Biomarkers of everolimus efficacy in breast cancer therapy

Abstract

Objective

Everolimus is an inhibitor of serine/ threonine kinase mTOR. The drug is approved for the treatment of metastatic ER positive, HER2 negative breast cancers and benefits a subset of patients with these breast cancers in combination with hormonal therapies. Despite extensive efforts, no additional predictive biomarkers to guide therapeutic decisions for everolimus have been introduced in clinical practice.

Data sources

This paper discusses predictive biomarkers for everolimus efficacy in breast cancer. A search of the medline and web of science databases was performed using the words “everolimus” and “biomarkers”. References of retrieved articles were manually scanned for additional relevant articles.

Data Summary

Everolimus benefits a subset of patients with metastatic ER positive, HER2 negative breast cancers in combination with hormonal therapies. Despite extensive efforts no additional predictive biomarkers to guide therapeutic decisions for everolimus therapy have been confirmed for use in clinical practice. However, promising biomarker leads for everolimus efficacy in breast cancer have been suggested and include expression of proteins in the mTOR pathway in ER positive, HER2 negative breast cancers. In HER2 positive cancers PIK3CA mutations, and PTEN expression loss are prognostic. Other clinical predictive biomarkers with more limited data include characteristics derived from whole genome sequencing, subsets of circulating leukocytes and changes in Standardized Uptake Values (SUV) of Positron Emission Tomography (PET) scans.

Conclusions

Putative predictive biomarkers for everolimus efficacy in breast cancer patients, both genomic and clinical, deserve further study and could lead to a better selection of responsive patients.

Keywords

Rapalogs predictive biomarkers next generation sequencing PI3 k metabolism

Introduction

Breast cancer represents the most prevalent female carcinoma and despite advances in treatment that have led to improved survival outcomes, it remains a disease with significant morbidity and mortality.¹ In addition, metastatic disease, albeit treatable remains incurable. A better characterization of sub-types of breast cancer has provided a classification into groups with therapeutic implications. Clinical biomarkers used for sub-type classification, including the Estrogen Receptor (ER) and the HER2 receptor (Human EGFR family Receptor 2), serve also as predictive biomarkers to targeted therapies.² Beyond therapies that inhibit these receptors, other targeted therapies have been introduced and found their place for the treatment of sub-sets of breast cancers. These include CDK kinase inhibitors and PI3 K kinase inhibitors for the treatment of metastatic ER positive breast cancers, PARP inhibitors for cancers with BRCA1 or BRCA2 mutations and immune checkpoint inhibitors for cancers with microsatellite instability.^3–6 Several other targeted therapies are in development.⁷

The mTOR kinase inhibitor everolimus is another targeted therapy that is indicated in metastatic ER positive, HER2 negative breast cancers in combination with hormonal therapy (Figure 1). It was the first targeted therapy to be introduced in this most common sub-type of breast cancer, after progression on hormonal therapy alone.⁸ Everolimus is indicated in combination with hormonal therapy in patients with ER positive, HER2 negative metastatic breast cancer irrespective of other molecular defects. Since the introduction of everolimus, other targeted therapies including CDK kinase inhibitors and PI3 K kinase inhibitors have been added in the metastatic ER positive, HER2 negative breast cancer armamentarium. The latter is indicated even as first line in newly diagnosed or newly progressed patients and the latter has an indication specifically for the subset of ER positive, HER2 negative patients with mutations in the target PI3 K kinase.⁹ As a result, everolimus has often been delegated to second line treatment in breast cancer patients with no PI3 K mutations progressing on hormonal therapy and CDK inhibitors combinations or third line treatment for patients with PI3 K mutations. In these settings everolimus competes with chemotherapy, as some physicians consider patients progressing on one type of hormonal therapy combination to be at higher risk for resistance to other hormonal therapy combinations.¹⁰ In later line of treatment, the effectiveness of everolimus is not well documented by randomized trials but it would be expected to be lower than the effectiveness in patients not exposed to other targeted agents, given that molecular alterations may confer cross-resistance. However, real world retrospective data suggest that hormonal therapy with or without everolimus retain activity after progression on hormonal therapies with CDK inhibitors.¹¹ The combination of exemestane with everolimus as a second line or third line treatments following hormonal therapy and CDK inhibitors is associated with shorter times to next treatment compared with exemestane and everolimus treatment following hormonal therapy alone.¹²

A better characterization of patient groups with tumors more sensitive to everolimus treatments would facilitate the reintegration of the drug in earlier treatment of ER positive, HER2 negative metastatic breast cancer and rekindle earlier efforts to introduce everolimus in the therapy of other breast cancer sub-types. This paper discusses predictive factors for everolimus efficacy in breast cancer with the goal to reposition its use in patients more likely to benefit.

