PI3K signaling in breast cancer: therapeutic advances versus adaptive resistance

Somatic alterations in the phosphoinositide 3-kinase (PI3K) pathway are frequent oncogenic drivers in cancer and are particularly relevant in estrogen receptor–positive breast cancer, where PI3K signaling integrates hormonal, growth factor, and metabolic inputs. Therapeutic targeting of PI3K has been limited by toxicity, pathway crosstalk, and adaptive resistance. Recent advances in structural and functional characterization of PIK3CA mutations have enabled the development of more selective inhibitors, including PI3K isoform–selective and emerging allosteric PI3Ka mutant–targeted compounds. These approaches aim to improve the therapeutic index while maintaining pathway inhibition. In parallel, combination strategies with endocrine therapy and CDK4/6 inhibitors have demonstrated clinical benefit in selected patient populations. However, responses are often transient owing to compensatory signaling and pathway reactivation. Additional metabolic and transcriptional adaptations further contribute to sustaining tumor cell survival. Overall, PI3K-targeted therapy is evolving along two complementary paths: increased mutant-selective inhibition to improve therapeutic index and broader pathway-level strategies designed to overcome adaptive resistance and achieve more durable responses. Ongoing efforts focus on optimizing combination strategies, enhancing development of predictive biomarkers, and improving therapeutic sequencing to achieve durable responses while minimizing toxicity.

PI3Ks are lipid kinases that translate extracellular stimuli into intracellular programs regulating proliferation, survival, metabolism, and cell fate (21).

PI3K isoforms and activation. Based on structural organization and substrate specificity, PI3Ks are classified into three classes. Class I PI3Ks are further subdivided into class IA and IB, each consisting of a p110 catalytic subunit associated with a regulatory subunit. In the human genome, class IA PI3Ks comprise three catalytic isoforms — p110α (encoded by the PIK3CA gene), p110β (PIK3CB), and p110δ (PIK3CD) — and five regulatory isoforms or splice variants: p85α, p85β, p55γ, p55α, and p50α. Upon activation, class I PI3Ks catalyze the generation of phosphatidylinositol 3,4,5-trisphosphate (PIP3) from PIP2. PIP3 is the main second messenger generated by PI3K. At the plasma membrane, PIP3 acts as the docking site for several pleckstrin homology domain–containing signaling proteins, such as phosphoinositide-dependent kinase-1 (PDK1) and AKT (PKB) (22, 23). The lipid phosphatase PTEN antagonizes the function of PI3K by dephosphorylating PIP3 back to PIP2 (7, 8). Upon PIP3-mediated recruitment to the plasma membrane, AKT is phosphorylated at Thr308 by PDK1 and at Ser473 by mTOR complex 2 (mTORC2), achieving full activation (24). Activated AKT phosphorylates multiple substrates that regulate protein synthesis, glucose metabolism, apoptosis, and cell survival, including TSC2, PRAS40, FOXO, and others (25) (Figure 1A).

Figure 1

PI3K signaling and crosstalk with ERα and MAPK pathways. (A) RTK activation recruits class IA PI3K (p85α/p110α), promoting conversion of PIP2 to PIP3 at the plasma membrane. PIP3 recruits PDK1 and AKT, enabling AKT phosphorylation by PDK1 (on Thr308) and by mTORC2 (on Ser473). AKT inhibits the TSC1/2 complex, relieving suppression of Rheb and activating mTORC1, which drives mRNA translation, lipid and nucleotide synthesis, and inhibits autophagy through S6K1 signaling. PDK1 can activate SGK1, which can also phosphorylate and inhibit TSC2, thereby sustaining AKT-independent residual mTORC1 activity. Both AKT and SGK1 phosphorylate FOXO3, promoting its cytoplasmic sequestration and functional inactivation. When not phosphorylated, FOXO3 translocates to the nucleus where it induces transcriptional programs that compensate for PI3K inhibition. Additional regulatory inputs from LKB1/AMPK signaling and PTEN-mediated PIP3 dephosphorylation are shown as negative modulators of the pathway. (B) Crosstalk between PI3K signaling, MAPK, and ERα happens at multiple nodes. PI3K can activate MAPK via interaction with KRAS. Activated AKT can phosphorylate ERα, activating ligand-independent transcription of ER-target genes. Ligand activated and/or phosphorylated ERα can bind RTKs via adaptor proteins (e.g., Src) and favor phosphorylation of RTKs at tyrosines in their C-terminus. These residues provide docking sites for p85 and activation of p110α, the catalytic subunit of PI3K.

ERα/MAPK signaling pathways crosstalk. In ER⁺ breast cancer, PI3K signaling operates as a central signaling hub that integrates ERa activity with growth factor, metabolic, and stress-response pathways. This integration relies on bidirectional crosstalk between PI3K and ERa signaling (Figure 1B). ERα can rapidly activate PI3K through nongenomic mechanisms, either via RTKs or through direct interaction with the p85 regulatory subunit (26, 27). Conversely, PI3K/AKT signaling enhances ER transcriptional output via phosphorylation of ERa and its coregulators (28, 29).

PI3K signaling also converges with the MAPK/ERK pathway downstream of shared RTKs (Figure 1B), with reciprocal feedback loops shaping proliferative responses and therapeutic adaptation (30). Coupling of PI3K to mTOR and AMPK coordinates estrogen-driven growth with nutrient availability and cellular energy status, whereas interactions with the WNT/β-catenin, NF-κB, and TGF-β pathways further modulate differentiation, survival, and immune-related programs. Collectively, these interconnected signaling networks enable ER⁺ breast cancer cells to maintain transcriptional and metabolic plasticity and underlie adaptive resistance to both antiestrogen and PI3K-targeted therapies.

Oncogenic PIK3CA alterations. PIK3CA encodes p110α, which consists of five distinct domains (Figure 2A): the p85 adaptor-binding domain (ABD), the Ras-binding domain (RBD), the C2 domain, the helical domain, and the kinase domain (31). Its regulatory subunit, p85α, is also composed of five domains, referred to as SH3, breakpoint-cluster region homology (BH), N-terminal SH2 (nSH2), inter-SH2 domain (iSH2), and C-terminal SH2 (cSH2) (31).

Figure 2

Landscape of PIK3CA mutations and their predicted functional consequences. (A) Lollipop plot of the PIK3CA gene, illustrating exons, domain organization, and position of point mutations. The GENIE cohort v19.0-public was interrogated for PIK3CA mutation frequency, type, and position in the breast cancer cohort (n = 6,812 specimens/6,044 patients) (5). PIK3CA (p110α) comprises five functional domains: the adaptor-binding domain (ABD, in green), which mediates interaction with the regulatory p85 subunit; the Ras-binding domain (RBD, in red), which binds active RAS proteins and contributes to enzyme activation; the C2 domain (in blue), involved in membrane association and interaction with p85; the helical domain (in yellow), which regulates kinase activity through interactions with the p85 nSH2 domain; and the kinase domain (in purple), which contains the catalytic site responsible for phosphorylation of PIP2. (B) Schematic representation of PI3K activation and oncogenic PI3Kα mutations. In the basal state (left), PI3K is maintained in an inactive conformation through inhibitory intramolecular interactions within the regulatory p85 subunit (N- and C-terminal SH2 domains) and the catalytic p110α subunit. Upon growth factor stimulation, activated RTKs recruit p85 via SH2-phosphotyrosine interactions, thus releasing the nSH2-mediated inhibition of p110α. This allows A-loop exposure and engagement of PIP2 in the membrane, resulting in enzymatic activation and conversion of PIP2 to PIP3. RAS-GTP can further enhance membrane recruitment and activation. Oncogenic mutations promote constitutive PI3K activation through distinct mechanisms. Helical domain mutations (E542K/E545K) disrupt the inhibitory interaction between the p85 nSH2 domain and p110α, enabling PI3K activation even in absence of RTK activation. In contrast, the H1047R mutation enhances membrane interaction and A-loop protrusion, increasing catalytic activity.

Gain-of-function (hypermorphic) mutations in PIK3CA are among the most common alterations across tumor types (32). In breast cancer, the most common mutations (“hotspots”) are H1047R in the kinase domain, followed by E545K and E542K in the helical domain (33–35). Recent biochemical and molecular studies have revealed that different mutations activate PI3K via distinct mechanisms. Moreover, accumulating evidence indicates that a smaller but biologically important subset of PIK3CA alterations may exhibit neomorphic properties, including altered regulatory dependencies and mutant-specific signaling outputs that extend beyond simple amplification of PI3K signaling (7).

In the prevailing model for PI3K activation (Figure 2B), p85 couples with phosphotyrosine (PY) residues in RTKs, which releases the inhibitory nSH2 contacts with the p110α catalytic subunit (19). Notably, this conformational change is mimicked by the helical domain mutations (19). As a result, the exposed nSH2 domain binds with higher affinity to tyrosine residues in pYXXM motifs on activated RTKs and/or their adaptor proteins (36). Conversely, mutations in the C-terminus of the kinase domain, especially H1047R, are associated with increased membrane binding through a conformational change reorientating its Wnt-inhibitory factor (WIF) binding motif and with enhanced kinase activity due to the flexibility of the activation loop (37). Importantly, this conformational change exposes a cryptic pocket near position H1047 (37, 38), which has been exploited as a potential binding site for allosteric inhibitors (39, 40).

