibited [³H]-2-deoxy-glucose uptake at concentrations as low as 100 nM, with near-complete inhibition at 10 μM. In contrast, the concentration of STX-478 required to achieve 50% inhibition (EC50) of glucose uptake was ≥10 μM, an approximately 100-fold rightward shift [1]. The maximum inhibitory effect (Emax) of STX-478 on glucose uptake was 38%, versus 88% for alpelisib, indicating that even at saturating concentrations STX-478 only partially suppresses WT PI3Kα-mediated glucose transport [1].

Evidence Dimension	Inhibition of insulin-stimulated glucose uptake in primary human subcutaneous adipocytes
Target Compound Data	EC50 ≥ 10 μM; Emax = 38%
Comparator Or Baseline	Alpelisib: EC50 ≈ 100 nM (estimated from concentration-response; significant inhibition at 100 nM with near-complete inhibition at 10 μM); Emax = 88%
Quantified Difference	~100-fold higher EC50 for STX-478 vs. alpelisib; Emax: 38% (STX-478) vs. 88% (alpelisib), a 50-percentage-point lower maximal suppression of glucose uptake
Conditions	Primary human subcutaneous adipocytes; pretreated with compound; stimulated with 10 nM insulin; [³H]-2-deoxy-glucose uptake assay

Why This Matters

This data directly quantifies STX-478's functional sparing of WT PI3Kα in a clinically relevant metabolic tissue. For procurement decisions, this is the single most predictive preclinical assay of whether a PI3Kα inhibitor will cause clinical hyperglycemia—alpelisib's dose-limiting toxicity that affects the majority of patients.

• [1] Buckbinder L, St Jean DJ Jr, Tieu T, Ladd B, Hilbert B, Wang W, et al. STX-478, a mutant-selective, allosteric PI3Kα inhibitor spares metabolic dysfunction and improves therapeutic response in PI3Kα-mutant xenografts. Cancer Discov. 2023;13(11):2432–2447. View Source

In Vivo Metabolic Safety: Absence of Insulin Resistance with STX-478 vs. Dose-Dependent Insulin Resistance with Alpelisib in Mice

In a direct head-to-head in vivo metabolic comparison, non-tumor-bearing BALB/c nude mice received repeat dosing of STX-478 (30, 100, or 300 mg/kg q.d.) or alpelisib (20 or 50 mg/kg q.d.) for 5 days, followed by insulin tolerance test (ITT) and oral glucose tolerance test (OGTT). Alpelisib caused a dose-dependent reduction in glucose disposal in both ITT and OGTT, consistent with overt insulin resistance, whereas STX-478 treatment was not associated with significant changes in glucose AUC at any dose, with only a non-statistically significant increase at the 300 mg/kg dose in ITT [1]. Notably, STX-478 at 100 mg/kg q.d.—a dose that produced antitumor efficacy comparable to alpelisib 50 mg/kg in xenograft models—had no effect on body weight or fasting plasma glucose, while alpelisib at 50 mg/kg caused clear metabolic dysregulation [1].

Evidence Dimension	Insulin sensitivity (ITT and OGTT) and metabolic parameters after 5-day repeat dosing in mice
Target Compound Data	STX-478 (30, 100, 300 mg/kg q.d.): no significant change in glucose AUC in ITT or OGTT at any dose; no effect on body weight or fasting plasma glucose at 100 mg/kg
Comparator Or Baseline	Alpelisib (20, 50 mg/kg q.d.): dose-dependent reduction in glucose disposal in both ITT and OGTT; overt insulin resistance at 50 mg/kg dose
Quantified Difference	STX-478 at efficacious dose (100 mg/kg) produced no metabolic dysfunction; alpelisib at efficacious dose (50 mg/kg) caused significant insulin resistance and impaired glucose disposal
Conditions	Female BALB/c nude mice; 5-day repeat dosing; ITT and OGTT performed after treatment; body weight and fasting glucose measured

Why This Matters

This in vivo evidence demonstrates that the mutant-selective mechanism translates into a genuine metabolic safety advantage at efficacious exposures. For researchers designing combination regimens or chronic dosing studies, this metabolic sparing profile means STX-478 can be combined with other agents without the confounding metabolic toxicities and counterregulatory hyperinsulinemia that limit alpelisib-based protocols.