The PI3 K/ AKT/ mTOR pathway and molecular defects in breast cancers

The PI3 K/ AKT/ mTOR pathway starts with phosphoinositide 3-kinase (PI3 K), receiving signals of activation from various receptor tyrosine kinases such as the EGFR family kinases, VEGFR family kinases, Insulin/ Insulin-like Growth Factor Receptors, c-MET and others.¹³ PI3 K phosphorylates the lipid phosphatidylinositol-4,5-biphosphate (PIP2) at position 3 to produce phosphatidylinositol-3,4,5-triphosphate. PIP3 in membranes is a docking site for proteins with a pleckstrin homology domain such as kinase AKT.¹³ In membranes, AKT undergoes 2 phosphorylations for activation by kinase PDK1 and by kinase mTOR as part of the mTORC2 complex. Then, AKT phosphorylates and regulates several downstream substrates including mTORC1, with a role in translation and cell growth regulation as well as Glycogen Synthase Kinase 3 (GSK3) (Figure 2). GSK3 is a negative regulator of the WNT/ APC/ beta-catenin pathway and of c-myc and is negatively regulated by AKT phosphorylation. mTORC1 activation by AKT kinase is indirect and is mediated through inhibition of the TSC1/ TSC2 (Tuberous Sclerosis 1 and 2) complex, a negative regulator of small GTPase Rheb (RAS Homologous Enriched in Brain). Rheb is the direct activator of mTORC1 (Figure 2). AKT also inhibits transcription factors of the FOXO family which have an anti-proliferative and pro-apoptotic function. The ubiquitin ligase MDM2, a main post-translational negative regulator of p53, is another target of AKT. MDM2 is activated by AKT phosphorylation. Closely interrelated to the PI3K-AKT cascade runs the KRAS/ RAF/ MEK/ ERK pathway, also activated by receptor tyrosine kinases. KRAS/ RAF/ MEK/ ERK pathway cross-talks at multiple levels with the PI3 K/ AKT/ mTOR pathway.¹³

mTOR is a serine/ threonine kinase of the PI3 K related kinase family.¹⁴ The kinase domain of mTOR is located on the carboxyterminal part of the protein. The FRB (FKBP12- Rapamycin Binding) domain through which mTOR interacts with the FKBP12 (FK506 Binding Protein 12)- Rapamycin complex is located just aminoterminally to the kinase domain. mTOR is constitutively active and its catalytic activity is regulated by access of its substrates to the kinase cleft which is restricted by FRB domain and an additional inhibitory helix.¹⁵ Binding of immunophilin FKBP12, but not of FKBP12 in complex with rapalogs, facilitates access of substrates to the active enzyme site. In mTORC1 complex, mTOR interacts with the scaffold protein Raptor (Regulatory protein associated with mTOR) through its HEAT repeats domain occupying the aminterminal part of the mTOR protein. In mTORC2, mTOR interacts with adaptor protein Rictor (Rapamycin insensitive companion of mTOR). In both complexes the kinase domain of mTOR binds mLST8 (mammalian Lethal with Sec13 protein 8), a protein that plays a facilitator role in kinase activation. The two complexes that utilize mTOR as the catalytic protein, mTORC1 and mTORC2 have distinct conformations and only mTORC1 is inhibited by rapalogs, including everolimus.¹⁴

Besides activation by AKT kinase, mTORC1 may be regulated by the LKB1 (Liver Kinase B1, also known as STK11- Serine Threonine Kinase 11)/ AMPK1 (AMP activated Kinase 1) cascade (Figure 2). The cell energy status is sensed by AMPK kinase through the intracellular levels of AMP. AMPK is activated by high AMP levels corresponding to energy depletion and after phosphorylation by LKB1 leads to mTORC1 inhibition for adjustment of protein production.¹⁶ AMPK inhibits mTORC1 directly through Raptor and indirectly through TSC1/ TSC2/ Rheb. Supply of cellular aminoacids also regulates mTORC1 and is sensed through interaction of mTORC1 with the RagA/ RagC/ Ragulator complex on lysosome cell membranes. The RagA/ RagC/ Ragulator complex receives input from various adaptor proteins that scavenge the concentrations of aminoacids methionine, arginine, and glutamine.¹⁴