Aberrant activation of the PI3K pathway can occur through additional mechanisms, including amplification of oncogenic RTKs (41), genetic inactivation of tumor suppressors regulating the pathway (42), or stimulation by upstream growth factor receptors such as insulin receptor tyrosine kinase (INSR) and the related insulin-like growth factor 1 receptor (IGF1R) (41).

Finally, although deletions in PIK3CA are rare, most reported deletions cluster in a narrow region of critical residues within the C2 domain. Such deletions represent about 10% of all PIK3CA mutations, and disrupt binding with the iSH2 domain of p85a, thereby derepressing p110α activity (43, 44). Tumor-associated mutations in PIK3R1 and PIK3R2, the genes that encode p85a and p85b, respectively, have also been described (45). By destabilizing the interaction with their corresponding p110 isoforms or by destabilizing PTEN, these mutations aberrantly activate PI3K and generate tumor cell dependence on PI3K signaling.

The high prevalence of PIK3CA mutations across solid tumors and the identification of recurrent “hotspot” activating alterations established PI3Kα as a compelling therapeutic target. However, translating this biological rationale into clinical success has proved challenging, with decades of research resulting in only few PI3K pathway–targeting compounds approved by the US FDA. This limited therapeutic armamentarium largely reflects the central role of PI3K signaling in normal cellular homeostasis and physiology, such that its inhibition frequently leads to substantial, dose-limiting toxicities (DLTs) (46). Based on their mechanism of action, PI3K inhibitors can be broadly classified into (a) pan-PI3K inhibitors, (b) orthosteric PI3Kα inhibitors that directly bind the ATP-binding pocket of the p110α catalytic subunit and block its kinase activity, and (c) a newer class of allosteric PI3Kα inhibitors (Table 1 and Figure 3).

Figure 3

PI3K/AKT/mTOR signaling pathway inhibitors and their sites of action. Therapeutics have been developed targeting PI3K, AKT, mTOR, or all three (PAM inhibitors).

Table 1

Therapeutic PI3K/AKT pathway inhibitors

First generation: pan-PI3K inhibitors. Early compound development focused on pan-PI3K inhibitors with the goal of achieving total pathway blockade. These trials collectively failed to translate into clinical practice but provided critical lessons that fundamentally reshaped the development of subsequent PI3K-targeted therapies. Among the candidate compounds in this class, buparlisib was tested in patients with HR⁺/HER2^– breast cancer in the BELLE clinical trials (47, 48). Preclinical investigations had shown that compensatory PI3K crosstalk with ERα required a combination approach in HR⁺ breast cancers (49, 50). Thus, the selective ER degrader (SERD) fulvestrant was tested in combination with buparlisib, but the strategy showed an unfavorable risk-benefit ratio: a limited improvement in progression-free survival (PFS) was accompanied by serious adverse effects, such as altered liver enzymes, hyperglycemia, and neuropsychiatric symptoms (48, 51). Collectively, these clinical and preclinical findings highlighted two priorities for future drug development: improving isoform selectivity to enhance the therapeutic index and combining PI3K inhibitors with endocrine therapies to overcome adaptive signaling.

Second generation: orthosteric PI3Kα-selective inhibitors. The second generation of PI3K inhibitors consisted of compounds selectively targeting the PI3Kα isoform. These PI3Kα-selective agents demonstrated improved pharmacologic tolerability and on-target pathway suppression, resulting in meaningful clinical activity in patients with molecularly selected tumors (52, 53). The clinical activity of taselisib was evaluated in the SANDPIPER trial, which met its primary endpoint of improved PFS in patients with PIK3CA-mutant HR⁺/HER2^– breast cancer. However, the magnitude of benefit was modest, and treatment was associated with a high incidence of adverse events — including diarrhea, hyperglycemia, and colitis — leading to frequent dose reductions and treatment discontinuations (54). This toxicity profile may partly reflect the incomplete isoform selectivity of taselisib. Despite showing greater activity against mutant PI3Kα than against the wild-type enzyme, this drug also inhibits PI3Kβ, heavily expressed in the intestinal and colonic epithelium, and PI3Kγ and PI3Kδ, which are predominantly expressed in immune cells (54).

Alpelisib was the first PI3Kα inhibitor approved by FDA, following the results of the phase III SOLAR-1 trial (55). SOLAR-1 enrolled patients with HR⁺/HER2^– advanced breast cancer who had progressed on or after prior endocrine therapy. Patients were stratified in two cohorts based on PIK3CA mutation status and randomized to receive fulvestrant in combination with alpelisib or placebo. SOLAR-1 established PIK3CA mutation status as a predictive biomarker of response. While no difference in PFS was observed between treatment arms in the wild-type cohort, in patients with PIK3CA-mutated tumors, median PFS was 11 months in the alpelisib–fulvestrant group compared with 5.7 months in the control group (55). Median overall survival (OS) was 39.3 months in the alpelisib arm versus 31.4 months in the placebo arm, although this difference did not reach statistical significance (52). However, wide clinical use of alpelisib has been limited by metabolic toxicity, particularly hyperglycemia, and by the modest durability of response (34). Notably, SOLAR-1 completed patient enrollment before CDK4/6 inhibitors became the standard of care, therefore only 5.9% of patients in this trial had received a prior CDK4/6 inhibitor. To address this point, the phase II single-arm BYLIEVE and the phase III EPIK-B5 studies were designed, evaluating the efficacy of alpelisib in the post-CDK4/6 inhibitor setting (56, 57). In the BYLIEVE trial, in the patients receiving alpelisib median PFS was 8.0 months and median OS was 27.3 months (56). In the EPIK-B5, the median PFS was 7.4 months in the fulvestrant plus alpelisib arm versus 2.8 months in the fulvestrant plus placebo arm, while OS is not yet available (57). However, the emergence of AKT inhibitors, which have demonstrated clinically meaningful efficacy with a favorable therapeutic index in patients with PIK3CA/AKT1/PTEN-altered tumors after CDK4/6 inhibitors, may limit the clinical applicability of these findings (58, 59).

Inavolisib (GDC-0077) is a PI3Kα-selective, PI3Kb-sparing agent with improved pharmacologic properties and enhanced target engagement compared with alpelisib (53). Indeed, inavolisib also promotes degradation of the mutant p110α protein, potentially enabling prolonged suppression of oncogenic signaling (53). This dual mechanism enhances antitumor activity while minimizing off-target inhibition of other PI3K isoforms. Importantly, the rationale for sustained PI3K pathway suppression is further supported by the extensive functional crosstalk among ER, CDK4/6, and PI3K signaling networks (60, 61), which cooperatively drive endocrine therapy resistance. These observations provided the biological basis for combined targeting of all three pathways in the phase III INAVO120 trial (62), where the addition of inavolisib to palbociclib and fulvestrant significantly improved PFS compared with palbociclib and fulvestrant in patients with PIK3CA-mutated, endocrine-resistant HR⁺/HER2^– advanced breast cancer. Notably, the median PFS was 15 months in the triplet-combination arm versus 7.3 months in the control arm of fulvestrant plus palbociclib (62). There was also a statistically significant OS benefit of 7 months, the first to be shown in a phase III trial evaluating a PI3K inhibitor–based combination (62). Collectively, these findings suggest that simultaneous blockade of ERa, CDK4/6, and PI3K more efficiently eliminates pathway crosstalk, thus abrogating treatment resistance and disease progression. However, hyperglycemia, which can result from blockade of AKT2-mediated glucose uptake in muscle and adipose cells together with increased hepatic glucose production, was reported by approximately 63% of patients (all grades), of which 6.8% were grade ≥3 (62). Stomatitis/mucosal inflammation emerged as a notable toxicity with inavolisib, occurring at substantially higher rates than observed with alpelisib in the SOLAR-1 or BYLIEVE studies. This was primarily managed with dexamethasone mouthwash, with dose modifications required in a minority of patients. Despite the addition of a third agent, treatment discontinuation due to adverse events with the inavolisib triplet remained low and compared favorably with historical rates reported for alpelisib-based doublets (62).

Allosteric mutant-selective PI3Kα inhibitors. Therapeutic selectivity against mutant PIK3CA is intended to preserve antitumor efficacy while reducing on-target metabolic toxicities, particularly hyperglycemia. The improved tolerability, in addition to enabling deeper pathway blockade, potentially broadens the repertoire of combinations with other therapeutic agents, such as SERDs, CDK4/6 inhibitors, anti-HER2 therapies, and others.

Structure analyses have linked H1047R-PIK3CA to an altered conformation of the C-terminal tail of p110α (Figure 2B) (19, 20). Subsequent studies leveraged these structure differences to develop allosteric mutant-selective inhibitors (39, 40). Unlike earlier compounds that inhibit both wild-type and mutant PI3Kα with comparable potency, these next-generation molecules are designed to preferentially target oncogenic PIK3CA-mutant isoforms while sparing the wild-type enzyme. Several compounds are included in this category, such as zovegalisib (RLY-2608) (39), tersolisib (STX-478) (40), LOXO-783 (63), and OKI-219 (64). The clinical development of LOXO-783 was halted following the emergence of dose-limiting gastrointestinal toxicity, predominantly diarrhea, raising questions about its degree of selectivity against mutant PI3Ka (63). These findings also reinforce the need for strategies that optimize not only molecular selectivity, but also tissue exposure and dosing schedules.

