PI3K Inhibitors in Cancer: Clinical Implications and Adverse Effects

1PI3K/AKT signaling is involved in important physiological and pathophysiological functions that drive tumor progression such as metabolism, cell growth, proliferation, angiogenesis and metastasis. Pharmacological or genetic suppression of this signaling causes cancer cell death and regression of tumor growth. The PI3K pathway is activated via point mutation of the PIK3CA gene or inactivation of the phosphatase and tensin homolog (PTEN) gene. Activation of this pathway occurs in approximately 30–50% human cancers and results in resistance to various anti-cancer therapies.

2PI3K proteins are classified into three main classes (I, II and III) based on the substrate specificities and structural characteristics. Class I PI3Ks are categorized into two subtypes (A and B) based on the mode of regulation. Class IA PI3Ks form dimers containing a regulatory (p85α, p85β, p55α, p55γ, p50α) and a catalytic (p110α, p110β, p110δ) subunit. Class I PI3Ks act downstream of both G protein-coupled receptors (GPCRs) and receptor tyrosine kinases (RTKs). In addition, these regulatory subunits play an important role in stabilization of the p110 catalytic subunits and in suppression of the basal lipid kinase activity. Class IB PI3Ks which are activated downstream of GPCRs form heterodimers consisting of p110γ as catalytic subunit with other regulatory subunits such as p101, p87 or p84. This class consists of three isoforms (PI3K-C2α, PI3K-C2ß and PI3K-C2γ). These isoforms have a RAS-binding domain (RBD), a helical domain, and a catalytic domain but lack a regulatory domain. The class III PI3Ks also known as vacuolar protein sorting 34 (VPS34) heterodimerize with membrane-associated VPS15 regulatory subunit. This VPS34-VPS15 complex is ubiquitously expressed in mammals regulating several functions such as phagocytosis, autophagy, endocytosis and intracellular trafficking. Some reports have considered mTOR as class IV PI3K-related kinase whose catalytic domain is associated with ataxia-telangiectasia (ATM), fluorescence recovery after photo bleaching (FRAP) and transformation/transcription domain-associated protein (TRRAP) and FK506-binding protein 12 (FKBP12)/rapamycin (FRB)-binding domain and a C-terminal FAT domain (FATC). There are two Huntington elongation factor 3, PR65/A subunit of protein phosphatase 2A, TOR (HEAT) repeats in the N-terminal region which modulate the protein–protein interactions in mTOR signaling.

3Small-molecule inhibitors targeting PI3K consist of pan, isoform-specific and dual PI3K/mTOR inhibitors. Wortmanin and LY294002 were the first-generation PI3K inhibitors that belong to the non-isoform-specific category. Wortmanin with irreversible inhibition, lack of selectivity and adverse effects resulted in termination of its clinical trials. Poor solubility, bioavailability and rapid degradation of LY294002 prevented its further biological evaluation and clinical studies. To date, five PI3K inhibitors (Copanlisib, Idelalisib, Umbralisib, Duvelisib and Alpelisib) have been approved by the United States Food and Drug Administration (FDA). This has encouraged clinicians and researchers to test other PI3K inhibitors in both preclinical and clinical settings with the goal to identify a potent PI3K inhibitor with significant clinical efficacy, low toxicities and optimum bioavailability. In this review, we discuss the challenges associated with successful development of PI3K inhibitors including the aberrant mutations/amplification of genes such as PI3K and/or PI3K-driven molecules, the role of these mutations and other molecules in PI3K-mediated drug sensitivity, and compensatory signaling mechanisms resulting in PI3K inhibitor resistance causing treatment failure in the clinic. In addition, we have summarized active, not-recruiting and completed trials along with detailed clinical outcomes for PI3K inhibitors in a wide variety of cancers along with the adverse events.

4PI3K signaling is deregulated in a variety of cancers. There are several factors that ultimately result in resistance to PI3K inhibitors including (1) inactivation or loss of PTEN activity, (2) mutations and amplification of PI3K, (3) drug-related toxicities, (4) feedback upregulation leading to compensatory mechanisms, (5) non-coding RNA in regulating PI3K signaling, (6) enhanced insulin production upon PI3K inhibition, (7) selection of patient population in clinical trials and (8) miscellaneous resistance mechanisms, as summarized below.

5PTEN is a well-characterized tumor suppressor which is a negative modulator of the PI3K pathway. PTEN regulates PI3K via dephosphorylating phosphatidylinositol-3, 4, 5-trisphosphate (PIP3). The loss of or gain in function is seen in several cancers including breast, brain and prostate. Increased PI3K signaling and tumorigenesis is observed in PTEN knock-in mice with PTEN mutations. Loss of PTEN lipid phosphatase activity causes the activation of AKT in PTEN-null cancers. A study by Juric et al. indicates that loss of PTEN can lead to clinical resistance to a PI3K inhibitor, Alpelisib in breast cancer. These studies indicate the possible role of PTEN in modulating response to PI3K inhibitors in different cancers.

6The p110α subunit of PIK3CA is frequently mutated and amplified (~30%) in a variety of cancers. There are hotspot mutations such as PIK3CAE545K in exon 9 in the helical PI3K homology domain that suppresses inhibition of p110α by p85 regulatory subunit. Another mutation is PIK3CAH1047 present in the catalytic subunit in exon 20 which enhances the interaction of p110α with the lipid membrane. E542K is another crucial PI3K mutation with significant oncogenic potential associated with elevated in vitro catalytic activity.The helical and kinase mutations in p110α domain induce oncogenic activity based on different interaction between the PI3K regulatory subunit, p85 with RAS-GTP. The oncogenic activity induced by helical domain mutations is independent of binding to p85 subunit. However, requires interaction with RAS-GTP. However, the kinase domain mutations are active without RAS-GTP binding but dependent on their interaction with p85. In addition, co-existence of both the domain mutations increases the p110 function and tumorigenic activity synergistically. A recent study has shown that C2 domain deletions in PIK3CA activate PI3K signaling significantly and also enhance the sensitivity to PI3Kα inhibitors. The deregulation of PI3K signaling leads to several oncogenic activities such as cancer cell proliferation, invasion, migration, glucose transport and angiogenesis that regulate tumor progression.

7Genetically engineered mouse models (GEMMs) using PIK3CA mutations demonstrate important tumorigenic potential associated with PI3K signaling. For example, the PIK3CAH1047R knock-in mouse model promoted conditional activation of the PI3K pathway and induced breast tumorigenesis. Another study indicated that PIK3CA H1047R mutation with PTEN deletion led to the development of ovarian adenocarcinoma and granulosa cell tumor in a mouse model. Further, overexpression of wild-type catalytic subunits (p110ß, p110δ and p110γ) is known to induce an oncogenic phenotype. The role of p110ß is well established in cancers including breast and prostate. Although the mechanism of action of this subunit is not well known, some reports still indicate that it works via GPCR signaling. E633K is a p110ß helical mutation first reported in HER2+ breast cancer. PIK3CAD1067V is another recurrent somatic mutation in the p110β subunit which induced in vitro and in vivo cancer cell growth via the activated PI3K signaling pathway. Pharmacologic inhibition using a specific p110β inhibitor (TGX-221) inhibited growth in patient-derived renal cell carcinoma cells with endogenous PIK3CAD1067V and epidermal growth factor receptor (EGFR)-mutant lung adenocarcinoma cells as well as in NIH-3T3 cells engineered to express this mutation. In addition, expression of this mutation is known to promote Erlotinib resistance in EGFR-mutant lung adenocarcinoma cells. PI3Kδ plays a crucial role in myeloid cell activities such as inflammation-driven cell infiltration, neutrophil oxidative burst, immune complex-mediated macrophage activation, mast cell maturation and degranulation. It also plays an important role in inducing cancer cell proliferation and AKT activation in myeloid leukemia. PI3Kδ point mutations have been identified in a panel of diffuse B-cell lymphomas. It is known to regulate in vitro chemotactic migration in response to EGF and share similar biological function in breast cancer cells and macrophages.