mTORC1 activation leads to regulation of several targets. The prototypic targets that have been discussed extensively in the literature are kinase p70S6 K and the protein 4EBP1 (Eukaryotic transcription initiation factor 4E binding protein 1), both involved in 5’ cap-dependent translation. 4EBP1 is an inhibitor of translation initiation factor eIF4E and is inhibited by mTORC1 phosphorylation. Two isoforms of p70S6 K, S6K1 and S6K2 exist, have high sequence homology and are both targets of mTORC1.¹⁷ p70S6 K is activated by mTORC1 and activates translation by positively regulating initiation factor eIF4B directly, and indirectly by destabilizing the negative regulator of eIF4B, PDCD4 (Programmed Cell Death 4).¹⁸ In addition, p70S6 K promotes lipid synthesis by regulating transcription factor SREBP (Sterol Responsive Element Binding Protein) and nucleotide synthesis by activation of the de novo pyrimidine synthesis enzyme carbamoyl-phosphate synthetase.¹⁹ SREBP is also involved in glycose metabolism. p70S6 K regulates glycose metabolism by activating transcription factor HIF1α (Hypoxia Inducible Factor 1α). mTORC1 further promotes cell anabolism by inhibition of the catabolic process of autophagy through negative regulation of autophagy factors ULK1 (Unc-51 Like autophagy activating Kinase 1) and ATG14L (Autophagy related 14- Like).²⁰

The immunophilin family member FKBP12, a peptidyl-prolyl isomerase, is the primary cellular ligand of rapalogs such as rapamycin (also called sirolimus), temsirolimus and everolimus.²¹ FKBP12 interacts with another immunosuppressive drug, FK506 (also known as tacrolimus), but in this case the complex binds calcineurin instead of mTOR.²² The FKBP12-rapalogs complex binds the FRB domain of mTOR and inhibits its kinase activity by restricting access of substrates to the kinase pocket of the enzyme. Other members of the immunophilin family such as FKBP51 and FKBP52 bind rapalogs and can inhibit mTOR.²³ FKBP12 and other immunophilins are involved in TGFβ and ryanodine receptors signaling as well as NF-κB signaling. Thus, rapalogs may in fact have broader pharmacologic effects and may affect a broad array of cellular processes beyond the effects mediated by inhibition of mTOR pathways, through modulating immunophilins.

Rapalogs display a differential capacity for inhibition of the various substrates of mTORC1, consistent with their allosteric mechanism of action that depends on interference of substrates access to the active kinase site. Differential sensitivity relates to a differential ability of mTOR to phosphorylate various physiologic substrates.²⁴ Phosphorylation of p70S6 K is effectively inhibited by these drugs, while persistent 4EBP1 phosphorylation is insensitive to rapalogs.²⁵ Insensitivity of 4EBP1 inhibition may be related to the fact that KRAS/RAF/ MEK cascade which is unaffected by rapalogs play a role in 4EBP1 phosphorylation through kinase ERK.²⁶ From a predictive biomarker point of view, insensitivity of 4EBP1 to rapalogs inhibition suggests that the ratio of eIF4E, the target of 4EBP1, to 4EBP1, that has been shown to be a predictor of sensitivity to mTOR kinase inhibitors would not be a predictor of rapalogs response.²⁷

Besides mutations in the gene encoding for the alpha catalytic unit of kinase PI3 K, PIK3CA, which is mutated in about one third of cases, other genes of the mTORC1 complex or the immediate upstream pathway that regulates it are not commonly mutated in breast cancers (Figure 3A).^28,29 Besides PIK3CA, the most commonly mutated genes of the pathway in the breast cancer TCGA series include PTEN in 5.4% of cases, NF1 in 3.6% of cases and AKT1 in 2.5% of cases (Figure 3A). Copy number alterations of genes in the pathway include amplifications of DEPTOR in 12.4% of cases, amplifications of LAMTOR2, encoding for Ragulator, in 7.9% of cases, amplifications of TSC2 and of MLST8 both occurring in 4.2% of cases and of RPTOR occurring in 3.8% of breast cancer cases in TCGA (Figure 3B). Deletions in genes of the pathway are rare except for deletions of PTEN that occur in 5.1% of breast cancers.