Currently, zovegalisib is being investigated in combination with fulvestrant with and without CDK4/6 inhibitors in the phase I/II ReDiscover trial. This PIK3CA mutant selective inhibitor recently obtained Breakthrough Therapy Designation by the FDA. Encouraging data were presented at the American Society of Clinical Oncology 2025 meeting, indicating that, in patients with metastatic disease, the median PFS was 11.0 months when they were treated with fulvestrant plus zovegalisib in the second-line setting and 10.3 months in those treated beyond the second-line setting (65). The ongoing ReDiscover-2 Study aims to assess the efficacy and safety of zovegalisib plus fulvestrant versus the pan-AKT inhibitor capivasertib plus fulvestrant in patients with PIK3CA-mutant HR⁺/HER2⁻ breast cancer following progression on or after treatment with CDK4/6 inhibitors (NCT06982521). In parallel, tersolisib is being tested in the phase I/II PIKALO-1 trial (NCT05768139), including a first cohort where the treatment is administered as monotherapy in patients with advanced solid tumors; a second cohort of patients with HR⁺/HER2^– breast cancers is treated with tersolisib plus fulvestrant; and a third cohort is treated with endocrine therapy (aromatase inhibitors, fulvestrant, or imlunestrant) and a CDK4/6 inhibitor. The phase III PIKALO-2 trials will investigate endocrine therapy (aromatase inhibitors or fulvestrant) plus CDK4/6 inhibitors with tersolisib or placebo, both in treatment-naive, endocrine therapy–sensitive as well as pretreated endocrine therapy–resistant patients. The results of these trials are expected to better define the role of next-generation PI3Kα inhibitors within the evolving treatment landscape of HR⁺/HER2⁻ metastatic breast cancer.

The efficacy of targeted therapies is frequently limited by complex adaptive processes occurring both within cancer cells and in their interactions with the tumor microenvironment. Genetic alterations, signaling rewiring, and phenotypic plasticity collectively contribute to resistance and shape treatment sensitivity. A deeper understanding of the mechanisms by which cancer cells evade PI3K inhibition may inform the development of rational combination strategies as well as predictive biomarkers of response.

Downstream PI3K pathway reactivation. Resistance to PI3K inhibitors frequently converges on restoration of PI3K pathway signaling. Among these molecular alterations, loss of the tumor suppressor PTEN is one of the most frequent events, resulting in derepression and thus hyperactivation of PI3K/AKT signaling (Figure 4A) (66, 67). PI3K signaling in PTEN-null cancers has been shown to depend the p110β isoform (68). Recently, it was reported that PTEN loss promotes PI3Kβ phosphorylation and the subsequent formation of a signaling complex with ephrin receptor A2 (EPHA2) and SRC (69). This complex sustains PI3Kβ phosphorylation and drives oncogenic signaling through p-ERK/c-MYC and p-AKT (69). In this context, the FDA-approved SRC/EPHA2 inhibitor dasatinib was able to suppress tumor progression in mouse models. However, PI3Kβ inhibitors have demonstrated only modest efficacy across PTEN-null tumor models (70, 71), suggesting that pathway dependency cannot be fully explained by catalytic inhibition alone and prompting investigation into alternative mechanisms sustaining PI3Kβ signaling.

Figure 4

Representative mechanisms of adaptation and resistance to PI3K inhibitors. (A) Downstream pathway reactivation. Loss of PTEN causes increased levels of PIP3 and thus hyperactivation of PI3K signaling. As a result, PI3Kβ is phosphorylated/activated by forming a complex with EPHA2 and SRC that, in turn, induces cMYC via ERK. (B) Upstream RTK feedback. Treatment with PI3K inhibitors reduces AKT activity and inhibits FOXO3 phosphorylation in the cytoplasm. Unphosphorylated FOXO3 translocates to the nucleus where it induces transcription of RTK genes, thus favoring RKT-mediated feedback reactivation of PI3K and dampening inhibition of PIP3 formation. (C) Metabolic adaptation. Inhibition of PI3K impairs glucose uptake in host tissues (muscle, adipose tissue, liver) via modulation of AKT2 activity. Reduced glucose uptake by the liver activates gluconeogenesis, further favoring hyperglycemia. This stimulates insulin secretion by the pancreas resulting in activation of INS/IGFR1 receptors in cancer cells, partial reactivation of PI3K signaling, and dampening inhibition of PIP3 formation.

Another mechanism of resistance involves residual mTOR activity sustained by pyruvate dehydrogenase kinase 1 (PDK1) and serum/glucocorticoid-regulated kinase 1 (SGK1). SGK1, a downstream effector activated by PDK1, sustains residual mTORC1 activity by directly phosphorylating and inhibiting TSC2 (Figure 1A). This residual mTORC1 activity is sufficient to weaken the antitumor activity of PI3Kα inhibition. Thus, simultaneous suppression of both AKT and PDK1 is required to achieve complete mTORC1 inhibition and antitumor activity in tumors resistant to PI3Kα antagonists (72).

RTK feedback and ERα reactivation. Upon PI3K inhibition, reduced AKT kinase activity leads to decreased phosphorylation of FOXO3A, allowing its translocation to the nucleus (Figure 4B). In turn, nuclear FOXO3A can modulate transcriptional programs promoting adaptive survival and pathway rewiring, such as HER3 and other RTKs (73, 74), as well as expression of ERα and ERα target genes (75). In HER2-overexpressing tumors, PI3K inhibition triggers compensatory upregulation of HER3, which promotes the formation of HER2-HER3 heterodimers and enhances HER3 phosphorylation (73). Phosphorylated HER3 provides multiple docking sites for p85 resulting in partial restoration of PI3K pathway activity (76). Other RTKs, including IGF1R and INSR, are upregulated following PI3K inhibition, reflecting relief of negative feedback loops that normally restrain upstream signaling. Another mechanism of adaptive resistance involves ERα signaling (Figure 1B). PI3K inhibition is associated with increased ERa transcription (50) mediated by enhanced activity of KMT2D, a histone H3 lysine 4 methyltransferase (77). Toska and colleagues showed that, under basal conditions, AKT1 binds and phosphorylates KMT2D, attenuating its methyltransferase activity and ERα function, whereas PI3Kα inhibition enhances KMT2D activity resulting in an open chromatin state at ERa target loci (77).

Metabolic adaptation. Under physiological conditions, insulin and IGF-1 receptor signaling rely predominantly on the p110α catalytic subunit of PI3K to generate PIP3 and activate AKT-mediated glucose uptake and survival pathways. Consequently, pharmacologic inhibition of PI3Kα not only suppresses tumor-intrinsic signaling but also disrupts glucose homeostasis by impairing glucose uptake in insulin-responsive tissues, such as liver, muscle, and adipose tissue, and promoting hepatic gluconeogenesis, ultimately causing hyperglycemia (Figure 4C). This metabolic perturbation stimulates compensatory insulin secretion from pancreatic β cells that, in turn, reactivates INSR and IGF1R, which are ubiquitous in tumor cells, partially restoring PI3K pathway activity through p110α and/or alternative PI3K isoforms. Thus, this adaptive feedback loop attenuates pathway suppression and dampens the antitumor action of PI3Kα inhibitors (78). Moreover, hyperglycemia in patients is a DLT that results in dose reduction and treatment interruption (79), potentially contributing to treatment inefficacy. However, numerous preclinical studies have demonstrated that adaptive resistance can emerge despite near-complete pharmacologic inhibition of PI3K signaling under saturating experimental conditions (70, 78, 80). Therefore, adaptive rewiring represents an intrinsic response to pathway inhibition that is distinct from, although potentially exacerbated by, DLTs. Unless PI3Kα-mutant selective inhibitors could completely spare wild-type PI3Ka, this adaptive compensatory signaling may also occur in tumors treated with such drugs, impairing their full antitumor effect.

Other compensatory signaling pathways. Resistance to PI3K inhibitors may also happen through activation of compensatory pathways. Indeed, inhibition of PI3K can lead to increased RAS/RAF/MEK/ERK activity in HER2-overexpressing breast cancer cells (81), and vice versa, blocking MAPK can increase phosphorylated AKT levels (82). Crosstalk between the PI3K/AKT/mTOR (PAM) and MAPK pathways happens at multiple points of these networks (Figure 1). RAF is inhibited by AKT or the small GTPase Rheb (83, 84). TSC2, a repressor of mTORC1, is inhibited by ERK (85). In breast cancer cells, these interactions are further complicated by additional MAPK and PI3K crosstalk with ERα (see above). Preclinical studies have addressed the implications of this crosstalk by combining PI3K and MEK inhibitors, demonstrating synergistic effect but also major side effects (82, 86, 87). On a similar note, KRAS activation mediates resistance to PI3K inhibition (88). MAPK signaling mutations occur in approximately 10% of breast cancers, with the most frequent alterations being inactivating mutations in NF1, resulting in loss of the GTP-activating protein neurofibromin and subsequently activation of RAS, and activating mutations in KRAS (89), occurring particularly in endocrine-resistant tumors (90). As the predominant clinical use of PI3K inhibitors is in endocrine therapy–resistant tumors, these data warrant further clinical research with recently approved RAS inhibitors.