8PI3Kγ is expressed in immune cells of myeloid origin that regulates innate immunity in both cancer and inflammatory cells. PI3Kγ regulates solid tumor neovascularization, contributes to chemotactic response and reactive oxygen species (ROS) production in neutrophils. All these studies indicate the key role of PI3K and PI3K isoform-mediated signaling along with site-specific driver mutations that leads to oncogenic transformation of cancer cells which could ultimately lead to drug resistance.

9One major hurdle in the development PI3K inhibitors is the inability to achieve optimal drug-target blockade in tumors due to drug-related toxicities in patients. Toxicities from small-molecule PI3K inhibitors depend on their PI3K isozyme specificity. For example, the adverse effects associated with PI3Kα inhibitors are mostly rash and hyperglycemia, and the side effects associated with δ subunits are mostly gastrointestinal, transaminitis and myelosuppression. The pan-PI3K inhibitors share common dose-dependent toxicities such as fatigue, diarrhea, rash and hyperglycemia. Dual PI3K inhibitors have a broader toxicity profile compared to isoform or pan-PI3K inhibitors. Isoform-specific PI3Kδ inhibitors have a better effect in B-cell malignancy compared to pan or dual PI3K inhibitors because of the toxicity profile. Several studies have indicated that intermittent dosing schedules of PI3K drugs have a better safety profile over a continuous dosing pattern. A study by Hudson et al. showed that when administrated intermittently PI3Kα/δ inhibitor AZD8835 showed better anti-tumor efficacy in breast cancer xenograft models as a monotherapy and in combination with an ER inhibitor (Fulvestrant) and CDK4/6 inhibitor (Palbociclib) that suppressed AKT activation and induced cell death. Another study demonstrated that intermittent administration of mitogen-activated protein kinase kinase (MEK) inhibitor, GDC-0973 with a PI3K inhibitor (GDC-0941) triggered cell death and tumor growth inhibition. However, dosing schedules, suboptimal dose selection and on-target toxicities of several PI3K inhibitors failed to cause the complete and sustained inhibition of PI3K signaling, challenging the efficacy of these PI3K inhibitors in the clinic.

10Drug resistance associated with pharmacological inhibition of PI3K inhibitors is caused due to feedback upregulation of the PI3K/AKT/mTOR pathway involving RTKs, growth factors and downstream transcription factors. One such transcription factor is forkhead box (FOXO), which represses RTKs or other adaptors that activate the PI3K pathway, such as human epidermal growth factor receptor 3 (HER3), insulin-like growth factor 1 receptor (IGF1R), EGFR, fibroblast growth factor receptor (FGFR) and insulin receptor (InsR). Therefore, inhibiting PI3K/AKT signaling suppresses the FOXO phosphorylation, causing FOXO-dependent repression of RTKs leading to derepression of molecules downstream of AKT such as S6K and growth factor receptor-bound protein 10 (GRB10), ultimately resulting in activation of multiple RTKs and partial maintenance of PIP3. Partial maintenance of PIP3 is also mediated by PI3K isoform p110ß in luminal breast cancer cells where PI3K is hyperactivated due to HER2-HER3 dimers. Another study showed the key role of FOXO-mediated adaptive resistance in matrix-attached cancer cells in response to inhibition of PI3K/mTOR signaling. Another FOXO-dependent adaptive response is via upregulation of Rictor, resulting in enhanced AKT activation in renal cancer cells.

11Another transcription factor, signal transducer and activator of transcription 5 (STAT5) plays a crucial role in mediating resistance to PI3K inhibitors in vitro. A study by Britschgi et al. showed that STAT5 is activated by Janus kinase 2 (JAK2) that evoked a positive feedback loop to dampen the efficacy of PI3K/mTOR inhibition via secretion of interleukin-8 (IL-8) in several cell lines and primary breast tumors. The data also showed that pharmacological and genetic inhibition of JAK2 combined with PI3K/mTOR inhibition abrogated this feedback loop. Treatment with the AKT inhibitor AZD5363 activates several RTKs, estrogen receptor 1 (ESR1) mRNA, and ERα-mediated transcription of insulin-like growth factor 1 (IGF1) and IGF2 ligands. In another study, expression of ESR1 mRNA and ER-related proteins are increased upon PI3K inhibition using Alpelisib. These drug-related transcriptional alterations were abrogated via anti-ER drugs. Treatment with Alpelisib in ER+ breast cancer cells or patients with primary tumors resulted in activation of ERα via lysine methyltransferase 2D (KMT2D)-dependent FOXA1-PBX1 complex. All these data highlight the role of ERα-mediated feedback upregulation in response to PI3K inhibition that might result in drug insensitivity in PIK3CA-driven cancers.

12Non-coding RNAs (ncRNAs) are commonly employed for RNAs that do not encode proteins which are involved in cancer drug resistance. These ncRNAs include siRNA, piRNA, miRNA and lncRNA. In general, ncRNAs function to regulate gene expression at the transcriptional and post-transcriptional level. Several lncRNAs are known to interact with PI3K to activate signaling. For example, lncRNA CRNDE is overexpressed in several cancers to activate cell proliferation and growth via PI3K-dependent pathways. lncRNA OIP5-AS1 is another ncRNA that activates PI3K signaling and cause Cisplatin resistance in osteosarcoma. OIP5-AS1 loss causes miR-410 accumulation that facilitates cell cycle progression, proliferation and apoptosis inhibition by targeting Krüppel-Like Factor-10 (KLF10) via activating the PI3K/AKT/mTOR pathway in multiple myeloma. lncRNA CCAT1 regulates thyroid and squamous cancer cell migration via activation of the PI3K/AKT pathway. lncRNA H19 inhibits melanoma cell migration and invasion via inactivating PI3K/AKT/NFĸB signaling. lncRNA HOTAIR promotes Cisplatin resistance in gastric cancer by targeting miR-126/miR-34a through activation of the PI3K/AKT pathway. It is also involved in multidrug resistance to Imatinib in a PI3K/AKT-dependent signaling in myeloid leukemia cells. It also regulates cancer cell growth, metastasis and apoptosis via PI3K/AKT-dependent pathway in melanoma, adenocarcinoma and glioma. lncRNA NEAT1 plays a key role in PI3K/AKT-mediated tumorigenesis in several cancer types. HULC is another lncRNA that is involved in bladder cancer cell proliferation via regulation of ZIC2 and PI3K/AKT signaling. It induces leukemia cell proliferation and suppresses angiogenesis in gliomas via an AKT-dependent pathway. AB073614 and PTTG3P are known to promote cancer cell proliferation, migration and invasion via PI3K/AKT-dependent signaling in colorectal and hepatocellular carcinoma, respectively. MALAT1 is another crucial lncRNA which is upregulated in several cancers and induces tumorigenic phenomenon via miRNA-mediated PI3K/AKT-dependent signaling. lncRNA ATB induces bladder cancer cell proliferation, migration, and invasion via AKT/mTOR-mediated signaling. lncRNA BC087858 induces acquired resistance to EGFR-TKIs by activating PI3K/AKT and epithelial–mesenchymal transition (EMT)-dependent signaling in lung cancer.