Molecular biomarkers from the PI3 K/ AKT/ mTOR pathway

Clinical studies of everolimus in breast cancer have incorporated biomarker exploration to define subsets of patients who are more responsive and derive greater benefit from addition of the drug in their treatment plan.³⁰ The randomized phase II TAMRAD trial of the French GINECO group compared everolimus plus tamoxifen with tamoxifen alone in 111 ER positive, HER2 negative metastatic breast cancer patients who had been previously treated with aromatase inhibitors.³¹ The study showed a prolongation of time to progression from 4.5 months with tamoxifen monotherapy to 8.6 months with the combination. Fifty-five patients from the TAMRAD trial were included in a biomarker study that investigated mutations in PIK3CA gene and expression of proteins of the PI3 K/ AKT/ mTOR pathway as predictors of everolimus response (Table 1).³² PIK3CA mutations were mostly located at the kinase domain and did not appear to be associated with response to everolimus, as assessed by time to progression, although the number of patients with mutations was small. Immunohistochemistry staining for proteins of the PI3 K/ AKT/ mTOR pathway showed that low levels of 4EBP1 protein or high levels of the phosphorylated form p4EBP1 and low levels of kinases LKB1 and PI3 K and of the phosphorylated form of kinase AKT were associated with greater benefit from addition of everolimus to tamoxifen.³²

Table 1.

Predictive biomarkers of everolimus efficacy and molecular lesions found not to be predictive of everolimus benefit in clinical studies. CIN: Chromosomal Instability.

Study	Number of patients	Biomarkers of response	Not predictive
TAMRAD³²	55	Low expression of 4EBP1, pAKT, LKB1 and PI3 K, high level of expression of p4EBP1	PIK3CA mutations
BOLERO 2³³	302	Low CIN	PIK3CA mutations, pathway activation, CCND1, CDK4, CDK6, CDKN2A alterations
Kruger et al.³⁴	145	High expression of p4EBP1 from metastatic site associated with worse PFS	pAKT, 4EBP1, p70S6 K, S6RP expressions
BOLERO 1 and 3³⁵	549	PIK3CA mutations, loss of PTEN expression, pathway activation
Jovanovic et al.³⁶	145	None	Phosphorylated S6 protein

BOLERO 2 was a randomized phase III trial in patients with ER positive, HER2 negative metastatic breast cancer that compared the steroidal aromatase inhibitor exemestane as monotherapy with the combination of exemestane and everolimus.⁸ Patients were randomized in a 2 to 1 ratio to the combination and monotherapy arms, respectively. The trial included 724 post-menopausal patients previously treated with a non-steroidal aromatase inhibitor in the adjuvant or metastatic setting. Half of the patients had also previously received tamoxifen and about two thirds had received chemotherapy. Results showed that the combination of exemestane with everolimus prolonged PFS from 2.8 months in the monotherapy arm to 6.9 months, but not OS, that was a secondary end point in the trial.^8,37 BOLERO 2 investigators presented biomarker data for prediction of efficacy of everolimus in a subset of 302 patients who participated in the study.³³ Almost half of the patients (47.6%) had mutations in PIK3CA, mostly in the known hotspot sites in the helical and catalytic kinase domains. Prevalence of these mutations was similar independently of whether the tested specimen was from the primary tumor or a metastatic site. Benefit from addition of everolimus to exemestane was observed independently of the presence of PIK3CA mutations (Table 1). Benefit seemed to be more pronounced in patients with exon 9, helical domain mutations.³³ Analyses of PIK3CA mutations through droplet digital PCR of plasma derived cell-free DNA of patients from the BOLERO2 study confirmed a similar benefit of everolimus independently of the presence or absence of hotspot PIK3CA mutations.³⁸ Everolimus benefit was also present independently of whether the PI3 K/ AKT/ mTOR pathway was activated, defined as at least one mutation of PIK3CA, PTEN, AKT1, PIK3R1 (the gene encoding for the regulatory sub-unit of PI3 K) or decreased expression of PTEN with a H-score of less than 10 by immunohistochemistry. Patients with ER positive, HER2 negative breast cancers have also common amplifications of cell cycle regulator cyclin D and alterations of its gene CCND1 were observed in 31.3% of examined cases in BOLERO 2. However, benefit of everolimus was independent of alterations in CCND1 and of other cell cycle regulators such as CDK4, CDK6 and CDKN2A.³³

In a series of 145 ER positive, HER2 negative metastatic breast cancer who participated in a study of exemestane with everolimus therapy, expression of phosphorylated forms of five proteins (2 phosphorylated forms of AKT at positions T308 and S473, 4EBP1, p70S6 K, and S6RP) of the PI3 K/ AKT/ mTOR pathway by immunohistochemistry of primary tumor samples did not correlate with treatment benefit either individually or when examined together grouped in an activated pathway group and a normal group.³⁴ In contrast, in a sub-group of 21 patients participating in the study who had material from a biopsy of a metastatic site done just before the start of the exemestane/ everolimus therapy, patients with high phosphorylated 4EBP1 (p4EBP1) expression had worse PFS compared with patients with low p4EBP1. Since all patients in this study received exemestane with everolimus, it is not possible to deduce from the data whether p4EBP1 expression is a predictive biomarker of therapy efficacy or a prognostic biomarker.