Acquired genetic resistance. Longitudinal plasma circulating tumor DNA (ctDNA) analysis from the phase I/II trial of alpelisib plus aromatase inhibitors showed that mutations in the DNA binding domain of ESR1 (e.g., Y537S, D538G), the gene encoding ERα, expanded under treatment with antiestrogen therapy and PI3K inhibitors and that the presence of these mutations at baseline was significantly associated with lack of clinical benefit (67). The use of SERDs, active against ESR1 mutations, may overcome this limitation, as suggested by the ctDNA analysis in the ReDiscover trial using zovegalisib (65). In addition to PIK3CA mutations and PTEN loss, PI3K signaling may be activated through AKT1 mutations, present in approximately 3% of breast cancers. The most common mutation is AKT1 E17K, which results in the protein being constitutively localized at the plasma membrane where it is activated by PDK1 and TORC2, while strongly binding PIP3 (91). As a result, AKT1 E17K is resistant to PI3K inhibitors (92, 93) but highly sensitivity to AKT inhibitors (94).

Although an infrequent event, PI3K activity can also increase via amplification or gain of PIK3CA copy number (95). In HER2⁺ breast cancer cells, high-level amplification of mutant PIK3CA was shown to cause resistance to the pan-class I PI3K inhibitor pictilisib (GDC-0941) (96). Amplified c-Myc has also been associated with resistance to PI3K inhibitors (97–99). A recent study in colorectal cancer cells showed that, upon treatment with PI3K inhibitors, an internal ribosome entry site within the MYC upstream region induces a FOXO3a-dependent transcriptional program that ultimately upregulates MYC, thus causing PI3K inhibitor resistance via enhanced autophagic flux (100).

Using a combination of structural modeling and free energy perturbation simulation, Varkaris and colleagues uncovered secondary mutations in the PIK3CA gene that promoted acquired resistance to PI3Kα orthosteric inhibitors (93). Notably, mutations in the W780 and Q859 positions, located within the PI3Kα ATP-binding site, confer resistance to orthosteric PI3K inhibitors, but this resistance can be overcome by allosteric PI3K inhibitors. The authors also found that some secondary PIK3CA mutations in cis with the primary mutation, e.g., E726K in cis with H1047R may decrease the effect of the allosteric inhibitor zovegalisib (93), while retaining sensitivity to orthosteric inhibitors (alpelisib, inavolisib).

Not all complex PIK3CA genotypes confer resistance. A study from Vasan et al. showed that in 12%–13% of cases, double PIK3CA mutants amplify signaling and enhance oncogenicity but also pathway dependence; such tumors retain exquisite sensitivity to alpelisib and inavolisib compared with single mutants. The double mutants generally include a major mutation in either the helical or kinase domain and a minor mutation in cis (e.g. E453, E726, M1043) (101). By contrast, mutations in trans (e.g. E726K/H1047R) are less sensitive than the mutations in cis, responding similarly to the single mutants (101). This study highlights limitations of the established practice of sequencing only certain variants or exons, which underestimates the frequency of double mutants at <1%, without position information (102).

As the field continues to evolve, the future clinical role of PI3K inhibitors will be determined not only by improvements in their molecular target selectivity, but also by development of drugs against other key components in this signaling pathway and rational combinations that abrogate adaptive mechanisms of resistance.

As the therapeutic landscape in HR⁺ breast cancer becomes more complex, treatment selection and sequencing strategy of PI3K inhibitors will be increasingly challenging. PI3K inhibitor efficacy is limited in tumors with PTEN loss-of-function and activating AKT1 mutations, suggesting that patients whose tumors harbor these alterations require alternative therapeutic strategies. One such strategy is use of pan-AKT inhibitors such as capivasertib, which has demonstrated clinical benefit in tumors harboring activating AKT1 mutations and PTEN loss, supporting its use as a biologically rational option for these molecular subsets as well as tumors relying on wild-type PIK3CA, potentially ahead of PIK3CA mutant–selective inhibitors with a more limited coverage of the PI3K/AKT pathway.

Indeed, AKT small-molecule antagonists have also been developed to target tumors that depend on aberrant PI3K signaling. The ATP-competitive pan-AKT inhibitor capivasertib has received FDA approval, in combination with fulvestrant, for patients with ER⁺/HER2^– breast cancer harboring mutations in PIK3CA, AKT, or PTEN based on results of the CAPItello-291 trial (58, 103). This study showed a longer median PFS in patients treated with fulvestrant plus capivasertib versus fulvestrant plus placebo (7.2 months versus 3.6 months; HR, 0.60) (58). Although capivasertib has similar side effects as alpelisib, e.g., rash and diarrhea, it is associated with a lower incidence of hyperglycemia, which may be attributable to its intermittent dosing schedule (4 days on/3 days off). Further, by blocking a key signaling hub downstream of all PI3K isozymes, AKT inhibitors may suppress a broader range of PI3K-driven signaling than PI3Ka inhibitors. Currently, capivasertib is increasingly used in the second-line setting. The future role of capivasertib in the treatment landscape of tumors with somatic alterations in the PI3K pathway may in part depend on the results of the ongoing ReDiscover-2 trial comparing fulvestrant plus capivasertib versus fulvestrant plus zovegalisib (NCT06982521).

Recently, the therapeutic landscape has further expanded with the development of the PAM pathway inhibitors, agents capable of targeting multiple alterations in the pathway — including nodes that mediate resistance to more selective PI3K inhibitors. The small-molecule PAM inhibitor gedatolisib was shown to be more effective than “single-node” PAM inhibitors, such as alpelisib, capivasertib, and the mTORC1 inhibitor everolimus, in tumor cells with either mutated or wild-type PAM signaling (104). Gedatolisib was associated with manageable side effects, possibly because of the reported lower IC₅₀ of the compound in cancer cells compared with nontumor cells and pulsatile pathway inhibition as a result of weekly i.v. administration, potentially decoupling efficacy from metabolic toxicity through this dosing strategy rather than molecular selectivity (104). Gedatolisib was tested in combination with endocrine therapy with or without palbociclib in a phase I/Ib trial in patients with or without PI3KCA mutations, where it was associated with a reduced rate of hyperglycemia compared with other single-node PAM inhibitors (105). The phase III trial VIKTORIA-1 evaluated the efficacy of the triplet gedatolisib/palbociclib/fulvestrant and the doublet gedatolisib/fulvestrant versus fulvestrant alone, in patients with wild-type PIK3CA (study 1) or versus fulvestrant/alpelisib in patients with PIK3CA-mutant cancers (study 2). Enrolled patients had HR⁺/HER2^– advanced breast cancer, which had progressed during or after aromatase inhibitors plus CDK4/6 inhibitors (105). While the results for the PIK3CA-mutant cohort are forthcoming, the results from study 1 were reported (106), and the median PFS for the triple combination was 9.3 months versus 2.0 months in the fulvestrant monotherapy group, while fulvestrant plus gedatolisib median PFS was 7.4 months. The reduced risk of death in patients treated with the triplet combination versus single agent fulvestrant is so far the largest difference reported in this patient population (106).

If the results in the PIK3CA-mutant cohort in the VIKTORIA study are positive, multiple implications arise for future drug development. First, PAM inhibitors seem to work in absence of PIK3CA mutations. This finding is consistent with the BOLERO-2 trial results, where everolimus plus exemestane was active independently of presence of any mutation in the PI3K pathway (107). On the other hand, gedatolisib was shown to be less effective against PTEN-null cancers in preclinical studies (108), but these cases were not excluded from the VIKTORIA trials. In any case, the identification of other compounds that show clinical activity irrespective of PI3K/AKT pathway mutation status would imply the need for better therapeutically predictive biomarkers of pathway dependence and thus sensitivity to pathway-targeted inhibitors (109, 110). Indeed, integration of genomic, transcriptomic, epigenomic, and proteomic data generated through large-scale initiatives, such as The Cancer Genome Atlas and Clinical Proteomic Tumor Analysis Consortium, has shown that mutational status alone does not always correspond to aberrant PI3K pathway activation, highlighting the complementary value of phosphoproteomic and transcriptomic profiling for identifying tumors with functional PI3K dependence (109–113).

Another emerging therapeutic strategy involves disrupting oncogenic protein-protein interactions, which are increasingly recognized as druggable therapeutic targets (114, 115). One such example is the interaction between PI3Kα and RAS family members through the RBD. Compounds designed to disrupt this interaction have been referred to as “PI3K-KRAS breakers.” Mutations such as T208D and K227A in the PI3Kα RBD disrupt the interaction between PI3K and KRAS (116). Based on this observation, a new compound that phenocopies the effect of the PI3Kα T208D-K227A mutant, BBO-10203, has been developed (117). BBO-10203 is specific to PI3Kα and exhibits no affinity toward PI3Kβ, PI3Kδ, and PI3Kγ. BBO-10203 is effective in decreasing AKT phosphorylation in PIK3CA/KRAS-mutant cells, while PTEN-deficient models appear largely unresponsive to the treatment. HER2-amplified tumors showed a greater-than-expected sensitivity to BBO-10203, although the underlying mechanism remains unclear. In HER2-amplified tumors, PI3Kα activation is not strictly dependent on canonical RAS activity (118), suggesting that this sensitivity may be mediated through effects on noncanonical RAS-related proteins, including RRAS, RRAS2, and MRAS (117). Further investigation of this mechanism is warranted; nevertheless, the observed activity in HER2-amplified models suggests a potential therapeutic role for this compound in this subset of patients. Notably, these antitumor effects were achieved without the development of hyperglycemia, highlighting a unique therapeutic property, as insulin signaling mediates glucose uptake through mechanisms that are independent of RAS-driven PI3Kα activation.