13Linc00659 and Linc00152 induce cancer cell growth and inhibit apoptosis in colorectal and lung cancer, respectively, via PI3K/AKT-mediated pathway activation. In similar lines, Linc00462 and Linc01296 regulate different hallmarks in tumor progression via AKT signaling. There are several other lncRNAs including UCA1, ecCEBPA, Ftx, RMEL3, lncARSR, BDLNR, ANRIL, ROR, MYD88, RNA-422 and PlncRNA-1 that are upregulated and mediate PI3K-dependent signaling in different cancers. Other lncRNAs which are downregulated yet known to mediate tumorigenesis via PI3K/AKT-dependent signaling include linc003121 in thyroid cancer, RP4 in colorectal cancer, MEG3 in endometrial cancer, GAS5 in various cancers. In summary, there is a long list of ncRNAs including those discussed above which are promising candidates to be targeted to overcome PI3K inhibitor resistance in clinic.

14The p110α subunit of PI3K is known to mediate cellular responses to insulin signaling in different diseases including cancer. Targeting this subunit via specific PI3K inhibitors inhibits the insulin-dependent signaling, which in turn promotes glycogen breakdown in liver and inhibits glucose uptake in skeletal muscle and adipose tissues, resulting in hyperglycemia. However, this effect is compensated via the insulin feedback pathway that activates PI3K/AKT/mTOR signaling in tumor cells compromising the therapeutic response to PI3K inhibitors. It has been shown that use of dietary and pharmaceutical approaches can also target this insulin feedback, resulting in enhanced efficacy and reduced toxicities due to PI3K inhibitors. Molinaro et al. have shown that insulin-mediated PI3K/AKT signaling is dependent on redundant PI3Kα/β activities and is mediated by RAS. In another study in acute leukemia patients, the effects of PI3K/AKT/mTOR inhibitors and insulin varied among the patient subsets, and it was noted that the variations in insulin responsiveness are associated with differential susceptibility to metabolic targeting. An increase in the plasma levels of insulin that is associated with hyperglycemia is also observed in patients treated with PI3K inhibitors and hence enhanced insulin level serves as a pharmacodynamics surrogate marker for such patients.

15Selection of accurate patient population with active PIK3CA mutations determines the efficacy of PI3K inhibitors. For example, in a phase I study, the clinical benefit rate (CBR) to a PI3K inhibitor, Alpelisib was 44% in tumors with PIK3CA hotspot mutations versus 20% in those patients with WT PIK3CA. In another study, patients with PIK3CA H1047R mutation had a better survival rate (36%) over subjects with WT tumors. In addition, significant improvement in the progression-free survival (PFS) was seen for patients with PIK3CA mutant-specific ctDNA treated with Alpelisib, indicating that the efficacy of PI3Kα inhibitors is dependent on PIK3CA-mutant tumors. However, patients with other concurrent alterations in TP53, FGFR1 and KRAS did not benefit from Alpelisib, indicating that exclusion of these additional resistance biomarkers could enhance the efficacy of PI3K inhibitors in cancers. There is evidence suggesting that PIK3CA hotspot mutations do not always capture all PI3K-dependent tumor genotypes that potentially respond to PI3K inhibitors. Hence, selection of the optimal patient population with activating PIK3CA mutations and identification of other genotypes conferring PI3K pathway dependence including PIK3CA mutations or amplifications must be included in PI3K inhibitor-based trials.

16In addition to those mentioned above, there are other factors and mechanisms of resistance and/or compensatory pathways derived from both clinical and in vitro lab studies, which affect PI3K signaling. These include (a) HRAS and KRAS mutations which reduce susceptibility to PI3K inhibitors while knockdown of these has shown to improve sensitivity to PI3K inhibitors. (b) NOTCH-MYC pathway/MYC amplification: NOTCH pathway and downstream induction of c-MYC confer resistance to PI3K inhibitors whereas knocking down MYC reverses resistance to PI3K/mTOR dual inhibitors. Another report indicated that MYC amplification is associated with eIF4E-mediated resistance to dual PI3K inhibitors. (c) Other factors that confer resistance to PI3K inhibitors include ribosomal S6 kinases 3/4 (RSK3/4), p21-activated kinase1 (PAK1), mitogen- and stress-activated kinase 1 (MSK1), lysine demethylase 6B (KDM6B), insulin-like growth factor-binding protein 5 (IGFBP5) and cyclin-dependent kinase 4/6 (CDK4/6). (d) FGFR1amplifications. Overexpression and/or amplification of AXL, serum/glucocorticoid regulated kinase family member 3 (SGK3), PIM1, S-phase kinase-associated protein 2 (SKP2), phosphoinositide-dependent kinase-1-serine/threonine-protein kinase1 (PDK1-SGK1). In addition, there are other factors leading to resistance mechanisms such as a high level of purine-related metabolites, elevated glycolysis with dysregulated mitochondrial signaling leading to loss of PTEN and PPP2R2B expression in response to dual PI3K/mTOR inhibitors, macrophages in the tumor microenvironment contributing to PI3K inhibitor resistance via the NFĸB pathway, and lack of PI3K mutant-specific inhibitors currently available. These are other pathways which need to be explored and targeted in the development of successful PI3K inhibitors.

17There are several PI3K inhibitors including pan, isoform-specific and dual PI3K/mTOR inhibitors which are tested in various phases of human clinical trials. In this section, we describe the active not-recruiting and completed studies involving PI3K inhibitors. We have also briefly mentioned about novel pan-PI3K inhibitors yet to be tested in clinical trials including RIDR-PI-103 that is currently investigated in our laboratory. Figure 1 highlights various drugs targeting PI3K signaling.

18Pan-PI3K inhibitors target all four isoforms (α, ß, δ, and γ) of class I PI3K and are ATP-competitive inhibitors. These inhibitors are associated with several adverse effects due to unselective blockage of PI3K signaling. Herein, we have listed pan-PI3K inhibitors which are in different stages of human clinical trials. Out of all pan-PI3K inhibitors, only Copanlisib has been approved by the FDA.

19Buparlisib/NVP-BKM120/BKM120 is a potent oral reversible bioavailable pan-PI3K inhibitor targeting p110 -α,-ß, -δ, -γ isoforms with IC50 values of 52, 166, 116 and 262 nM, respectively. Few of the active not-recruiting and completed trials testing Buparlisib in phase I and phase II are mentioned below. Most of the clinical trials using Buparlisib are either sponsored or are in collaboration with Novartis Pharmaceuticals. For example, a phase I active trial is testing Buparlisib with Rituximab, a chimeric monoclonal antibody against CD20 in patients with relapsed or refractory (R/R) indolent B-cell lymphoma. The study was sponsored by the Ohio State University Comprehensive Cancer Center. The main aim of the trial was to determine the maximum tolerated dose (MTD) and dose-limiting toxicities (DLT) of this drug combination.