Mutations on the gene encoding for mTOR kinase, MTOR and upstream regulator TSC1 are rare in breast cancers and are encountered in 1.9% and 0.8% of cases, respectively, in TCGA cohort.²⁸ Case studies in other cancers have evaluated the role of MTOR and TSC1 mutations in rapalogs response in long-term responders.^39,40 A genomic analysis of a patient with bladder cancer who responded to everolimus for more than 2 years disclosed truncated mutations in TSC1, as well as in tumor suppressor NF1.³⁹ Three additional bladder cancer patients with TSC1 mutations were identified by the same investigators and showed responses to the drug, albeit more short-lived. Another report on five renal cell carcinoma patients with long responses to temsirolimus or everolimus found mutations in TSC1 or MTOR in three of them.⁴⁰ A case series of various non-breast cancer patients described clinical benefit from everolimus in patients with MTOR, TSC1 and TSC2 mutations.⁴¹ Another long-term responder of more than a year with metastatic bladder carcinoma was identified in a phase 1 study employing everolimus with pazopanib.⁴² The tumor of this patient harbored a double mutation in MTOR. Both mutations were in the kinase domain and resulted in increased mTOR kinase activity in vitro, with an additive activation effect when both mutations were concomitantly present.⁴² Long-term responders have also been reported in ER positive, HER2 negative breast cancer.⁴³ In the BOLERO2 trial, 10 patients with MTOR mutations were found and seem to derive benefit from everolimus.³³ However, small numbers precluded a formal statistical analysis. In a small series of ER positive metastatic breast cancer patients treated with everolimus and hormonal therapies, patients with absent staining for mTOR seemed to have prolonged PFS on therapy compared with patients who had weak or moderate mTOR expression.⁴⁴

Another investigation described three metastatic ER positive, HER2 negative breast cancer patients who derived a long-term benefit from treatment with anastrozole and everolimus.⁴⁵ All three had multiple previous lines of therapy, including chemotherapy, and their response duration with anastrozole and everolimus exceeded response duration with previous therapies. Alterations in the PI3 K/ AKT/ mTOR pathway were discovered in all three patients and included PTEN loss by immunohistochemistry, mutations in PIK3CA and the regulatory subunit gene PIK3R1 and amplification in the adaptor protein IRS2 gene. In addition, two patients had amplifications of the CCDN1 and FGFR1 loci at chromosomes 11q13 and 8p11 that include the genes for the target proteins of mTOR, p70S6 and 4E-BP1, respectively.⁴⁶ Of note, all three described patients were younger than 50 years-old and their menopause status or ovarian suppression therapy is not described in the report. Thus, they may have received sub-optimal hormonal therapy. Mutations in kinase AKT1 gene has also been described as a molecular abnormality associated with response to anastrozole and everolimus.⁴⁷

Mutations in the mTOR regulating kinase STK11, the cause of hereditary Peutz-Jeghers syndrome, have been reported in relation to everolimus response. A patient with biphenotypic ER positive and triple negative metastatic breast cancer was found to have a F354L STK11 mutation with loss of heterozygosity in the triple negative component.⁴⁸ The patient obtained an ongoing almost complete response with the combination of exemestane and everolimus. In another report, a patient with pancreatic cancer on a background of Peutz-Jeghers syndrome was treated with everolimus and obtained a partial response lasting for 9 months.⁴⁹ In contrast, a similar patient with Peutz-Jeghers syndrome and pancreatic cancer treated with everolimus showed progressive disease.⁵⁰ STK11 mutations have been described in Female Adnexal Tumors of probable Wolffian Origin (FATWO), a rare histologic type of adnexal tumors and a recent report described treatment with everolimus and anastrozole in a patient with recurrent FATWO.⁵¹ Treatment produced a partial response sustained for more than a year. These anecdotal data are not conclusive but may suggest that molecular lesions in STK11 contribute to everolimus sensitivity, while the specific tumor microenvironment needs be taken into consideration.