• [1] Buckbinder L, St Jean DJ Jr, Tieu T, Ladd B, Hilbert B, Wang W, et al. STX-478, a mutant-selective, allosteric PI3Kα inhibitor spares metabolic dysfunction and improves therapeutic response in PI3Kα-mutant xenografts. Cancer Discov. 2023;13(11):2432–2447. View Source

Clinical Hyperglycemia Grade Profile: 0% Grade ≥3 Hyperglycemia with STX-478 vs. 36.6% Grade ≥3 with Alpelisib in Controlled Trials

In the Phase 1 portion of the first-in-human trial (NCT05768139), 61 heavily pre-treated patients with PI3Kα-mutant advanced solid tumors received STX-478 monotherapy at 20–160 mg daily. Hyperglycemia of any grade occurred in 23% of patients, but all hyperglycemia events were Grade 1 or 2, and no Grade ≥3 hyperglycemia was observed. No patient discontinued due to an adverse event [1]. In contrast, the Phase 3 SOLAR-1 trial of alpelisib plus fulvestrant in HR+ advanced breast cancer reported Grade 3 hyperglycemia in 32.7% of patients and Grade 4 hyperglycemia in 3.9%, for a combined Grade ≥3 rate of 36.6% [2]. Critically, the STX-478 trial enrolled a high-risk population that included patients with pre-diabetes or controlled Type 2 diabetes (48%) and those with prior PI3K pathway inhibitor intolerance (41%), populations typically excluded from alpelisib trials [1].

Evidence Dimension	Incidence and severity of treatment-emergent hyperglycemia
Target Compound Data	STX-478 (20–160 mg q.d.): any-grade hyperglycemia 23%; Grade 1/2 only; Grade ≥3 hyperglycemia 0%; 0% discontinuation due to AE (n=61, all solid tumors)
Comparator Or Baseline	Alpelisib + fulvestrant (SOLAR-1 Phase 3): Grade 3 hyperglycemia 32.7%; Grade 4 hyperglycemia 3.9%; combined Grade ≥3 36.6% (n=284)
Quantified Difference	Grade ≥3 hyperglycemia: STX-478 0% vs. alpelisib 36.6%; any-grade hyperglycemia: STX-478 23% vs. alpelisib 66% (SOLAR-1 any-grade rate)
Conditions	STX-478: Phase 1 dose escalation (NCT05768139), patients with advanced PI3Kα-mutant solid tumors, including 48% pre-diabetic/diabetic and 41% with prior PI3K pathway inhibitor exposure. Alpelisib: Phase 3 SOLAR-1 trial, HR+/HER2− advanced breast cancer, patients with controlled diabetes excluded

Why This Matters

The complete absence of Grade ≥3 hyperglycemia with STX-478—even in a higher-risk population that included diabetic patients—versus a 36.6% rate with alpelisib represents a clinically meaningful differentiation in tolerability. For procurement decisions involving translational or clinical research programs, this safety profile directly impacts protocol feasibility, patient retention, and the ability to maintain continuous target coverage without dose interruptions.

• [1] Juric D, Giordano A, Jhaveri K, Munster P, Rodón Ahnert J, LoRusso P, et al. Abstract PS7-02: First-in-human results of STX-478, a mutant-selective PI3K alpha inhibitor, in HR+ breast cancer and advanced solid tumor patients. Clin Cancer Res. 2025;31(12_Supplement):PS7-02. View Source
• [2] Rugo HS, André F, Yamashita T, Cerda H, Toledano I, Stemmer SM, et al. Time course and management of key adverse events during the randomized phase III SOLAR-1 study of PI3K inhibitor alpelisib plus fulvestrant in patients with HR-positive advanced breast cancer. Ann Oncol. 2020;31(8):1001–1010. View Source

CNS Penetration: Blood-Brain Barrier Crossing Capability Absent in Most Clinically Available PI3Kα Inhibitors

STX-478 was specifically designed to cross the blood-brain barrier (BBB), a property not shared by alpelisib or most other orthosteric PI3Kα inhibitors [1]. This design feature addresses an important clinical need: PI3Kα-mutant breast cancers have a propensity for brain metastasis, yet currently approved PI3Kα inhibitors lack meaningful CNS exposure. The oral bioavailability and CNS-penetrant properties of STX-478 have been highlighted by clinical investigators as a key differentiating advantage for patients with or at risk of brain metastases [2]. Quantitative brain-to-plasma ratios or CSF pharmacokinetic data from the ongoing Phase 1/2 trial (NCT05768139) have not yet been publicly disclosed, so this evidence dimension carries a class-level inference rating pending full pharmacokinetic publication.