20There are several completed studies with Buparlisib without any significant clinical outcomes as summarized below. The trials with specific clinical outcomes and adverse events are summarized in Table 1. An early phase I study aimed to determine the grade of inhibition of PI3K/mTOR signaling in a pre-surgery setting with Buparlisib along with potential biomarker assessment for a pathologic complete response in breast cancer patients. This was sponsored by Grupo Hospital de Madrid. A phase II trial sponsored by University Hospital, Essen investigated the efficacy and safety of Buparlisib with tamoxifen (an ER inhibitor) in ER/PR+, HER2- breast cancer patients with prior exposure to anti-hormonal therapy. A phase I multicenter, non-randomized open-label dose-escalation study aimed to investigate the MTD of Buparlisib in combination with LEE011 (a CDK4/6 inhibitor) and Letrozole (a non-steroidal aromatase inhibitor) in HR+, HER2- post-menopausal women with locally advanced or metastatic breast cancer (MBC) was completed without significant outcomes. A phase I study tested the efficacy of Buparlisib in Chinese patients with advanced solid tumors.

21Another phase I study of Buparlisib was explored in combination with chemotherapeutics Cisplatin and Etoposide in patients with small-cell lung cancer (SCLC). The study was sponsored by University of California. The primary objective of the trial was to determine the safety and feasibility and the secondary objectives were to establish the MTD, DLT and pharmacokinetic parameters of this triple drug combination regimen. It was also investigated with LEE011 plus Fulvestrant along with another PI3K-α-specific class I inhibitor (BYL719) to explore the clinical utility of this drug combination in ER+/HER2- locally advanced or metastatic breast cancer patients. A phase I dose-escalation treatment regimen included Buparlisib administrated in a concurrent treatment arms with chemotherapy (Capecitabine) or anti-HER2 targeted therapy in patients with MBC. The first treatment arm tested Buparlisib plus Capecitabine, the second arm tested BYL719 plus Capecitabine, the third arm tested BKM120 with Capecitabine plus Trastuzumab (a monoclonal antibody targeting HER2) and the last arm tested the concurrent treatment of Buparlisib with Capecitabine plus Lapatinib (a HER2-specific inhibitor). The study was initiated by the UNC Lineberger Comprehensive Cancer Center in collaboration with Novartis Pharmaceuticals and aimed to achieve the tolerability, safety, and MTD of these regimens. A phase II study sponsored by US Oncology Research investigated the safety and efficacy of Buparlisib in combination with chemotherapy, Capecitabine in breast cancer patients with brain metastases. Another chemotherapy (Docetaxel) was tested with Buparlisib in an early phase I pilot study to identify the adverse effects and the recommended phase 2 dose (RP2D) of Buparlisib in patients with locally advanced metastatic solid tumors who cannot be treated with surgery. The study was sponsored by Roswell Park Cancer Institute in collaboration with NCI and Novartis Pharmaceuticals. A multicenter, open-label phase I study sponsored by the UNC Lineberger Comprehensive Cancer Center aimed to define the MTD of Buparlisib with a chemotherapy regimen folinic acid (leucovorin)-fluorouracil-oxaliplatin (mFOLFOX6) in advanced metastatic pancreatic cancer patients. A phase I trial aimed to evaluate the preliminary anti-tumor effect, the safety profile/tolerability of BKM120 or BEZ235 (another PI3K/mTOR inhibitor) with Letrozole (non-steroidal aromatase inhibitor) used as endocrine therapy in post-menopausal MBC patients. The study was sponsored by the Vanderbilt-Ingram Cancer Center. A phase I study sponsored by Washington University School of Medicine in collaboration with NCI aimed to determine the toxicity profile and the MTD of Buparlisib with Fulvestrant in estrogen receptor (ER)+ stage IV breast cancer patients.

22A phase I study evaluating the best dose of Buparlisib in combination with chemotherapies (Carboplatin and Pemetrexed disodium) was completed in patients with stage IV lung cancer. The study was sponsored by City of Hope Medical Center with NCI and aimed to figure out the RP2D, DLTs of this combination regimen. Other chemotherapies were also tested in combination with Buparlisib. For example, in a phase Ib study sponsored by Fondazione Michelangelo, Buparlisib was tested with Cisplatin (in group 1) or with Carboplatin (in group 2) to identify the RP2D of these drug combinations for patients with advanced solid tumors. A phase I pilot study aimed to identify the clinical benefits, overall response rate (ORR) and toxicities associated with Buparlisib treatment in patients with non-hodgkin lymphoma (NHL). The trial was sponsored by Mayo Clinic with NCI. Two phase I/II trials tested the efficacy of Buparlisib in combination with either thoracic radiotherapy or with Erlotinib, an EGFR inhibitor in non-small-cell lung cancer (NSCLC) patients. An open-label, phase II trial evaluated Buparlisib to determine the PFS for patients with R/R primary central nervous lymphoma (PCNSL) and R/R secondary central nervous system lymphoma (SCNSL).

23CH5132799/PA-79 is a novel oral pan-class I PI3K inhibitor with a strong inhibitory activity against the PI3Kα isoform (IC50 = 0.014 μM). This drug has promising anti-tumor activity in PI3K mutation-driven cancers and xenograft mice models. To date, there is only one phase I study that has tested CH5132799 as described in Table 1.

24Pilaralisib/XL147/SAR245408 is a reversible pan-class I PI3K inhibitor against the PI3Kα/δ/γ isoforms with respective IC50 values of 39 nM/36 nM/23 nM in cell-free assays although known to be less potent to PI3Kβ isoform. Most trials using Pilaralisib were sponsored by Sanofi. In an early phase I study sponsored by Sanofi in collaboration with Merrimack Pharmaceuticals, Pilaralisib was tested with MM-121/SAR256212, an anti-HER3 antibody with a goal to determine the MTD and RP2D of this drug combination in patients with advanced solid tumors. Another such phase I dose-escalation trial tested the clinical efficacy of Pilaralisib with MSC1936369B, an oral MEK inhibitor in locally advanced or metastatic solid tumors to evaluate the safety, tolerability and initial PK profile of this drug combination. Similar phase I and II studies sponsored by Sanofi, respectively, investigated Pilaralisib as monotherapy in patients with advanced solid tumors or recurrent endometrial cancer.

25ZSTK474 is an oral ATP-competitive pan-PI3K class I inhibitor with week effectiveness against mTOR. To date, Zenyaku Kogyo Co., Ltd., Tokyo, Japan has sponsored all the known clinical trials using this drug. ZSTK474 was tested as a monotherapy in open-label phase I studies in Japanese patients and other cancer patients with advanced solid tumors to evaluate the safety, tolerability, efficacy and initial PK profile.