Amplifications of members of the FGF (Fibroblast Growth Factor)/ FGFR ligand/ receptor tyrosine kinase pairs were examined in association with aberrations of the PI3 K/ AKT/ mTOR pathway as markers of response to drugs targeting the PI3 K/ AKT/ mTOR pathway.⁵² Among a series of 112 metastatic breast cancer patients studied by Next Generation Sequencing (NGS), 15 patients (13.4%) had amplifications in one or more FGF/ FGFR family members and concomitant alterations of PI3 K/ AKT/ mTOR pathway members, including PIK3CA mutations and PTEN loss, as the most prevalent alterations. Most patients were ER positive, HER2 negative and a few were triple negative. When treated with inhibitors of the PI3 K/ AKT/ mTOR pathway, 11 patients with concomitant FGF/ FGFR and PI3 K/ AKT/ mTOR pathway aberrations had higher clinical benefit rate, defined as complete response, partial response or stable disease lasting for more than 6 months, than patients with PI3 K/ AKT/ mTOR pathway alterations alone (73% vs. 34%). Patients with concomitant aberrations of the two pathways remained also longer on treatment with PI3 K/ AKT/ mTOR pathway inhibitors than counterparts bearing only PI3 K/ AKT/ mTOR pathway alterations.⁵² Unfortunately, the report did not clarify the specific PI3 K/ AKT/ mTOR inhibitor drugs used.⁵²

DYRK2 (Dual specificity tyrosine phosphorylated regulated kinase) is a kinase that phosphorylates mTOR facilitating subsequent ubiquitination for protein degradation.⁵³ Xenografts of breast cancer cells which had DYRK2 knocked down by shRNA were inhibited by everolimus treatment to a greater degree than xenografts of control cells. In addition, metastatic ER positive breast cancer patients with low DYRK2 tumor expression had longer responses to everolimus treatment than counterparts with high expression of DYRK2.⁵³

Besides ER positive, HER2 negative breast cancers, everolimus was investigated in trials of metastatic HER2 positive disease in the first line and second line settings in combinations with chemotherapy and trastuzumab.^54,55 BOLERO 1 trial randomized 719 HER2 positive patients who had not received previous therapy for metastatic disease to paclitaxel, trastuzumab, with or without everolimus. The addition of everolimus did not prolong PFS in these patients. However, in the ER negative subset, everolimus prolonged PFS by 7.2 months, albeit just short of statistical significance as prespecified in the trial protocol.⁵⁴ BOLERO 3 randomized 569 pretreated patients to vinorelbine and trastuzumab, with or without everolimus. In this population, the addition of everolimus prolonged PFS from a median of 5.8 months in the placebo arm to a median of 7 months, which was statistically significant.⁵⁵ Biomarker evaluation from the combined BOLERO1 and 3 trials from a total of 549 patients who had participated in the 2 trials was presented by the authors of the two studies.³⁵ A benefit from the addition of everolimus to chemotherapy and trastuzumab was detected in patients with PIK3CA mutations, loss of PTEN protein expression by immunohistochemistry or hyper-activation of the PI3 K pathway defined as mutations of PIK3CA, loss of expression/ mutations of PTEN or mutations in AKT1 (Table 1).³⁵ In contrast, the group of patients without these molecular abnormalities showed similar PFS independently of the addition of everolimus in their therapies. The subgroup of hormone receptor negative patients and presence of PIK3CA mutations, loss of PTEN protein expression by immunohistochemistry or hyper-activation of the PI3 K pathway derived benefit from everolimus similarly to the whole HER2 positive population, while patients with hormone receptor negative tumors but without PIK3CA mutations, loss of PTEN protein expression by immunohistochemistry or hyper-activation of the PI3 K pathway did not benefit from the addition of everolimus. For hormone receptor positive patients, a benefit from everolimus was observed only in patients with PTEN protein expression loss.³⁵

In triple negative breast cancer, a study that evaluated the addition of everolimus to neo-adjuvant chemotherapy with weekly cisplatin and paclitaxel for 12 weeks disclosed no benefit in pathologic complete response by the addition of everolimus.³⁶ Inhibition of mTOR by everolimus was confirmed by decreased phosphorylation of S6 protein by immunohistochemistry but did not correlate with response to therapy (Table 1). Higher responses in both arms were observed in patients with high Ki67 proliferation marker, with alterations in DNA damage response genes and with absence of expression of the Androgen Receptor (AR).

Molecular biomarkers outside the core genes of the PI3 K/ AKT/ mTOR pathway and whole genome biomarkers

Molecular biomarkers that are indirectly related to the PI3 K/ AKT/ mTOR pathway or that are related to general characteristics of the cancer cells, such as proliferation rate have also been examined as predictors of everolimus clinical response. In a small series of metastatic ER positive, HER2 negative breast cancer patients, those patients with a low Ki-67 proliferation index displayed a longer mean PFS (109 weeks), while high Ki-67 was associated with shorter mean PFS of 19 weeks.⁵⁶

Expression of c-MET, the receptor for Hepatocyte Growth Factor, was associated with breast cancers with high grade and resistance to everolimus compared with cancers with low expression of c-MET by immunohistochemistry.⁵⁷ An individual patient that was studied at baseline before everolimus treatment and at progression, was found to develop c-MET upregulation and responded to the c-MET inhibitor cabozantinib.