Identification of effective pathway inhibitors has important implications for optimizing treatment sequencing. PIK3CA mutations associated with resistance to the orthosteric inhibitor alpelisib do not appear to impair sensitivity to the allosteric inhibitor zovegalisib (93). AKT1-E17K mutations have been found in patients resistant to either allosteric or orthosteric PI3K inhibitors (92, 93), but those patients are sensitive to AKT inhibitors (94). In this context, PAM pathway inhibitors are likely to be of particular interest. More broadly, a deeper understanding of resistance mechanisms may inform the development of rational combination strategies and improve patient selection for clinical trials. A clear example of this approach is the exclusion of patients with PTEN-null tumors from trials evaluating PI3Ka inhibitors. Whether patients harboring tumors with activating RAS mutations should also be excluded is not clear. Achieving these advances will require well-designed ancillary studies to generate the necessary biological and clinical insights, ultimately enabling more durable and profound responses while carefully balancing treatment-related toxicities.

Another promising area for PI3K inhibition is the neoadjuvant setting. As PIK3CA mutations are truncal events present from diagnosis, targeting this pathway early in the disease course is biologically compelling. A previous study adding alpelisib to endocrine therapy did not improve objective response rates, pathological complete response rates, or Ki-67 suppression compared with endocrine therapy alone, suggesting that the impact of PI3K inhibition may differ between early and metastatic disease (119). However, the results of simultaneous blockade of ERa, CDK4/6, and PI3K in the INAVO120 trial suggest that early use of this combination in some patients with high-risk disease is worthy of clinical testing. In this context, the role of inavolisib in combination with letrozole and palbociclib in endocrine therapy–sensitive, PIK3CA-mutant breast cancer is being investigated in the INAVO123 trial (NCT06790693). In the future, replacing inavolisib with a PIK3CA-mutant selective inhibitor may improve the tolerability and efficacy of this combination.

PIK3CA mutations are highly prevalent in lobular breast cancers, often at higher frequency than in ductal carcinoma (120, 121). While no PI3K inhibitor trial has been specifically designed for this subtype, there is increasing interest in targeting the PI3K pathway in studies enriched for this population. In this context, an ongoing trial (NCT06607757) is evaluating the AKT inhibitor capivasertib in combination with fulvestrant versus fulvestrant alone in primary high-risk lobular breast cancer.

Finally, in addition to patient selection defined by molecular criteria, it is important to consider the differences in trial eligibility criteria, especially in terms of HbA1c thresholds and acceptable glycemic control. Earlier trials of alpelisib, such as SOLAR-1 (52), largely excluded patients with diabetes or prediabetes, whereas more recent studies of PI3K pathway inhibitors have adopted more permissive eligibility criteria, allowing patients with well-controlled diabetes. In particular, PIKALO-1 (investigating tersolisib/STX-478) is the most permissive trial to date, with 54% of patients enrolled with baseline prediabetes or diabetes, providing a more clinically representative assessment of hyperglycemia risk (122). Despite this broader eligibility, metabolic toxicity remained limited: among patients with prediabetes or diabetes, rates were 37% for any-grade hyperglycemia and only 1% for grade ≥3 events. No grade 3 hyperglycemia occurred among patients with normal baseline glycemic parameters. Conversely, patients with type 1 diabetes or uncontrolled type 2 diabetes are excluded from the VIKTORIA-1 and ReDiscover2 trials and, in the latter, even patients with type 2 diabetes requiring medication are excluded.

In conclusion, PI3K/AKT pathway inhibition is a key therapeutic strategy in breast cancer, with drug development still evolving across clinical disease contexts and molecular subgroups. Ongoing clinical and translational efforts will be crucial to improve patient selection, overcome resistance mechanisms, and maximize clinical benefits. In particular, enhancing biomarker-driven approaches and integrating novel combinations will be central to unlocking its full therapeutic potential.

NU declares personal fees from Pfizer, Lilly, and Gilead Sciences. CLA reports grants from Pfizer and Eli Lilly and scientific expert advisory activities for Lilly and Pfizer.

This work is the result of NIH funding, in whole or in part, and is subject to the NIH Public Access Policy. Through acceptance of this federal funding, the NIH has been given a right to make the work publicly available in PubMed Central.

• National Cancer Institute grants P30 CA142543 and R01 CA273246.
• Breast Cancer Research Foundation.

The authors would like to acknowledge the American Association for Cancer Research and its financial and material support in the development of the AACR Project GENIE. Interpretations are the responsibility of study authors.

Address correspondence to: Carlos L. Arteaga, MD, Simmons Comprehensive Cancer Center, UT Southwestern Medical Center, 5323 Harry Hines Blvd., Dallas, TX 75390-8590, USA. Email: carlos.arteaga@utsouthwestern.edu.

Copyright: © 2026, Napolitano et al. This is an open access article published under the terms of the Creative Commons Attribution 4.0 International License.

Reference information: J Clin Invest. 2026;136(20):e207635. https://doi.org/10.1172/JCI207635.