26Sonolisib/PX-866 is an improved semi-synthetic analog of Wortmannin, which is an oral, irreversible pan-PI3K inhibitor that had demonstrated significant anti-tumor effect in squamous cell carcinoma of head and neck (SCCHN) patients with PI3K mutation. It exhibits strong inhibitory activity against PI3K110 -α, -δ and -γ isoforms but poor activity against the ß isoform. A phase I/II study tested Sonolisib in combination with Docetaxel in patients with solid tumors. The study was sponsored by Cascadian Therapeutics which also evaluated Sonalisib as monotherapy in patients with advanced solid tumor in yet another early phase I trial. A phase I/II open-label study determined the safety and efficacy of Sonolisib with Cetuximab-based chemotherapy in a combination treatment regimen. In the phase I part of the trial, Sonolisib plus Cetuximab was administrated in patients with castration resistant colorectal (CRC) cancer or progressive, recurrent SCCHN. The phase II part of the study was a randomized evaluation of the anti-tumor activity and safety of Sonolisib plus Cetuximab versus Cetuximab alone in patients with either incurable metastatic CRC who had a prior history of progression or recurrence on chemotherapy-based treatment regimens including Irinotecan and Oxaliplatin, or were resistant to Irinotecan treatment (group 1) or in metastatic SCCHN (group 2).

27Pictilisib/GDC-0941/RG7321 is an oral bioavailable potent PI3Kα/δ inhibitor with an IC50 of 3 nM and has weak selectivity against the p110β (10-fold) and p110γ (25-fold) isoforms. It binds and competes with the ATP-binding pocket to inhibit PI3K signaling pathway. It also had a better tolerability and safety profile in Japanese patients with advanced NSCLC. Most of the trials using Pictilisib are sponsored by Genentech. In one such open-label, multicenter, phase Ib dose-escalation study, Pictilisib was administered with one of three chemotherapies regimens: Arm A: Paclitaxel plus Carboplatin in Bevacizumab-ineligible NSCLC patients, Arm B: Paclitaxel, Carboplatin plus Bevacizumab in Bevacizumab-eligible NSCLC patients and Arm C: Pemetrexed, Cisplatin with Bevacizumab in Bevacizumab-eligible NSCLC patients to assess the tolerability, safety and PK profile of pictilisib. Another phase II study tested Pictilisib with the above chemotherapy regimens in NSCLC patients. A phase Ib open-label, three arm and dose-escalation study investigated the safety, tolerability, PK and activity of oral Pictilisib administered with either intravenous (IV) infusion of humanized monoclonal antibody Trastuzumab covalently linked to the cytotoxic agent DM1 (T-DM1) or IV infusion of Trastuzumab in HER+ breast cancer progressed on Transtuzumab therapy. Another phase I study tested the safety and efficacy of Pictilisib in advanced or metastatic tumors.

28Copanlisib/BAY 80-6946/Aliqopa is an intravenous highly potent and reversible pan-class I PI3K inhibitor with significant activity against α and δ isoforms. Copanlisib was brought into market by Bayer and approved by the FDA for treatment of patients with relapsed follicular lymphoma (FL). In a phase I trial, Copanlisib was tested with Paclitaxel in patients with advanced cancer to determine its MTD and RP2D.

29There are several active trials which are not currently enrolling patients for Copanlisib testing as summarized below. For example, a phase Ib/II active trial is investigating Copanlisib in Japanese patients with relapsed indolent B-cell NHL. In another phase I study (ROCOCO), Copanlisib is currently being explored with Rogaratinib (a FGFR inhibitor) to evaluate the safety, tolerability and the MTD of this drug combination in patients with locally advanced or metastatic solid tumors with a primary objective to find the RP2D. The secondary objective of the trial is to characterize the PK profile and anti-tumor efficacy. A phase I/II study is testing the safety and efficacy of Copanlisib with chemotherapy, Gemcitabine in patients with R/R peripheral T-cell or NK/T-cell lymphoma. The study is sponsored by Chonnam National University Hospital in collaboration with Consortium for improving survival of lymphoma and Bayer to determine the DLTs, MTD and RP2D. In a large-scale randomized phase III study (CHRONOS-2 and 3), Copanlisib is treated alone or in combination with Rituximab in Rituximab-refractory indolent NHL patients. In further extension of this phase III study (CHRONOS-4) Copanlisib is explored with standard immune-chemotherapy versus placebo or immunotherapy control to assess the PFS in the above patients. A phase Ib/II study (COPAN-ORL06) sponsored by UNICANCER is assessing the MTD, efficacy and RP2D of Copanlisib in combination with Cetuximab in patients with recurrent and/or metastatic head and neck squamous cell carcinoma (HNSCC) with a PIK3CA mutation/amplification and/or a PTEN loss. A phase I/II trial is investigating the best dose and side effects of Copanlisib when given with Letrozole and Palbociclib in HR+ and HER2- stage I–IV breast cancer patients. A phase II study is testing the safety and efficacy of Copanlisib plus Gemcitabine and Cisplatin in patients with advanced cholangiocarcinoma. The study is sponsored by the H. Lee Moffitt Cancer Center and Research Institute in collaboration with Bayer.

30B591, a dihydrobenzofuran-imidazolium salt is a novel pan-PI3K inhibitor with potent inhibitory activity against class I PI3K. This has shown significant inhibition of cellular PI3K/AKT signaling with robust anti-tumor activity in a set of cancer cell lines. However, it has not entered into clinical trials.

31TG-100-115 is a pan-PI3K inhibitor, developed by Sanofi, which has significant activity against γ/δ isoforms with IC50 values of 83 nM/235 nM in cell-free assay. However, TG-100–115 has less effect against α/β isoforms. This drug is in phase I and phase II trials in other diseases but not in cancer. However, Song et al. have shown that TG100-115 can be used as a potent TRPM7 kinase inhibitor and a potent suppressor of breast cancer cell migration.

32The pyridinylfuranopyrimidine inhibitor, PI-103 is a pan-PI3K inhibitor which targets p110α/β/δ/γ isoforms with IC50 values of 2 nM/3 nM/3 nM/15 nM in cell-free assays and has a strong activity against PI3Kα isoform. However, it failed in clinical trials due to toxicity and poor bioavailability issues. Therefore, a bioisostere of PI-103 (RIDR-PI-103) was synthesized with a structural modification containing a boronate in place of a phenolic hydroxyl group which enhanced the bioavailability of PI-103. RIDR-PI-103 differs from the boron-containing compound as it releases PI-103 under oxidative stress in tumor microenvironment. This self-cyclizing prodrug has shown significant in vitro anti-cancer efficacy in acute myeloid leukemia. In addition, our laboratory has shown that RIDR-PI-103 in combination with the chemotherapeutic, doxorubicin impaired PI3K signaling, increased DNA damage response and suppressed in vitro breast cancer cell growth. Active not-recruiting and completed trials with specific clinical outcomes for pan-PI3K inhibitors are summarized in Table 1.

33Isoform-specific PI3K inhibitors cause selective inhibition of specific PI3K isoforms (α/β/γ or δ). Out of all the isoform-specific drugs, four drugs which included Alpelisib, Idelalisib, Umbralisib and Duvelisib are approved by the FDA. These inhibitors reduce off-target toxicities although these require narrow patient selection.