Evidence Dimension	Blood-brain barrier penetration capability
Target Compound Data	STX-478: designed as CNS-penetrant; confirmed BBB crossing in preclinical models; clinical confirmation pending full PK data disclosure
Comparator Or Baseline	Alpelisib and most orthosteric PI3Kα inhibitors: not designed for CNS penetration; limited BBB crossing due to physicochemical properties and transporter efflux; no established CNS activity
Quantified Difference	Qualitative differentiation: STX-478 is CNS-penetrant by design; alpelisib and related orthosteric PI3Kα inhibitors are not. Quantitative brain-to-plasma ratios not yet publicly available
Conditions	Preclinical BBB penetration assays; clinical CNS PK pending full reporting from NCT05768139

Why This Matters

For research programs focused on brain metastasis models, leptomeningeal disease, or glioblastoma with PI3Kα mutations, the CNS-penetrant property of (S)-STX-478 makes it the only viable PI3Kα-targeted tool compound currently available, as alternative PI3Kα inhibitors lack sufficient BBB penetration to achieve pharmacologically relevant CNS exposures.

• [1] Juric D, Giordano A, Jhaveri K, Munster P, Rodón Ahnert J, LoRusso P, et al. Abstract PS7-02: First-in-human results of STX-478, a mutant-selective PI3K alpha inhibitor, in HR+ breast cancer and advanced solid tumor patients. Clin Cancer Res. 2025;31(12_Supplement):PS7-02. View Source
• [2] Montero AJ, interviewed by OncLive. Dr Montero on potential advantages with STX-478 in PI3Kα-mutant solid tumors. OncLive. Published November 5, 2024. View Source

Kinome-Wide Selectivity: Only 1 of 373 Kinases Inhibited >50% at 10 µM, Contrasting with Broader Kinome Activity of ATP-Competitive PI3Kα Inhibitors

STX-478 was profiled against a biochemical panel of 373 kinases representing approximately 70% of the human kinome, including PI3Kβ, PI3Kδ, and PI3Kγ isoforms. At a concentration of 10 μM, a single off-target kinase—Aurora B (AurB)—was inhibited by >50% (IC50 = 1.6 μM); follow-up cellular assays confirmed negligible AurB inhibition at concentrations up to 10 μM [1]. This exquisite kinome-wide selectivity is a direct consequence of STX-478's allosteric binding mode: it occupies a cryptic pocket unique to PI3Kα rather than the highly conserved ATP-binding site targeted by orthosteric inhibitors [1]. Class-level inference from published kinome profiling of ATP-competitive PI3Kα inhibitors such as alpelisib indicates broader off-target kinase activity due to ATP-site conservation across the kinome [2].

Evidence Dimension	Kinome-wide selectivity profile at 10 μM compound concentration
Target Compound Data	STX-478: 1 kinase inhibited >50% (AurB, IC50 = 1.6 μM) out of 373 kinases tested; AurB cellular activity negligible at ≤10 μM; PI3Kβ, PI3Kδ, PI3Kγ isoforms not inhibited >50% at 10 μM
Comparator Or Baseline	ATP-competitive PI3Kα inhibitors (class reference): broader kinome activity profiles due to ATP-binding site homology across kinases; specific comparator kinome data for alpelisib in the same 373-kinase panel not available
Quantified Difference	STX-478 hits only 0.27% (1/373) of kinases at >50% inhibition at 10 μM. Alpelisib kinome profiling data from separate panels reported multiple off-target kinases with sub-μM activity, though direct panel-to-panel comparison is not possible
Conditions	Biochemical kinase inhibition screen; 373 kinases; STX-478 tested at 10 μM; cellular follow-up for AurB up to 10 μM

Why This Matters

The near-absolute kinome selectivity of STX-478 reduces the probability of off-target kinase-driven confounding effects in cellular and in vivo experiments. For researchers using STX-478 as a chemical probe to interrogate PI3Kα-mutant biology, this clean selectivity profile provides higher confidence that observed phenotypes are attributable to PI3Kα mutant inhibition rather than polypharmacology.