1. Millis SZ, et al. Landscape of phosphatidylinositol-3-kinase pathway alterations across 19784 diverse solid tumors. JAMA Oncol. 2016;2(12):1565–1573.
   View this article via: CrossRef PubMed Google Scholar
2. Wiese W, et al. PI3K/Akt/mTOR signaling pathway in blood malignancies-new therapeutic possibilities. Cancers (Basel). 2023;15(21):5297.
   View this article via: CrossRef PubMed Google Scholar
3. Belli C, et al. The emerging role of PI3K inhibitors for solid tumour treatment and beyond. Br J Cancer. 2023;128(12):2150–2162.
   View this article via: CrossRef PubMed Google Scholar
4. Canaud G, et al. A review of mechanisms of disease across PIK3CA-related disorders with vascular manifestations. Orphanet J Rare Dis. 2021;16(1):306.
   View this article via: CrossRef PubMed Google Scholar
5. The AACR Project GENIE Consortium, et al. AACR Project GENIE: powering precision medicine through an international consortium. Cancer Discov. 2017;7(8):818–831.
   View this article via: CrossRef PubMed Google Scholar
6. Hanker AB, et al. Challenges for the clinical development of PI3K inhibitors: strategies to improve their impact in solid tumors. Cancer Discov. 2019;9(4):482–491.
   View this article via: CrossRef PubMed Google Scholar
7. Goncalves MD, et al. Phosphatidylinositol 3-kinase, growth disorders, and cancer. N Engl J Med. 2018;379(21):2052–2062.
   View this article via: CrossRef PubMed Google Scholar
8. Keppler-Noreuil KM, et al. PIK3CA-related overgrowth spectrum (PROS): diagnostic and testing eligibility criteria, differential diagnosis, and evaluation. Am J Med Genet A. 2015;167A(2):287–295.
   View this article via: CrossRef PubMed Google Scholar
9. Kuentz P, et al. Molecular diagnosis of PIK3CA-related overgrowth spectrum (PROS) in 162 patients and recommendations for genetic testing. Genet Med. 2017;19(9):989–997.
   View this article via: CrossRef PubMed Google Scholar
10. Yehia L, et al. Extended spectrum of cancers in PTEN hamartoma tumor syndrome. NPJ Precis Oncol. 2025;9(1):61.
    View this article via: CrossRef PubMed Google Scholar
11. O’Leary B, et al. The genetic landscape and clonal evolution of breast cancer resistance to palbociclib plus fulvestrant in the PALOMA-3 Trial. Cancer Discov. 2018;8(11):1390–1403.
    View this article via: CrossRef PubMed Google Scholar
12. Basho RK, et al. Comprehensive analysis identifies variability in PI3K pathway alterations in triple-negative breast cancer subtypes. JCO Precis Oncol. 2024;8(8):e2300124.
    View this article via: CrossRef PubMed Google Scholar
13. Majewski IJ, et al. PIK3CA mutations are associated with decreased benefit to neoadjuvant human epidermal growth factor receptor 2-targeted therapies in breast cancer. J Clin Oncol. 2015;33(12):1334–1339.
    View this article via: CrossRef PubMed Google Scholar
14. Loibl S, et al. PIK3CA mutations are associated with reduced pathological complete response rates in primary HER2-positive breast cancer: pooled analysis of 967 patients from five prospective trials investigating lapatinib and trastuzumab. Ann Oncol. 2016;27(8):1519–1525.
    View this article via: CrossRef PubMed Google Scholar
15. Zagami P, et al. Association of PIK3CA mutation with pathologic complete response and outcome by hormone receptor status and intrinsic subtype in early-stage ERBB2/HER2-positive breast cancer. JAMA Netw Open. 2023;6(12):e2348814.
    View this article via: CrossRef PubMed Google Scholar
16. Miller TW, et al. Phosphatidylinositol 3-kinase and antiestrogen resistance in breast cancer. J Clin Oncol. 2011;29(33):4452–4461.
    View this article via: CrossRef PubMed Google Scholar
17. Koboldt DC, et al. Comprehensive molecular portraits of human breast tumours. Nature. 2012;490(7418):61–70.
    View this article via: CrossRef PubMed Google Scholar
18. Mandelker D, et al. A frequent kinase domain mutation that changes the interaction between PI3Kalpha and the membrane. Proc Natl Acad Sci U S A. 2009;106(40):16996–17001.
    View this article via: CrossRef PubMed Google Scholar
19. Liu X, et al. Cryo-EM structures of cancer-specific helical and kinase domain mutations of PI3Kα. Proc Natl Acad Sci U S A. 2022;119(46):e2215621119.
    View this article via: CrossRef PubMed Google Scholar
20. Hart JR, et al. Nanobodies and chemical cross-links advance the structural and functional analysis of PI3Kα. Proc Natl Acad Sci U S A. 2022;119(38):e2210769119.
    View this article via: CrossRef PubMed Google Scholar
21. Engelman JA, et al. The evolution of phosphatidylinositol 3-kinases as regulators of growth and metabolism. Nat Rev Genet. 2006;7(8):606–619.
    View this article via: CrossRef PubMed Google Scholar
22. Thorpe LM, et al. PI3K in cancer: divergent roles of isoforms, modes of activation and therapeutic targeting. Nat Rev Cancer. 2015;15(1):7–24.
    View this article via: CrossRef PubMed Google Scholar
23. Park WS, et al. Comprehensive identification of PIP3-regulated PH domains from C. elegans to H. sapiens by model prediction and live imaging. Mol Cell. 2008;30(3):381–392.
    View this article via: CrossRef PubMed Google Scholar
24. Sarbassov DD, et al. Phosphorylation and regulation of Akt/PKB by the rictor-mTOR complex. Science. 2005;307(5712):1098–1101.
    View this article via: CrossRef PubMed Google Scholar
25. Manning BD, Cantley LC. AKT/PKB signaling: navigating downstream. Cell. 2007;129(7):1261–1274.
    View this article via: CrossRef PubMed Google Scholar
26. Blakely B, et al. Overview of the therapeutic strategies for ER positive breast cancer. Biochem Pharmacol. 2023;212:115552.
    View this article via: CrossRef PubMed Google Scholar
27. Simoncini T, et al. Interaction of oestrogen receptor with the regulatory subunit of phosphatidylinositol-3-OH kinase. Nature. 2000;407(6803):538–541.
    View this article via: CrossRef PubMed Google Scholar
28. Yamnik RL, et al. S6 kinase 1 regulates estrogen receptor alpha in control of breast cancer cell proliferation. J Biol Chem. 2009;284(10):6361–6369.
    View this article via: CrossRef PubMed Google Scholar
29. Campbell RA, et al. Phosphatidylinositol 3-kinase/AKT-mediated activation of estrogen receptor alpha: a new model for anti-estrogen resistance. J Biol Chem. 2001;276(13):9817–9824.
    View this article via: CrossRef PubMed Google Scholar
30. Gustin JP, et al. Knockin of mutant PIK3CA activates multiple oncogenic pathways. Proc Natl Acad Sci U S A. 2009;106(8):2835–2840.
    View this article via: CrossRef PubMed Google Scholar
31. Huang CH, et al. The structure of a human p110alpha/p85alpha complex elucidates the effects of oncogenic PI3Kalpha mutations. Science. 2007;318(5857):1744–1748.
    View this article via: CrossRef PubMed Google Scholar
32. Samuels Y, et al. High frequency of mutations of the PIK3CA gene in human cancers. Science. 2004;304(5670):554.
    View this article via: CrossRef PubMed Google Scholar
33. Glaviano A, et al. PI3K/AKT/mTOR signaling transduction pathway and targeted therapies in cancer. Mol Cancer. 2023;22(1):138.
    View this article via: PubMed CrossRef Google Scholar
34. Wylaź M, et al. Exploring the role of PI3K/AKT/mTOR inhibitors in hormone-related cancers: A focus on breast and prostate cancer. Biomed Pharmacother. 2023;168:115676.
    View this article via: CrossRef PubMed Google Scholar
35. Martínez-Sáez O, et al. Frequency and spectrum of PIK3CA somatic mutations in breast cancer. Breast Cancer Res. 2020;22(1):45.
    View this article via: CrossRef PubMed Google Scholar
36. Song KW, et al. RTK-dependent inducible degradation of mutant PI3Kα drives GDC-0077 (Inavolisib) efficacy. Cancer Discov. 2022;12(1):204–219.
    View this article via: CrossRef PubMed Google Scholar
37. Jenkins ML, et al. Oncogenic mutations of PIK3CA lead to increased membrane recruitment driven by reorientation of the ABD, p85 and C-terminus. Nat Commun. 2023;14(1):181.
    View this article via: CrossRef PubMed Google Scholar
38. Hon WC, et al. Regulation of lipid binding underlies the activation mechanism of class IA PI3-kinases. Oncogene. 2012;31(32):3655–3666.
    View this article via: CrossRef PubMed Google Scholar
39. Varkaris A, et al. Discovery and clinical proof-of-concept of RLY-2608, a first-in-class mutant-selective allosteric PI3Kα inhibitor that decouples antitumor activity from hyperinsulinemia. Cancer Discov. 2024;14(2):240–257.
    View this article via: CrossRef PubMed Google Scholar
40. 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.
    View this article via: CrossRef PubMed Google Scholar
41. Jiang W, Ji M. Receptor tyrosine kinases in PI3K signaling: The therapeutic targets in cancer. Semin Cancer Biol. 2019;59:3–22.
    View this article via: CrossRef PubMed Google Scholar
42. Lee YR, et al. The functions and regulation of the PTEN tumour suppressor: new modes and prospects. Nat Rev Mol Cell Biol. 2018;19(9):547–562.
    View this article via: CrossRef PubMed Google Scholar
43. Gymnopoulos M, et al. Rare cancer-specific mutations in PIK3CA show gain of function. Proc Natl Acad Sci U S A. 2007;104(13):5569–5574.
    View this article via: CrossRef PubMed Google Scholar
44. Croessmann S, et al. PIK3CA C2 domain deletions hyperactivate phosphoinositide 3-kinase (PI3K), generate oncogene dependence, and are exquisitely sensitive to PI3Kα inhibitors. Clin Cancer Res. 2018;24(6):1426–1435.
    View this article via: CrossRef PubMed Google Scholar