34Alpelisib/BYL719/Piqray is a potent oral selective PI3Kα inhibitor targeting PIK3CA-mutated cancers. The role of Alpelisib has also been identified outside cancer in PIK3CA-related overgrowth syndromes (PROS) which are caused by mosaic gain-of-function mutations in the PIK3CA gene. It is shown that daily oral administration of Alpelisib in the mouse model of PROS post-tamoxifen administration, effectively suppressed PIK3CA pathway activation, which prevented the PROS phenotype and enhanced survival. Alpelisib was approved by the FDA for PIK3CA-mutated, HR+, HER2- advanced breast cancer who had received endocrine therapy previously. Most of the studies using Alpelisib are sponsored by Novartis Pharmaceuticals. One such trial tested Alpelisib in a phase I study with a HSP90 inhibitor (AUY922) to determine the MTD and/or recommended dose of expansion (RDE) of this drug combination in patients with advanced gastric cancer. A dose-escalation phase Ib trial involved Alpelisib with Imatinib (a RTK inhibitor) in gastrointestinal stromal tumor (GIST) patients who had relapsed with Imatinib and Sunitinib (a RTK inhibitor). A phase Ib/II investigated Alpelisib with HER3 monoclonal antibody (LJM716) versus Taxane or Irinotecan with previously treated esophageal squamous cell carcinoma patients. The aim of the study was to evaluate the MTD and/or RP2D of this drug combination. A phase Ib/II study investigated Alpelisib with an oral pan-PIM kinase inhibitor (LGH447/PIM447) in relapsed myeloma patients. The study aimed to explore the MTD and RP2D of this combination. Another phase II trial tested Alpelisib in Chinese patients with NSCLC.

35There are several active not-recruiting studies testing Alpelisib in various stages of clinical trials as mentioned below. A phase II study (SAFIR PI3K) is underway in patients with PIK3CA-mutated advanced breast cancer which compares Alpelisib and ER inhibitor (Fulvestrant) versus chemotherapy in these patients. The trial sponsored by UNICANCER aims to determine whether this drug combination could prolong the PFS versus maintenance chemotherapy in the above subjects. A phase I study is currently testing Alpelisib with a potent, oral bioavailable selective estrogen receptor degrader (SERD), LSZ102 in advanced or metastatic ER+ breast cancer patients progressed after endocrine therapy. The study aims to identify the safety, tolerability, PK and anti-tumor efficacy of this combined regimen. A phase I study is testing Alpelisib in locally advanced squamous cell carcinoma head and neck cancer (LA-SCCHN) patients in combination with standard radiation and chemotherapy, Cisplatin. The study is sponsored by University Health Network, Toronto in collaboration with Novartis.

36Serabelisib/INK-1117/MLN-1117/TAK-117 is a potent oral PI3K inhibitor with an IC50 of 21 nmol/L against PI3Kα isoform. The first in human phase I study evaluated the safety and efficacy of Serabelisb in patients with advanced solid tumors. A phase II study is investigating Serabelisib with oral mTORC1/2 inhibitor Sapanisertib (TAK-228) in TNBC patients to target the DNA repair pathway. The study is sponsored by Baylor Research Institute in collaboration with Takeda. A phase II study evaluated the efficacy and safety of MLN0128 and the combination of MLN0128 plus Serabelisib versus Everolimus in the treatment of metastatic clear-cell renal cell carcinoma (mccRCC) patients who progressed on vascular endothelial growth factor (VEGF)-targeted therapy. A phase I study tested Serabelisib with mTORC1/2 inhibitor Sapanisertib (TAK-228) in patients with advanced non-hematological patients to determine the DLTs, MTD and/or RP2D of this drug combination.

37GSK2636771 is an orally bioavailable PI3Kβ inhibitor with >900-fold selectivity over the PI3Kα/γ isoforms and >10-fold selectivity over the PI3Kδ isoform. A phase I study tested GSK2636771 with Enzalutamide in male patients with CRPC. The study was sponsored by GlaxoSmithKline and aimed to evaluate the safety and clinical efficacy to help guide the selection of the RP2D. To date, there are few active non-recruiting trials using GSK2636771. For example, a phase Ib/IIa, open-label, dose-finding study is evaluating the safety, efficacy and clinical activity of GSK2636771 with Paclitaxel in patients with higher grade of gastric adenocarcinoma. NCI is currently testing GSK2636771 in cancer patients with a PTEN alteration.

38Idelalisib/GS-1101/CAL-101/Zydelig, an oral PI3Kδ inhibitor with an IC50 value of 2.5 nM in cell-free assays was the first FDA-approved PI3K inhibitor for the treatment of patients with R/R CLL or LL/FL who progressed on prior therapies. A phase Ib study sponsored by Novartis Pharmaceuticals tested Idelalisib in a dose-escalation trial with BCL201 (elective BH3-mimetic inhibitor of BCL-2) in patients with FL and MCL. The primary objective of the study was to determine the safety and tolerability of this drug combination. Other completed trials with published clinical outcomes and adverse events for Idelalisib treatment are summarized in Table 2.

39There are several active not-recruiting trials using Idelalisib. For example, a pilot phase I study is investigating Idelalisib with ACY-1215/Ricolinostat (a histone deacetylase inhibitor) in patients with R/R CLL. The study was sponsored by Dana-Farber Cancer Institute in collaboration with Acetylon Pharmaceuticals. Idelalisib is currently being tested in combination therapies with either Rituximab plus TRU-016, mAb targeting CD37 (phase I) or with Tirabrutinib, an oral bruton’s tyrosine kinase (BTK) inhibitor (phase II) in such patients. Another phase II trial (ILIAD) is testing Idelalisib in patients with relapse DLBCL. A phase II (COSMO) is evaluating the safety and efficacy of MOR00208 with Idelalisib or Venetoclax in R/R CLL/SLL patients treated earlier with a BTK inhibitor.

40Zandelisib/ME-401/PWT-143 is an oral, selective PI3Kδ inhibitor which is in early phase of clinical trials mostly in lymphoma patients. Zandelisib is being investigated in Japanese patients with B-cell NHL in a phase I study sponsored by Kyowa Kirin Co. Ltd., Tokyo, Japan.

41AMG 319 is a PI3Kδ-specific inhibitor developed by Amgen. AMG 319 was tested in patients with R/R lymphoid malignancies to evaluate its tolerability, safety and PK profile.

42YY-20394/Linperlisib is an oral highly selective PI3Kδ inhibitor which is studied in patients with R/R B-cell malignancies. This drug was developed by Shanghai Ying Li Pharmaceutical Co. Ltd., Shanghai, China.

43INCB050465/Parsaclisib/IBI-376 is a next-generation highly selective potent PI3Kδ inhibitor initially developed by Incyte Corporation, which sponsored a phase I study to evaluate the safety and tolerability of a pan-PIM inhibitor (INCB053914) in combination with Parsaclisib in R/R DLBCL patients.

44There are two active non-recruiting phase I trials investigating Parsaclisib in B-cell lymphoma patients. Another phase I study is testing Parsaclisib with Pembrolizumab in higher-grade solid tumors including lung cancer. In addition, several phase II studies are exploring Parsaclisib as monotherapy in FL patients (CITADEL 203) and in MCL patients (CITADEL-204). Parsaclisib is being tested ± BTK inhibitor (CITADEL 205) and in combination with Bendamustine plus Obinutuzumab in patients with B-cell malignancies (CITADEL 102).