ESR1 mutations at hotspots Y537 and D538, leading to constitutive activation of the Estrogen Receptor, were tested by droplet digital PCR in cell free plasma DNA from a sub-set of 541 patients that participated in the BOLERO 2 trial.⁵⁸ As the trial included patients that had progressed on non-steroidal aromatase inhibitors, ESR1 mutations prevalence was high (28.8%). Although presence of ESR1 mutations were prognostic for worse OS (median OS 20.7 months in patients with mutations vs. median OS of 32.1 in patients who were wild type for ESR1), they were not predictive for everolimus benefit. However, when analyzed by specific site of mutation, there was a suggestion that while patients with ESR1 D538G mutations derived benefit from the addition of everolimus, similarly to wild type patients, patients with Y537S mutations or mutations in both sites did not benefit from the addition of everolimus to exemestane.⁵⁸ These data suggest that reversal of resistance by mTOR inhibition depends on the specific mutant ER protein.

Besides biomarkers based on individual genes and proteins, attempts for prediction of everolimus efficacy from genomic evaluation showed that breast cancer genomic sub-types respond differently. Genomic data from a sub-set of patients from the BOLERO 2 trial disclosed that, among ER positive, HER2 negative metastatic breast cancer patients, about one fourth are not luminal and show worse survival outcomes compared with luminal patients.⁵⁹ Most non-luminal cancers are HER2-enriched and only rarely they are basal-like. Both luminal and HER2-enriched patients obtained benefit from the addition of everolimus to exemestane.⁵⁹ The benefit was somewhat greater in luminal cancers (hazard ratio for PFS of everolimus vs. placebo 0.37) than in non-luminal cancers (hazard ratio for PFS of everolimus vs. placebo 0.47).

Chromosomal Instability (CIN) was also examined in BOLERO 2 patients as a biomarker of everolimus benefit.³³ CIN was measured by a CIN score derived from genomic data by calculation taking into consideration rearrangements and copy number alterations. Patients were classified as high CIN if they were in the highest quartile of the CIN score. Benefit of everolimus addition to exemestane was observed in the low CIN score group patients who had a prolongation of their mean PFS from 2.8 months to 8.4 months but not in the high CIN group where mean PFS was 4.1 months with monotherapy and 5.6 months with exemestane and everolimus combination (Table 1).³³

Clinical biomarkers

Clinical and radiologic predictors of everolimus response are of potential interest as they are non-invasive and may use information already available to the clinician without the need of additional studies with the associated dedicated resources and cost. Given the central role of mTOR and the PI3 K/ AKT/ mTOR pathway in metabolism, biomarkers such as Body Mass Index (BMI) and fasting glycose have attracted attention.^60,61 In a study of 102 postmenopausal ER positive, HER2 negative metastatic breast cancer patients treated with exemestane and everolimus, lower fasting glycose (less than 107 mg/ dL) at the time of best response was associated with a longer PFS.⁶⁰ In contrast, BMI was not associated with outcomes. In a phase IIIb expanded access study of postmenopausal ER positive metastatic breast cancer patients who received exemestane and everolimus, BMI at baseline was not associated with PFS.⁶¹ However, changes of BMI during therapy affected outcomes, with patients losing weight in the first 4 weeks of therapy having worse PFS and patients having lost weight at the end of therapy having a better PFS than patients who did not lose weight. It is unclear from the study whether early weight loss in non-responders was merely a biomarker of progressive disease and worsening general status or a genuine metabolic marker, pathophysiologically related to benefit from mTOR inhibition.

Peripheral blood lymphocyte subsets were examined in a report that used data from a neoadjuvant everolimus trial.⁶² Responders to everolimus had a higher number of circulating T cells as well as CD4 + and CD8 + T cells compared to patients not responding to the drug. In contrast, numbers of circulating B cells and NK cells were not different between everolimus responders and non-responders and the number of T regulatory lymphocytes with the phenotype CD4 + / CD25 + / CD157- showed only a trend for lower numbers in responders that did not reach statistical significance (p = 0.07).⁶² Data from a sub-set of patients from the phase IIIb trial mentioned in the previous paragraph disclosed that the neutrophil to lymphocyte ratio in peripheral blood was predictive of outcomes to treatment with exemestane and everolimus.^62,63 Patients with a neutrophil to lymphocyte ratio below 4.4 had a superior PFS compared with patients with baseline neutrophil to lymphocyte ratio above 4.4 (Log Rank test p = 0.01).⁶²