45. Liu Y, et al. Pan-cancer analysis on the role of PIK3R1 and PIK3R2 in human tumors. Sci Rep. 2022;12(1):5924.
    View this article via: CrossRef PubMed Google Scholar
46. Jhaveri KL, et al. Clinical management of common toxicities with inhibitors targeting the PI3K/AKT/mTOR pathway in breast cancer. ESMO Open. 2026;11(2):105936.
    View this article via: CrossRef PubMed Google Scholar
47. Bendell JC, et al. Phase I, dose-escalation study of BKM120, an oral pan-Class I PI3K inhibitor, in patients with advanced solid tumors. J Clin Oncol. 2012;30(3):282–290.
    View this article via: CrossRef PubMed Google Scholar
48. Baselga J, et al. Buparlisib plus fulvestrant versus placebo plus fulvestrant in postmenopausal, hormone receptor-positive, HER2-negative, advanced breast cancer (BELLE-2): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Oncol. 2017;18(7):904–916.
    View this article via: CrossRef PubMed Google Scholar
49. Miller TW, et al. ERα-dependent E2F transcription can mediate resistance to estrogen deprivation in human breast cancer. Cancer Discov. 2011;1(4):338–351.
    View this article via: CrossRef PubMed Google Scholar
50. Bosch A, et al. PI3K inhibition results in enhanced estrogen receptor function and dependence in hormone receptor-positive breast cancer. Sci Transl Med. 2015;7(283):283ra51.
    View this article via: CrossRef PubMed Google Scholar
51. Di Leo A, et al. Buparlisib plus fulvestrant in postmenopausal women with hormone-receptor-positive, HER2-negative, advanced breast cancer progressing on or after mTOR inhibition (BELLE-3): a randomised, double-blind, placebo-controlled, phase 3 trial. Lancet Oncol. 2018;19(1):87–100.
    View this article via: CrossRef PubMed Google Scholar
52. André F, et al. Alpelisib plus fulvestrant for PIK3CA-mutated, hormone receptor-positive, human epidermal growth factor receptor-2-negative advanced breast cancer: final overall survival results from SOLAR-1. Ann Oncol. 2021;32(2):208–217.
    View this article via: CrossRef PubMed Google Scholar
53. Hanan EJ, et al. Discovery of GDC-0077 (Inavolisib), a highly selective inhibitor and degrader of mutant PI3Kα. J Med Chem. 2022;65(24):16589–16621.
    View this article via: CrossRef PubMed Google Scholar
54. Dent S, et al. Phase III randomized study of taselisib or placebo with fulvestrant in estrogen receptor-positive, PIK3CA-mutant, HER2-negative, advanced breast cancer: the SANDPIPER trial. Ann Oncol. 2021;32(2):197–207.
    View this article via: CrossRef PubMed Google Scholar
55. André F, et al. Alpelisib for PIK3CA-mutated, hormone receptor–positive advanced breast cancer. N Engl J Med. 2019;380(20):1929–1940.
    View this article via: CrossRef PubMed Google Scholar
56. Rugo HS, et al. Alpelisib plus fulvestrant in PIK3CA-mutated, hormone receptor-positive advanced breast cancer after a CDK4/6 inhibitor (BYLieve): one cohort of a phase 2, multicentre, open-label, non-comparative study. Lancet Oncol. 2024;25(12):e629–e638.
    View this article via: CrossRef PubMed Google Scholar
57. De Laurentiis M, et al. Abstract RF7-02: Alpelisib plus fulvestrant for PIK3CA-mutated, HR-positive, HER2-negative advanced breast cancer after a CDK4/6 inhibitor (EPIK-B5): Phase III, randomized, double-blind, placebo-controlled, multicenter study. Clin Cancer Res. 2026;32(4_suppl):RF7.
    View this article via: CrossRef Google Scholar
58. Turner NC, et al. Capivasertib in hormone receptor-positive advanced breast cancer. N Engl J Med. 2023;388(22):2058–2070.
    View this article via: CrossRef PubMed Google Scholar
59. Luboff AJ, DeRemer DL. Capivasertib: a novel AKT inhibitor approved for hormone-receptor-positive, HER-2-negative metastatic breast cancer. Ann Pharmacother. 2024;58(12):1229–1237.
    View this article via: CrossRef PubMed Google Scholar
60. Vora SR, et al. CDK 4/6 inhibitors sensitize PIK3CA mutant breast cancer to PI3K inhibitors. Cancer Cell. 2014;26(1):136–149.
    View this article via: CrossRef PubMed Google Scholar
61. Herrera-Abreu MT, et al. Early adaptation and acquired resistance to CDK4/6 inhibition in estrogen receptor-positive breast cancer. Cancer Res. 2016;76(8):2301–2313.
    View this article via: CrossRef PubMed Google Scholar
62. Turner NC, et al. Inavolisib-based therapy in PIK3CA-mutated advanced breast cancer. N Engl J Med. 2024;391(17):1584–1596.
    View this article via: CrossRef PubMed Google Scholar
63. Jhaveri K, et al. Abstract PS7-03: A first-in-human phase 1a/b trial of LOXO-783, a potent, highly mutant-selective, brain-penetrant, allosteric PI3Kα H1047R inhibitor in PIK3CA H1047R-mutant advanced breast cancer and other solid tumors: Results from the PIKASSO-01 study. Clin Cancer Res. 2025;31(12_suppl):PS7.
    View this article via: CrossRef Google Scholar
64. Taylor M, et al. Abstract P4-12-20: OKI-219 enhances activity of SOC therapies in double and triple combinations in pre-clinical PI3Kα H1047R mutant breast cancer models. Clin Cancer Res. 2025;31(12_suppl):4–12.
    View this article via: CrossRef Google Scholar
65. Sammons SL, et al. Updated efficacy of mutant-selective PI3Kα inhibitor RLY-2608 in combination with fulvestrant in patients with PIK3CA-mutant HR+HER2- advanced breast cancer: ReDiscover trial. J Clin Oncol. 2025;43(16_suppl):1086–1086.
    View this article via: CrossRef Google Scholar
66. Juric D, et al. Convergent loss of PTEN leads to clinical resistance to a PI(3)Kα inhibitor. Nature. 2015;518(7538):240–244.
    View this article via: CrossRef PubMed Google Scholar
67. Razavi P, et al. Alterations in PTEN and ESR1 promote clinical resistance to alpelisib plus aromatase inhibitors. Nat Cancer. 2020;1(4):382–393.
    View this article via: CrossRef PubMed Google Scholar
68. Wee S, et al. PTEN-deficient cancers depend on PIK3CB. Proc Natl Acad Sci U S A. 2008;105(35):13057–13062.
    View this article via: CrossRef PubMed Google Scholar
69. Tang S, et al. PTEN loss promotes PI3Kβ phosphorylation and EPHA2/SRC/p-PI3KβY962 complex assembly to drive tumorigenesis. Cancer Discov. 2026;16(3):521–540.
    View this article via: CrossRef PubMed Google Scholar
70. Lynch JT, et al. Combined inhibition of PI3Kβ and mTOR inhibits growth of PTEN-null tumors. Mol Cancer Ther. 2018;17(11):2309–2319.
    View this article via: CrossRef PubMed Google Scholar
71. Schwartz S, et al. Feedback suppression of PI3Kα signaling in PTEN-mutated tumors is relieved by selective inhibition of PI3Kβ. Cancer Cell. 2015;27(1):109–122.
    View this article via: CrossRef PubMed Google Scholar
72. Castel P, et al. PDK1-SGK1 signaling sustains AKT-Independent mTORC1 activation and confers resistance to PI3Kα inhibition. Cancer Cell. 2016;30(2):229–242.
    View this article via: CrossRef PubMed Google Scholar
73. Chandarlapaty S, et al. AKT inhibition relieves feedback suppression of receptor tyrosine kinase expression and activity. Cancer Cell. 2011;19(1):58–71.
    View this article via: CrossRef PubMed Google Scholar
74. Chakrabarty A, et al. Feedback upregulation of HER3 (ErbB3) expression and activity attenuates antitumor effect of PI3K inhibitors. Proc Natl Acad Sci U S A. 2012;109(8):2718–2723.
    View this article via: CrossRef PubMed Google Scholar
75. Guo S, Sonenshein GE. Forkhead box transcription factor FOXO3a regulates estrogen receptor alpha expression and is repressed by the Her-2/neu/phosphatidylinositol 3-kinase/Akt signaling pathway. Mol Cell Biol. 2004;24(19):8681–8690.
    View this article via: CrossRef PubMed Google Scholar
76. Soltoff TP, et al. ErbB3 is involved in activation of phosphatidylinositol 3-kinase by epidermal growth factor. Mol Cell Biol. 1994;14(6):3550–3558.
    View this article via: CrossRef PubMed Google Scholar
77. Toska E, et al. PI3K pathway regulates ER-dependent transcription in breast cancer through the epigenetic regulator KMT2D. Science. 2017;355(6331):1324–1330.
    View this article via: CrossRef PubMed Google Scholar
78. Hopkins BD, et al. Suppression of insulin feedback enhances the efficacy of PI3K inhibitors. Nature. 2018;560(7719):499–503.
    View this article via: CrossRef PubMed Google Scholar
79. Rugo HS, 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 this article via: CrossRef PubMed Google Scholar
80. Napolitano F, et al. Abstract 3003: Targeting mechanisms of adaptive resistance to the PI3Kαmutant selective inhibitor RLY-2608 in HR+/PIK3CA mutant breast cancer. Cancer Res. 3003;85(8_suppl_1):3003.
    View this article via: CrossRef Google Scholar
81. Serra V, et al. PI3K inhibition results in enhanced HER signaling and acquired ERK dependency in HER2-overexpressing breast cancer. Oncogene. 2011;30(22):2547–2557.
    View this article via: CrossRef PubMed Google Scholar
82. Hoeflich KP, et al. In vivo antitumor activity of MEK and phosphatidylinositol 3-kinase inhibitors in basal-like breast cancer models. Clin Cancer Res. 2009;15(14):4649–4664.
    View this article via: CrossRef PubMed Google Scholar
83. Zimmermann S, Moelling K. Phosphorylation and regulation of Raf by Akt (protein kinase B). Science. 1999;286(5445):1741–1744.
    View this article via: CrossRef PubMed Google Scholar
84. Karbowniczek M, et al. Rheb inhibits C-raf activity and B-raf/C-raf heterodimerization. J Biol Chem. 2006;281(35):25447–25456.
    View this article via: CrossRef PubMed Google Scholar
85. Ma L, et al. Phosphorylation and functional inactivation of TSC2 by Erk implications for tuberous sclerosis and cancer pathogenesis. Cell. 2005;121(2):179–193.