45Umbralisib/TGR-1202/RP5264 is a next-generation highly specific, oral inhibitor targeting the δ isoform of PI3K. There are many trials testing Umbralisib sponsored by TG Therapeutics. One such phase I trial evaluated the safety and efficacy of Umbralisib as a monotherapy or with nab-Paclitaxel plus Gemcitabine or FOLFOX in patients with R/R solid tumors. A phase I study evaluated the efficacy and safety of Umbralisib with Obinutuzumab (anti-CD20 monoclonal antibody) and Chlorambucil (chemotherapy) in CLL patients. Another phase I study tested the efficacy of Umbralisib with Brentuximab vedotin, an anti-body drug conjugate (ADC) against CD30 in patients with HL.

46Phase II/III studies are actively testing Umbralisib as a monotherapy in patients with non-follicular iNHL and with Ublituximab, anti-CD20 monoclonal antibody in combinational therapy in previously untreated CLL patients. A phase I study sponsored by the Vanderbilt-Ingram Cancer Center is currently testing Umbralisib with Ruxolitinib in patients with myeloproliferative neoplasm. Umbralisib has been recently approved by the FDA for treatment of patients with R/R marginal zone lymphoma (MZL) who had at least one prior anti-CD20-based therapy and for adults with R/R follicular lymphoma (FL) relapsed on three prior lines of systemic therapy.

47Tenalisib/RP6530 is a novel, potent PI3Kδ/γ inhibitor with activity more than 300-fold and 100-fold over α and β isoforms, respectively. It was developed by Rhizen Pharmaceuticals and mostly tested against B-cell malignancies. An open-label, phase I/II trial is assessing the safety and efficacy of Tenalisib with a histone deacetylase (HDAC) inhibitor, romidepsin in patients with R/R T-cell lymphoma.

48GDC-0032/Taselisib/RG-7604 is a potent class I PI3K inhibitor having strong activity against the PI3K α, δ and γ isoforms. It exhibits less potency to the ß isoform. There are several ongoing and completed trials using Taselisib in human subjects. Genentech has sponsored all the trials for Taselisib. A phase I study is investigating the safety and tolerability of Taselisib alone in patients with advanced solid tumors or NHL and with Fulvestrant in advanced HER2- and HR+ breast cancer patients. Taselisib is being tested in a phase II study in patients with different cancers (excluding breast) with PIK3CA mutation but without KRAS mutation or PTEN loss. The study is sponsored by the NCI.

49AZD8186 is a specific ß/δ inhibitor, developed by AstraZeneca, which is currently in phase I/II clinical studies.

50AZD8835 is a δ/α-specific PI3K inhibitor, developed by AstraZeneca, with one trial listed so far in patients with solid tumors.

51Duvelisib/IPI-145/INK1197 is an oral highly selective PI3K δ/γ inhibitor with IC50 values of 243 pM and 50 nM, respectively, in cell-free assay. Duvelisib (September 2018, NDA 311155) was approved by the FDA for treatment of patients with R/R CLL/SLL. A multicenter open-label phase I study aimed to investigate the safety and PK of Duvelisib in Japanese subjects with R/R lymphoma was completed in 2017. The study was sponsored by AbbVie in collaboration with Infinity Pharmaceuticals. A phase II study evaluating the clinical activity, long-term safety, and overall survival rate of Duvelisib in patients with hematologic malignancies was recently completed without significant clinical outcomes.

52Leniolisib/CDZ173 is a potent and selective PI3Kδ inhibitor. Most of the studies using Leniolisib are sponsored by Novartis Pharmaceuticals. Although no studies are known in cancer but it is implicated in immunodeficiency disorders such as activated PI3Kδ syndrome/p110δ-activating mutation causing senescent T cells, lymphadenopathy and immunodeficiency (APDS/PASLI) where it has been tested in phase II/III trials. Rao et al. have shown that treatment with Leniolisib resulted in dose-dependent inhibition of PI3Kδ pathway hyperactivation in cells expressing APDS-causative p110δ variants and in T-cell blasts derived from patients. In addition it was shown that oral administration of Leniolisib in six APDS patients resulted in dose-dependent inhibition of the PI3K/AKT pathway evaluated via ex vivo assay and had enhanced immune dysregulation. Thus, Leniolisib is known to be well tolerated and has improved laboratory and clinical parameters in patients with APDS, supporting the specific inhibition of PI3Kδ as a promising new targeted therapy in patients with APDS and other diseases characterized by hyperactivation of the PI3Kδ pathway. Leniolisib is also tested in patients with primary sjögren’s syndrome (PSS) in phase II trials where its oral administration resulted in inhibition of phosphorylated AKT in ex vivo stimulated B cells, decreased the serum CXCL13 levels and reduced the frequency of circulating follicular T helper-like cells.

53Eganelisib/IPI-549 is a potent PI3K-γ inhibitor with >100-fold selectivity with an IC50 value of 16 nM. Eganelisib is the first-in-class, oral, selective PI3K-γ inhibitor identified in preclinical studies which reprogrammed macrophages from an immune-suppressive to an immune-activating phenotype and overcome resistance to checkpoint inhibitors. A phase I study is evaluating the safety, efficacy, PK and PD of Eganelisib in combination with Etrumadenant/AB928 (a selective dual adenosine receptor, A2aR/A2bR antagonist) plus pegylated liposomal doxorubicin (PLD) in patients with TNBC or gynecological tumors. Active not-recruiting and completed trials with specific clinical outcomes are summarized in Table 2.

54Dual PI3K/mTOR inhibitors target both PI3K and mTOR signaling. To date, none of these drugs have received FDA approval for treatment of any cancer. The detailed clinical outcomes of completed and active not-recruiting studies with adverse effects are summarized in Table 3.

55Dactolisib/BEZ235/NVPBEZ235 is a dual pan-class I PI3K inhibitor targeting p110α/ß/γ/δ as well as mTOR kinase inhibitor with IC50 values of 4 nM/5 nM/7 nM/75 nM and 20.7 nM for p110α/γ/δ/β and mTOR, respectively. Dactolisib, developed by Novartis Pharmaceuticals, is known to inhibit both mTORC1 and mTORC2 activity. A phase Ib study sponsored by the Vanderbilt-Ingram Cancer Center investigated the safety and efficacy of Dactolisib with BKM120 in combination with endocrine therapy in post-menopausal patients with HR+ metastatic breast cancer. Another phase I/II study investigated Dactolisib in patients with advanced breast cancer. A phase I trial tested Dactolisib with RAD001/Everolimus in patients with advanced solid tumors without any significant outcomes. Another such phase Ib study evaluated the safety, pharmacokinetics, and pharmacodynamics of Dactolisib with MEK162 in patients with selected advanced solid tumors. A phase Ib/II study of Dactolisib with Trastuzumab versus Lapatinib and Capecitabine was completed in patients with HER2+ locally advanced or metastatic breast cancer who failed on prior Trastuzumab treatment. The trial aimed to find the MTD and the RP2D of this drug combination. Another phase I trial investigated the MTD and RDE of Abiraterone Acetate in combination with Dactolosib and Abiraterone Acetate with BKM120 in CRPC patients. However, despite all these studies, Dactolisib was removed from clinical trials in 2015.