• [1] Buckbinder L, St Jean DJ Jr, Tieu T, Ladd B, Hilbert B, Wang W, et al. STX-478, a mutant-selective, allosteric PI3Kα inhibitor spares metabolic dysfunction and improves therapeutic response in PI3Kα-mutant xenografts. Cancer Discov. 2023;13(11):2432–2447. View Source
• [2] Fritsch C, Huang A, Chatenay-Rivauday C, Schnell C, Reddy A, Liu M, et al. Characterization of the novel and specific PI3Kα inhibitor NVP-BYL719 and development of the patient stratification strategy for clinical trials. Mol Cancer Ther. 2014;13(5):1117–1129. View Source

Optimal Research and Procurement Application Scenarios for (S)-STX-478 Based on Verified Differentiation Evidence

Mutant-Selective PI3Kα Chemical Probe Studies Requiring WT-Sparing Pharmacology

For research programs that require selective pharmacological interrogation of oncogenic PI3Kα signaling without confounding inhibition of WT PI3Kα-dependent metabolic pathways, (S)-STX-478 is the evidence-supported tool compound of choice. Its 14-fold biochemical selectivity for H1047R mutant over WT PI3Kα (IC50 = 9.4 vs. 131 nM), 15-fold K_D differential (4.7 vs. 70 nM), and functional sparing of insulin-mediated glucose uptake in primary human adipocytes (EC50 ≥10 μM, Emax = 38%) [1] provide a mechanistic foundation that no orthosteric PI3Kα inhibitor can replicate. Alpelisib and inavolisib both lack this mutant-vs-WT discrimination and will produce WT-driven metabolic phenotypes that confound interpretation of PI3Kα-mutant-specific biology [1].

In Vivo Xenograft Efficacy Studies Requiring Metabolic Safety for Combination Regimens

When designing in vivo efficacy studies involving prolonged dosing or combination with agents that themselves affect metabolism (e.g., CDK4/6 inhibitors, endocrine therapies), (S)-STX-478's demonstrated absence of insulin resistance at efficacious doses provides a critical experimental advantage. In head-to-head mouse studies, STX-478 at 100 mg/kg q.d. produced antitumor efficacy comparable to alpelisib 50 mg/kg without altering glucose AUC in ITT/OGTT, body weight, or fasting glucose, while alpelisib caused dose-dependent insulin resistance [1]. This metabolic safety profile enables cleaner interpretation of combination efficacy data and avoids the confounding effects of hyperinsulinemia—which has been shown to counteract PI3K inhibition efficacy in tumor models [1].

Brain Metastasis or CNS Tumor Models Leveraging Blood-Brain Barrier Penetration

(S)-STX-478 is specifically designed to cross the blood-brain barrier, a property absent from alpelisib and most clinically available PI3Kα inhibitors [1]. For preclinical models of breast cancer brain metastasis, leptomeningeal disease, or primary CNS tumors harboring PI3Kα mutations, (S)-STX-478 is the only PI3Kα-targeted tool compound with documented CNS-penetrant properties currently available for research procurement [2]. This application scenario is supported by clinical investigators highlighting STX-478's potential to address the unmet need of PI3Kα-mutant tumors with CNS involvement [2].

Clinical Translational Research with High-Risk Patient Populations Including Diabetics

For translational research programs or clinical trial design involving patient populations with pre-existing metabolic comorbidities, (S)-STX-478's clinical safety data showing 0% Grade ≥3 hyperglycemia—even in a cohort where 48% of patients were pre-diabetic or diabetic—establishes a tolerability profile that is fundamentally distinct from alpelisib's 36.6% Grade ≥3 hyperglycemia rate in SOLAR-1 [1]. The STX-478 Phase 1 trial also enrolled patients with prior PI3K pathway inhibitor intolerance (41%), demonstrating feasibility of use in populations for whom alpelisib-based therapy is contraindicated [1]. This evidence supports (S)-STX-478 as the PI3Kα inhibitor of choice for protocols requiring metabolic safety in high-risk or heavily pre-treated populations.

• [1] Buckbinder L, St Jean DJ Jr, Tieu T, Ladd B, Hilbert B, Wang W, et al. STX-478, a mutant-selective, allosteric PI3Kα inhibitor spares metabolic dysfunction and improves therapeutic response in PI3Kα-mutant xenografts. Cancer Discov. 2023;13(11):2432–2447. View Source
• [2] Juric D, Giordano A, Jhaveri K, Munster P, Rodón Ahnert J, LoRusso P, et al. Abstract PS7-02: First-in-human results of STX-478, a mutant-selective PI3K alpha inhibitor, in HR+ breast cancer and advanced solid tumor patients. Clin Cancer Res. 2025;31(12_Supplement):PS7-02. View Source