    View this article via: CrossRef PubMed Google Scholar
86. Haagensen EJ, et al. The synergistic interaction of MEK and PI3K inhibitors is modulated by mTOR inhibition. Br J Cancer. 2012;106(8):1386–1394.
    View this article via: CrossRef PubMed Google Scholar
87. Mirzoeva OK, et al. Basal subtype and MAPK/ERK kinase (MEK)-phosphoinositide 3-kinase feedback signaling determine susceptibility of breast cancer cells to MEK inhibition. Cancer Res. 2009;69(2):565–572.
    View this article via: CrossRef PubMed Google Scholar
88. Ihle NT, et al. Mutations in the phosphatidylinositol-3-kinase pathway predict for antitumor activity of the inhibitor PX-866 whereas oncogenic Ras is a dominant predictor for resistance. Cancer Res. 2008;69(1):143–150.
    View this article via: CrossRef PubMed Google Scholar
89. Rocca A, et al. The predictive and prognostic role of RAS-RAF-MEK-ERK pathway alterations in breast cancer: revision of the literature and comparison with the analysis of cancer genomic datasets. Cancers (Basel). 2022;14(21):5306.
    View this article via: CrossRef PubMed Google Scholar
90. Razavi P, et al. The genomic landscape of endocrine-resistant advanced breast cancers. Cancer Cell. 2018;34(3):427–438.
    View this article via: CrossRef PubMed Google Scholar
91. Carpten JD, et al. A transforming mutation in the pleckstrin homology domain of AKT1 in cancer. Nature. 2007;448(7152):439–444.
    View this article via: CrossRef PubMed Google Scholar
92. Beaver JA, et al. PIK3CA and AKT1 mutations have distinct effects on sensitivity to targeted pathway inhibitors in an isogenic luminal breast cancer model system. Clin Cancer Res. 2013;19(19):5413–5422.
    View this article via: CrossRef PubMed Google Scholar
93. Varkaris A, et al. Allosteric PI3Kα inhibition overcomes on-target resistance to orthosteric inhibitors mediated by secondary PIK3CA mutations. Cancer Discov. 2024;14(2):227–239.
    View this article via: CrossRef PubMed Google Scholar
94. Davies BR, et al. Tumors with AKT1E17K mutations are rational targets for single agent or combination therapy with AKT inhibitors. Mol Cancer Ther. 2015;14(11):2441–2451.
    View this article via: CrossRef PubMed Google Scholar
95. Gonzalez-Angulo AM, et al. Frequency of mesenchymal-epithelial transition factor gene (MET) and the catalytic subunit of phosphoinositide-3-kinase (PIK3CA) copy number elevation and correlation with outcome in patients with early stage breast cancer. Cancer. 2013;119(1):7–15.
    View this article via: CrossRef PubMed Google Scholar
96. Huw LY, et al. Acquired PIK3CA amplification causes resistance to selective phosphoinositide 3-kinase inhibitors in breast cancer. Oncogenesis. 2013;2(12):e83.
    View this article via: CrossRef PubMed Google Scholar
97. Liu P, et al. Oncogenic PIK3CA-driven mammary tumors frequently recur via PI3K pathway-dependent and PI3K pathway-independent mechanisms. Nat Med. 2011;17(9):1116–1120.
    View this article via: CrossRef PubMed Google Scholar
98. Muellner MK, et al. A chemical-genetic screen reveals a mechanism of resistance to PI3K inhibitors in cancer. Nat Chem Biol. 2011;7(11):787–793.
    View this article via: CrossRef PubMed Google Scholar
99. Ilic N, et al. PI3K-targeted therapy can be evaded by gene amplification along the MYC-eukaryotic translation initiation factor 4E (eIF4E) axis. Proc Natl Acad Sci U S A. 2011;108(37):E699–E708.
    View this article via: CrossRef PubMed Google Scholar
100. Bordone R, et al. MYC upstream region orchestrates resistance to PI3K inhibitors in cancer cells through FOXO3a-mediated autophagic adaptation. Oncogene. 2024;43(46):3349–3365.
     View this article via: CrossRef PubMed Google Scholar
101. Vasan N, et al. Double PIK3CA mutations in cis increase oncogenicity and sensitivity to PI3Kα inhibitors. Science. 2019;366(6466):714–723.
     View this article via: CrossRef PubMed Google Scholar
102. Saal LH, et al. PIK3CA mutations correlate with hormone receptors, node metastasis, and ERBB2, and are mutually exclusive with PTEN loss in human breast carcinoma. Cancer Res. 2005;65(7):2554–2559.
     View this article via: CrossRef PubMed Google Scholar
103. Dilawari A, et al. US Food and Drug Administration approval summary: capivasertib with fulvestrant for hormone receptor–positive, human epidermal growth factor receptor 2–negative locally advanced or metastatic breast cancer With PIK3CA/AKT1/PTEN alterations. J Clin Oncol. 2024;42(34):4103–4113.
     View this article via: CrossRef PubMed Google Scholar
104. Rossetti S, et al. Gedatolisib shows superior potency and efficacy versus single-node PI3K/AKT/mTOR inhibitors in breast cancer models. NPJ Breast Cancer. 2024;10(1):40.
     View this article via: CrossRef PubMed Google Scholar
105. Layman RM, et al. Gedatolisib in combination with palbociclib and endocrine therapy in women with hormone receptor-positive, HER2-negative advanced breast cancer: results from the dose expansion groups of an open-label, phase 1b study. Lancet Oncol. 2024;25(4):474–487.
     View this article via: CrossRef PubMed Google Scholar
106. Hurvitz SA, et al. VIKTORIA-1 trial of gedatolisib plus fulvestrant with or without palbociclib in hormone receptor-positive/HER2-/PIK3CA wild-type advanced breast cancer. J Clin Oncol. 2026;44(12):1108–1119.
     View this article via: CrossRef PubMed Google Scholar
107. Baselga J, et al. Everolimus in postmenopausal hormone-receptor–positive advanced breast cancer. N Engl J Med. 2012;366(6):520–529.
     View this article via: CrossRef PubMed Google Scholar
108. Alves CL, et al. Dual PI3K/mTOR inhibition is required to combat resistance to CDK4/6 inhibitor and endocrine therapy in PIK3CA-mutant breast cancer. Sci Transl Med. 2025;17(829):eadp5088.
     View this article via: CrossRef PubMed Google Scholar
109. Kwei KA, et al. Modulators of sensitivity and resistance to inhibition of PI3K identified in a pharmacogenomic screen of the NCI-60 human tumor cell line collection. PLoS One. 2012;7(9):e46518.
     View this article via: CrossRef PubMed Google Scholar
110. Creighton CJ. A gene transcription signature of the Akt/mTOR pathway in clinical breast tumors. Oncogene. 2007;26(32):4648–4655.
     View this article via: CrossRef PubMed Google Scholar
111. Jiang P, et al. Big data in basic and translational cancer research. Nat Rev Cancer. 2022;22(11):625–639.
     View this article via: CrossRef PubMed Google Scholar
112. Alam MS, et al. Integrated bioinformatics and statistical approaches to explore molecular biomarkers for breast cancer diagnosis, prognosis and therapies. PLoS One. 2022;17(5):e0268967.
     View this article via: CrossRef PubMed Google Scholar
113. Coleman WB. Next-generation breast cancer omics. Am J Pathol. 2017;187(10):2130–2132.
     View this article via: CrossRef PubMed Google Scholar
114. Xie X, et al. Recent advances in targeting the “undruggable” proteins: from drug discovery to clinical trials. Signal Transduct Target Ther. 2023;8(1):335.
     View this article via: CrossRef PubMed Google Scholar
115. Scott DE, et al. Small molecules, big targets: drug discovery faces the protein–protein interaction challenge. Nat Rev Drug Discov. 2016;15(8):533–550.
     View this article via: CrossRef PubMed Google Scholar
116. Murillo MM, et al. Disruption of the interaction of RAS with PI 3-kinase induces regression of EGFR-mutant-driven lung cancer. Cell Rep. 2018;25(13):3545–3553.
     View this article via: CrossRef PubMed Google Scholar
117. Simanshu DK, et al. BBO-10203 inhibits tumor growth without inducing hyperglycemia by blocking RAS-PI3Kα interaction. Science. 2025;389(6758):409–415.
     View this article via: CrossRef PubMed Google Scholar
118. Ruiz-Saenz A, et al. HER2 amplification in tumors activates PI3K/Akt signaling independent of HER3. Cancer Res. 2018;78(13):3645–3658.
     View this article via: CrossRef PubMed Google Scholar
119. Mayer IA, et al. A Phase II randomized study of neoadjuvant letrozole plus alpelisib for hormone receptor-positive, human epidermal growth factor receptor 2-negative breast cancer (NEO-ORB). Clin Cancer Res. 2019;25(10):2975–2987.
     View this article via: CrossRef PubMed Google Scholar
120. McCart Reed AE, et al. Invasive lobular carcinoma of the breast: morphology, biomarkers and ’omics. Breast Cancer Res BCR. 2015;17(1):12.
     View this article via: CrossRef PubMed Google Scholar
121. Buttitta F, et al. PIK3CA mutation and histological type in breast carcinoma: high frequency of mutations in lobular carcinoma. J Pathol. 2006;208(3):350–355.
     View this article via: CrossRef PubMed Google Scholar
122. Jhaveri KL, et al. A phase 1/2 trial of tersolisib (LY4064809/STX-478), a pan-mutant-selective PI3Kα inhibitor (PI3Kαi), in PIK3CA-mutant advanced solid tumors: Updated results from PIKALO-1. J Clin Oncol. 2026;44(16_suppl):1072.
     View this article via: CrossRef Google Scholar
123. Rugo HS, et al. Capivasertib and fulvestrant for patients with hormone receptor-positive advanced breast cancer: characterization, time course, and management of frequent adverse events from the phase III CAPItello-291 study. ESMO Open. 2024;9(9):103697.
     View this article via: CrossRef PubMed Google Scholar
124. Buckingham L, et al. Ipatasertib, an oral AKT inhibitor, inhibits cell proliferation and migration, and induces apoptosis in serous endometrial cancer. Am J Cancer Res. 2022;12(6):2850–2862.
     View this article via: PubMed Google Scholar