Tersolisib

• CAS: 2883540-92-7
• 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-/m1/s1
• InChIKey: LGPNQALKGDDVBD-CYBMUJFWSA-N

Tersolisib: Mutant-Selective PI3Kα Inhibitor

Tersolisib (STX-478, LY4064809) is a synthetic organic small molecule that functions as an allosteric, mutant-selective inhibitor of the phosphoinositide 3-kinase catalytic subunit alpha (PI3Kα, encoded by PIK3CA) [1]. It is chemically defined as (R)-1-(2-aminopyrimidin-5-yl)-3-(1-(5,7-difluoro-3-methylbenzofuran-2-yl)-2,2,2-trifluoroethyl)urea, with a molecular weight of 401.29 g/mol [2]. Tersolisib is designed to preferentially target oncogenic kinase-domain mutants of PI3Kα, such as H1047R, while sparing the wild-type enzyme, thereby addressing the on-target toxicity limitations of earlier PI3Kα inhibitors [3].

Mutant-selective PI3Kα (H1047R) inhibitor

Reported to spare wild-type PI3Kα signaling

Oral, predicted CNS-penetrant for brain metastasis research

• [1] Buckbinder L, et al. STX-478, a Mutant-Selective, Allosteric PI3Kα Inhibitor Spares Metabolic Dysfunction and Improves Therapeutic Response in PI3Kα-Mutant Xenografts. Cancer Discov. 2023 Nov 1;13(11):2432-2447. doi: 10.1158/2159-8290.CD-23-0396. View Source
• [2] GuidetoImmunopharmacology (GtoPdb). tersolisib ligand page. Ligand ID: 12903. Accessed 2026. View Source
• [3] National Cancer Institute Thesaurus. Mutant-selective PI3K-alpha H1047X Inhibitor STX-478. Accessed 2026. View Source

Why Tersolisib Selectivity Matters

Substituting Tersolisib with a generic or alternative PI3Kα inhibitor is not scientifically equivalent due to fundamental differences in mechanism and selectivity. First-generation PI3Kα inhibitors like alpelisib are orthosteric, ATP-competitive agents that inhibit both wild-type and mutant PI3Kα with equal potency, leading to significant on-target toxicities such as hyperglycemia and rash due to wild-type PI3Kα inhibition in metabolic tissues [1]. Tersolisib's allosteric, mutant-selective mechanism is designed to spare wild-type PI3Kα, thereby maintaining antitumor efficacy while avoiding these metabolic liabilities [2]. Furthermore, Tersolisib demonstrates CNS penetration, a property not shared by many other PI3Kα inhibitors, which is critical for targeting brain metastases . These distinct pharmacological profiles preclude simple interchangeability and underscore the need for compound-specific procurement based on experimental objectives.

Mechanism

Allosteric, mutant-selective

ATP-competitive, pan-PI3Kα

Metabolic impact

Reported insulin signaling sparing

Reported hyperglycemia & insulin resistance

CNS access

Predicted CNS penetration

Limited CNS penetration

• [1] Andre F, et al. Alpelisib for PIK3CA-Mutated, Hormone Receptor-Positive Advanced Breast Cancer. N Engl J Med. 2019;380(20):1929-1940. doi: 10.1056/NEJMoa1813904. View Source
• [2] Buckbinder L, 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. doi: 10.1158/2159-8290.CD-23-0396. View Source

Tersolisib: Comparative Evidence Guide

Mutant-Selective PI3Kα Potency vs. Alpelisib

Tersolisib demonstrates a 14-fold higher selectivity for the prevalent PI3Kα H1047R kinase-domain mutant over the wild-type enzyme, a property not shared by alpelisib. In biochemical assays, Tersolisib inhibited PI3Kα H1047R with an IC50 of 9.4 nmol/L, compared to an IC50 of 131 nmol/L for wild-type PI3Kα, representing a 14-fold selectivity window [1]. Under identical conditions, alpelisib showed no mutant selectivity, exhibiting equipotent inhibition of both mutant and wild-type PI3Kα [2].

Mutant vs. wild-type IC50

Head-to-head

Tersolisib
H1047R IC50: 9.4 nM
WT IC50: 131 nM
Selectivity: 14-fold

vs

Alpelisib
IC50: ~4.6 nM (both)
Selectivity: 1-fold

Reported mutant selectivity supports pathway-specific study with reduced wild-type perturbation.

Biochemical assay; recombinant enzymes.