56Apitolisib/GDC-0980/RG7422 is a potent, class I PI3K inhibitor targeting α/β/δ/γ isoforms with IC50 values of 5 nM/27 nM/7 nM/14 nM, respectively, in cell-free assays. Apitolisib also targets mTOR signaling. Genentech Inc. sponsored several phase I/II studies using Apitolisib. A phase Ib, dose-escalation study evaluated the safety, tolerability and PK properties of Apitolisib with Paclitaxel with or without Bevacizumab in patients with locally advanced breast cancer. A phase Ib study assessed the safety and PK properties of Apitolisib in combination with Fluoropyrimidine, Oxaliplatin, and Bevacizumab in patients with advanced solid tumors. A phase I study evaluated Apitolisib with either Paclitaxel and Carboplatin (with or without Bevacizumab) or Pemetrexed and Cisplatin in patients with locally advanced or metastatic solid tumors. Apitolisib was evaluated as a monotherapy in patients with R/R solid tumors or with NHL. The study was sponsored by Genentech Inc. and aimed to assess the safety, tolerability, and pharmacokinetics of escalating doses of Apitolisib.

57Gedatolisib/PF-05212384/PKI-587, developed by Pfizer, is a highly potent dual inhibitor of PI3Kα/γ and mTOR with IC50 values of 0.4 nM, 5.4 nM and 1.6 nM, respectively, in cell-free assays. Details of clinical studies with specific clinical outcomes are summarized in Table 3.

58SF1126, brought into market by SignalRx, is a small-molecule conjugate that selectively inhibits all PI3K class IA isoforms and other members of the PI3K superfamily, including DNA-PK and mTOR. It has been tested in a phase I study in patients with advanced solid tumors and B-cell malignancies. A phase II study sponsored by Semafore Pharmaceuticals in collaboration with SignalRX Pharmaceuticals, Inc. evaluated the safety and tolerability of SF1126 by assessing the DLTs, MTD, and RP2D in patients with advanced or metastatic tumors. A current non-recruiting phase I trial is studying SF1126 with immunotherapy (Nivolumab) in patients with advanced or metastatic hepatocellular carcinoma and Child-Pugh A-B7 cirrhosis.

59Omipalisib/GSK458/GSK2126458 is a reversible, selective, pan-PI3K ATP-competitive inhibitor with a Ki for the catalytic p110α subunit in the subnanomolar range that targets the mTOR pathway. The first-in-human phase I study tested Omipalisib in patients with advanced solid tumors. The study was sponsored by GlaxoSmithkline. A phase I open-label, dose-escalation study evaluated the safety, efficacy, and PK of Omipalisib in patients with solid tumors or lymphoma.

60Samotolisib/LY3023414 is a dual inhibitor of PI3Kα and mTOR with potential anti-neoplastic activity. A phase I trial evaluated the safety of Samotolisib in patients with advanced solid tumors. A phase II study is testing the effectiveness of Samotolisib in patients with endometrial cancer. Two phase I studies sponsored by Eli Lilly and Company evaluated Samotolisib in combination with chemotherapies or other targeted therapies in patients with advanced cancer or lung cancer. In addition, a phase II study tested Samotolisib as monotherapy or in combination with other agents versus standard of care (Gemcitabine or Capecitabine) in patients with metastatic pancreatic ductal adenocarcinoma (PDAC). A phase I study is currently exploring Samotolisib with a CDK4/6 inhibitor LY2835219 (Abemaciclib) in MBC patients. The study is also sponsored by Eli Lilly and Company and aims to investigate the safety of this drug combination.

61Bimiralisib/PQR309 is a potent, oral bioavailable, pan-class I PI3K inhibitor that also targets mTORC1 and mTORC2. A phase II study sponsored by PIQUR Therapeutics AG aimed to assess the safety, tolerability, and the RP2D of Bimiralisib in patients with R/R lymphoma. There are two other phase II studies that are completed in patients with R/R primary central nervous system lymphoma (PCNSL) and advanced lymphoma and HNSCC harboring NOTCH1 loss of function mutations. Another phase I/II study evaluated the safety and pharmacokentic properties of Bimiralisib in combination with chemotherapy, Eribulin in patients with locally advanced or metastatic HER2-negative cancer in the escalation part and in TNBC patients in the expansion part of the study.

62Paxalisib/GDC-0084/RG7666, developed by PIQUR therapeutics, is a potent dual inhibitor of PI3K and mTOR signaling. A phase I study is investigating Paxalisib in pediatric patients with newly diagnosed diffuse intrinsic pontine glioma or diffuse midline gliomas. The study is sponsored by St. Jude Children’s Research Hospital in collaboration with Kazia Therapeutics Ltd., Sydney, Australia which aims to determine the safety, MTD, RP2D, and the pharmacokinetic properties of Paxalisib in the pediatric population. A phase II study is also investigating the efficacy and the pharmacokinetic properties of Paxalisib in glioblastoma patients.

63Voxtalisib/SAR245409/XL765 is a dual inhibitor of PI3K/mTOR which targets the PI3K p110γ, DNA-PK and mTOR pathways. This drug was introduced into market by Sanofi. Sanofi sponsored a phase I/II study testing the safety and tolerability of Voxtalisib as a monotherapy or in combination regimen in patients who were benefiting from this treatment. A phase II study evaluated the efficacy of Voxtalisib with MEK1/2 inhibitor, Pimasertib in patients with ovarian cancer. Most of the studies using this drug are completed and the details of the clinical outcomes with adverse effects are listed in Table 3.

64Resistance to current PI3K therapies continues to push the need to expand our understanding of cancer treatment. Very few PI3K inhibitors are approved by the FDA as PI3K inhibitors suffer from many adverse effects, and poor solubility and permeability. However, these issues are addressed via improved PK techniques. Designing new PI3K inhibitors based on variable structural and binding sites (mutant selective inhibitors) could be used to increase selectivity and reduce toxicity for isoform-specific PI3K inhibitors. Targeting other pathways with ERK, EGFR, CDK4/6, SGLT2 inhibitors, anti-estrogen therapies or HDAC as a combinatorial therapy approach could increase the efficacy for PI3K inhibitors. The mechanism of action which includes the synergistic combination of PI3K with mTOR inhibitors must be explored for better targeting of the PI3K pathway. In this context, Yang et al. have already shown that addition of Dactolisib, a dual PI3K/mTOR inhibitor or ZSTK474, a pan-PI3K inhibitor, in combination with Temsirolimus, a mTORC1 inhibitor, synergistically inhibited cancer cell growth and overcame cellular resistance to Temsirolimus regardless of PTEN status. Emerging evidence in a study by Tang et al. suggested that a combinatorial approach using dual PI3K/mTOR inhibitors could achieve better anti-cancer outcomes. Other factors that could overcome the challenges to PI3K inhibitors in the clinic include selection of accurate patient population with activating PIK3CA mutations, optimizing the dosing schedules, etc. Another challenge in the clinic is a lack of selective PI3Kγ inhibitors. The biggest issue in developing small-molecule inhibitors against PI3Kγ (member of PI3K class IB) is to obtain inhibitors that have good selectivity for PI3Kγ against other PI3K isoforms as PI3Kγ show high sequence similarity in the conserved ATP-kinase-binding site with class I isoforms, leading to difficulties in developing specific γ inhibitors. Further research is needed to understand whether using a dual or multi-targeted approach inhibiting two or more PI3K enzymes or direct inhibition of PI3Kγ could help to achieve maximal efficacy against these inhibitors. In addition, there are several new P13K drugs such as RIDR-PI-103, B591 and TG-100–115 with limited in vitro and in vivo studies which need to be further explored using PK and PD models so that it could be translated to clinic.