Evidence Dimension	Biochemical inhibition of PI3Kα (H1047R mutant vs. wild-type)
Target Compound Data	IC50 = 9.4 nmol/L (H1047R mutant); IC50 = 131 nmol/L (wild-type); Selectivity ratio = 14-fold
Comparator Or Baseline	Alpelisib: IC50 ~4.6 nmol/L for both mutant and wild-type (equipotent); Selectivity ratio = 1
Quantified Difference	Tersolisib: 14-fold mutant selectivity; Alpelisib: 1-fold (non-selective)
Conditions	Biochemical PI3Kα activity assay using recombinant enzymes

Why This Matters

This mutant selectivity directly translates to a reduced risk of on-target toxicities (hyperglycemia, rash) in vivo by sparing wild-type PI3Kα in metabolic tissues, a key procurement advantage for studies requiring chronic dosing.

• [1] Buckbinder L, 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. doi: 10.1158/2159-8290.CD-23-0396. Data from Fig. 1B and text. View Source
• [2] Fritsch C, 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):1117-29. doi: 10.1158/1535-7163.MCT-13-0865. View Source

Metabolic Sparing in Adipocytes vs. Alpelisib

In primary human subcutaneous adipocytes, Tersolisib demonstrates a substantially higher therapeutic index by sparing insulin-stimulated glucose uptake, a key driver of hyperglycemia. Alpelisib inhibited glucose uptake with an EC50 of ~300 nmol/L and a maximal inhibition (Emax) of 88%, whereas Tersolisib achieved only 38% maximal inhibition, even at high concentrations [1]. This indicates that Tersolisib's effect on wild-type PI3Kα signaling in metabolic tissues is significantly attenuated.

Glucose uptake sparing

Head-to-head

Tersolisib
Emax: 38%

vs

Alpelisib
Emax: 88%

Lower maximal inhibition of glucose uptake may support metabolic endpoint review in chronic models.

Primary human adipocytes, 10 nM insulin.

Evidence Dimension	Insulin-stimulated glucose uptake in primary human subcutaneous adipocytes
Target Compound Data	Maximal inhibition (Emax) of glucose uptake = 38%
Comparator Or Baseline	Alpelisib: Maximal inhibition (Emax) of glucose uptake = 88%
Quantified Difference	Tersolisib achieves 50 percentage points less maximal inhibition of glucose uptake compared to alpelisib.
Conditions	Primary human subcutaneous adipocytes treated with 10 nmol/L insulin; glucose uptake measured via [3H]-2-deoxy-glucose incorporation.

Why This Matters

This cellular data provides a mechanistic basis for the avoidance of hyperglycemia in vivo, a critical differentiator that impacts the feasibility of long-term preclinical efficacy studies and potential clinical translation.

• [1] Buckbinder L, 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. doi: 10.1158/2159-8290.CD-23-0396. Data from Fig. 2F and text. View Source

In Vivo Metabolic Dysfunction vs. Alpelisib

In non-tumor-bearing mice, Tersolisib avoids the dose-dependent insulin resistance and hyperglycemia observed with alpelisib. Following 5 days of repeat dosing, alpelisib at 50 mg/kg caused a significant reduction in glucose disposal in both insulin tolerance tests (ITT) and oral glucose tolerance tests (OGTT), indicative of insulin resistance [1]. In contrast, Tersolisib at 100 mg/kg (a dose providing robust antitumor efficacy) did not produce significant changes in glucose AUC or fasting plasma glucose [2].

In vivo glucose homeostasis

Head-to-head

Tersolisib
100 mg/kg: no significant change

vs

Alpelisib
50 mg/kg: reduced glucose disposal

Avoidance of insulin resistance may enable longer-term high-dose research models.

BALB/c nude mice, 5-day repeat dosing.

Evidence Dimension	In vivo glucose homeostasis (ITT and OGTT) in non-tumor-bearing BALB/c nude mice
Target Compound Data	No significant change in glucose AUC at 100 mg/kg QD after 5 days
Comparator Or Baseline	Alpelisib: Significant dose-dependent reduction in glucose disposal at 50 mg/kg QD after 5 days
Quantified Difference	Tersolisib avoids alpelisib-induced insulin resistance at a dose 2-fold higher.
Conditions	Female BALB/c nude mice, 5-day repeat dosing, ITT and OGTT performed on day 5.

Why This Matters

This in vivo evidence confirms that Tersolisib's mutant selectivity translates to a meaningful reduction in on-target metabolic toxicity, a major limitation of non-selective PI3Kα inhibitors, thus enabling more prolonged and higher-dose treatment regimens in preclinical models.

• [1] Buckbinder L, 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. doi: 10.1158/2159-8290.CD-23-0396. Data from Fig. 3A-D and text. View Source
• [2] Buckbinder L, 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. doi: 10.1158/2159-8290.CD-23-0396. Text regarding glucose AUC and fasting glucose. View Source

Antitumor Efficacy in HNSCC Xenografts vs. Alpelisib

In a CAL-33 (H1047R PI3Kα mutant) human head and neck squamous cell carcinoma (HNSCC) xenograft model, Tersolisib demonstrated antitumor efficacy comparable to or better than alpelisib, but without metabolic dysregulation. Tersolisib at 30 mg/kg QD achieved similar tumor growth inhibition as alpelisib at 20 mg/kg QD, while Tersolisib at 100 mg/kg QD produced superior tumor regression [1]. Importantly, these efficacy doses of Tersolisib did not cause the hyperglycemia and insulin resistance observed with alpelisib in the same study [2].

Antitumor efficacy HNSCC

Head-to-head

Tersolisib
100 mg/kg: tumor regression

vs

Alpelisib
20 mg/kg: growth inhibition, limited by toxicity

Reported tumor regression with metabolic sparing suggests wider research window.

CAL-33 xenograft, daily oral dosing for 21 days.

Evidence Dimension	Antitumor efficacy in CAL-33 H1047R mutant HNSCC xenograft model
Target Compound Data	30 mg/kg QD: tumor growth inhibition comparable to alpelisib 20 mg/kg; 100 mg/kg QD: robust tumor regression, superior to alpelisib.
Comparator Or Baseline	Alpelisib: 20 mg/kg QD: tumor growth inhibition; higher doses limited by toxicity.
Quantified Difference	Tersolisib achieves similar or superior efficacy at doses that do not induce metabolic dysfunction, unlike alpelisib.
Conditions	CAL-33 HNSCC xenograft in nude mice, daily oral dosing for 21 days.

Why This Matters

This demonstrates that Tersolisib maintains robust antitumor activity while avoiding the dose-limiting toxicities of non-selective inhibitors, effectively widening the therapeutic window and making it a more suitable tool for long-term in vivo oncology studies.

• [1] Buckbinder L, 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. doi: 10.1158/2159-8290.CD-23-0396. Data from Fig. 4A-B and text. View Source
• [2] Buckbinder L, 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. doi: 10.1158/2159-8290.CD-23-0396. Text regarding metabolic parameters in efficacy study. View Source

Kinome-Wide Selectivity Profile

Tersolisib exhibits exquisite kinome-wide selectivity, minimizing off-target kinase inhibition. In a biochemical screen of 373 kinases representing approximately 70% of the human kinome, including PI3Kβ, PI3Kδ, and PI3Kγ isoforms, Tersolisib at 10 µmol/L inhibited only the Aurora B kinase (AurB) by more than 50% (IC50 = 1.6 µmol/L) [1]. Follow-up cellular assays confirmed minimal AurB inhibition at concentrations up to 10 µmol/L, indicating that the biochemical activity is not functionally relevant at therapeutic concentrations [2].

Kinome selectivity

Reported

1 of 373 kinases

AurB only (>50% at 10 µM)

Low off-target kinase inhibition supports precise pathway probing.

Biochemical panel, no functional PI3Kβ/δ/γ inhibition.

Evidence Dimension	Biochemical inhibition of a panel of 373 kinases
Target Compound Data	At 10 µmol/L, >50% inhibition only for AurB (IC50 = 1.6 µmol/L); no significant inhibition of PI3Kβ, PI3Kδ, PI3Kγ.
Comparator Or Baseline	Class-level inference: Many ATP-competitive PI3K inhibitors show broader kinome inhibition profiles.
Quantified Difference	>50% inhibition of only 1 out of 373 kinases tested.
Conditions	Biochemical kinase activity assay at 10 µmol/L compound concentration.

Why This Matters

This high degree of selectivity reduces the likelihood of confounding off-target effects in cellular and in vivo experiments, enhancing the interpretability of research findings and the compound's value as a precise chemical probe.

• [1] Buckbinder L, 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. doi: 10.1158/2159-8290.CD-23-0396. Data from Supplementary Fig. S1A. View Source
• [2] Buckbinder L, 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. doi: 10.1158/2159-8290.CD-23-0396. Data from Supplementary Fig. S1B and text. View Source

CNS Penetration Potential

Tersolisib is characterized as an oral, CNS-penetrant inhibitor, a feature that distinguishes it from many other PI3Kα inhibitors. The compound's physicochemical properties (molecular weight 401.29, XLogP 0.73, topological polar surface area 101.1 Ų) are consistent with blood-brain barrier permeability [1]. While direct quantitative brain-to-plasma ratios are not available in the primary literature, the compound is explicitly designed and described as CNS-penetrant . This property is a significant differentiator for research targeting brain metastases of PI3Kα-mutant cancers.

CNS penetration

Data to verify

XLogP 0.73 TPSA 101.1 Ų Described as CNS-penetrant

Predicted CNS penetration supports research on intracranial PI3Kα-mutant models.

In silico prediction; confirm brain-to-plasma ratio.

Evidence Dimension	Predicted CNS penetration based on physicochemical properties
Target Compound Data	XLogP = 0.73, TPSA = 101.1 Ų, MW = 401.29; meets criteria for CNS penetration.
Comparator Or Baseline	Alpelisib (MW 441.5, XLogP ~3.4) and other PI3Kα inhibitors may have lower predicted CNS penetration.
Quantified Difference	Favorable physicochemical properties for BBB penetration; described as CNS-penetrant in multiple sources.
Conditions	In silico prediction and vendor description.

Why This Matters

For researchers studying brain metastases or CNS tumors with PI3Kα mutations, Tersolisib offers a potential advantage over non-CNS-penetrant PI3K inhibitors, enabling direct targeting of intracranial disease.

• [1] GuidetoImmunopharmacology (GtoPdb). tersolisib ligand page. Ligand ID: 12903. Accessed 2026. Physicochemical properties section. View Source

Tersolisib: Application Scenarios

Chronic Dosing in PI3Kα-Mutant Tumor Models

Tersolisib's ability to avoid hyperglycemia and insulin resistance at efficacious doses (100 mg/kg QD) [1] makes it the preferred choice for long-term in vivo xenograft or PDX studies where metabolic toxicity from non-selective PI3Kα inhibitors (e.g., alpelisib) would confound results or limit dosing duration [2].

PI3Kα Biology: Avoiding Metabolic Perturbations

For researchers dissecting the specific role of oncogenic PI3Kα mutations in tumor cell signaling, Tersolisib's mutant selectivity (14-fold over wild-type for H1047R [1]) and its lack of effect on insulin-stimulated glucose uptake in primary adipocytes [2] provide a cleaner pharmacological tool compared to pan-PI3Kα inhibitors. This is particularly relevant in studies combining PI3Kα inhibition with metabolic modulators (e.g., insulin, metformin) or immunotherapies.

Targeting CNS Tumors and Brain Metastases

Given its CNS-penetrant properties [1], Tersolisib is uniquely suited for preclinical models of brain metastases from PI3Kα-mutant breast cancer or primary CNS tumors with PIK3CA mutations. Non-CNS-penetrant PI3Kα inhibitors are ineffective against intracranial disease, making Tersolisib the compound of choice for such research applications.

Combination Therapy: CDK4/6 & Endocrine Agents

Preclinical data demonstrate that Tersolisib combines effectively with fulvestrant and CDK4/6 inhibitors (e.g., palbociclib) to produce robust and durable tumor regression in ER+/HER2- breast cancer models [1]. Its favorable metabolic profile suggests these combinations can be administered chronically without exacerbating hyperglycemia, a potential concern with alpelisib-based combinations.

Application

Selection Property

Validation Focus

Chronic dosing in PI3Kα-mutant tumor models

Mutant-selective PI3Kα inhibition with metabolic sparing

Glucose homeostasis and insulin sensitivity endpoints

PI3Kα mutant signaling studies without metabolic perturbation

Selective inhibition of mutant PI3Kα, sparing wild-type insulin signaling

Adipocyte glucose uptake and pathway specificity

CNS tumor and brain metastasis research models

Predicted CNS penetration

Brain-to-plasma ratio and intracranial efficacy endpoints

Combination studies with CDK4/6 and endocrine agents

Compatibility with fulvestrant and palbociclib

Combination efficacy and metabolic endpoints in long-term models

• [1] Buckbinder L, 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. doi: 10.1158/2159-8290.CD-23-0396. View Source
• [2] Kearney AL, Vasan N. A New Wave of PI3Kα Inhibitors. Cancer Discov. 2023;13(11):2313-2315. doi: 10.1158/2159-8290.CD-23-0945. View Source