Changes in Standardized Uptake Values (SUV) of Positron Emission Tomography (PET) scans at 3 months compared with baseline SUV were examined in a small study of 31 patients with metastatic ER positive, HER2 negative breast cancer receiving exemestane and everolimus.⁶⁴ A change of SUV by more than 28.8% was found to be an optimal cut-off to predict PFS at 10 months. Patients with PET SUV changes above this cut-off point had a 10-month PFS of 63.2%, while patients with PET SUV changes below this cut-off point had a 10-month PFS of 16.7%. In addition, change of SUV by more than 53.8% was found to be an optimal cut-off to predict longer OS beyond 3 years. Patients with PET SUV decreases of more than 53.8% at 3 months had a 3-year OS of 82.5%, while patients with PET SUV changes below this cut-off point had a 3-year OS of 45.9%.⁶⁴ Although PET responses are promising as a biomarker of everolimus treatment, they require a treatment period and are not genuine pre-treatment predictors. In addition, both PET scans and peripheral blood lymphocyte sub-sets examinations are not a standard part of breast cancer patient response evaluation and will require additional costs and resources if they were to be incorporated in clinical prediction of response.

Potential biomarkers derived from in vitro studies

To determine molecular characteristics that could be associated with everolimus response, an extensive panel of breast cancer cell lines was examined in vitro.⁶⁵ Most everolimus sensitive cell lines (77%) were ER positive. However, only 38% of ER positive cell lines examined were sensitive to everolimus. A similar percentage (31%) of HER2 positive cell lines were sensitive to everolimus, while only 13% of basal-like/ post-EMT (Epithelial-Mesenchymal Transition) cell lines were sensitive to the drug. Among specific molecular aberrations, amplifications of ERBB2 gene, encoding for HER2 and of AURKA gene, encoding for Aurora Kinase A were associated with everolimus sensitivity.⁶⁵ These results confirm that ER and HER2 positive breast cancers are most sensitive to mTOR inhibition, albeit with intra-subtype heterogeneity.

Cell lines of various cancers origin bearing activating mutations in MTOR were sensitive to everolimus in vitro.⁶⁶ Cell lines with non-activating MTOR mutations showed no sensitivity. Activating mutations are observed in the carboxyterminal part of the protein, including the FRB and kinase domains and correspond to mutations that had been observed to offer everolimus sensitivity in some patients in vivo.⁶⁶

The activity of eukaryotic translation initiating factor eIF4E, a down-stream target of mTOR pathway was associated with sensitivity of breast, colorectal and lung cancer cell lines to everolimus and rapamycin.⁶⁷ Cell lines with higher eIF4E activity as measured by an in vitro translation assay were more sensitive to the drugs, implying that higher eIF4E activity is associated with addiction of cancer cells to the pathway, in vitro. In contrast, results in vivo from breast cancer patients treated with everolimus in the neoadjuvant setting showed that higher eIF4E expression in pre-treatment biopsies was not associated with better responses to therapy as measured by decrease of cell proliferation index from diagnosis to surgery.⁶⁷ In addition, everolimus treatment resulted in significant variability in expression of eIF4E and 4EBPs suggesting that the lack of response in vivo is associated with feedback adjustments in the pathway status. These observations are also consistent with the fact that 4EBP1 phosphorylation is not effectively inhibited by rapalogs.

A study of breast cancer cell lines isogenic for activating hotspot PIK3CA mutations at positions H1047 and E545 showed that cells with these mutations become sensitive to everolimus treatment in vitro and in mouse xenograft models in vivo.⁶⁸ Introduction of additional mutations in KRAS or BRAF reverses everolimus sensitivity and leads to treatment resistance. Moreover, similar results were observed in a group of 43 patients receiving everolimus, in whom mutations in PIK3CA or PTEN were associated with benefit from treatment but presence of concomitant mutations in KRAS or BRAF led to a lack of everolimus benefit.⁶⁸ In addition, as mentioned in a previous section, PIK3CA mutations have not been found to predict benefit from the combination of everolimus with exemestane in ER positive, HER2 negative metastatic breast cancer patients, progressing on hormonal monotherapy and participating in the BOLERO 2 study, implying that in the setting of hormone refractory ER positive, HER2 negative metastatic breast cancer, everolimus can resensitize cancer cells to hormonal therapy, independently of whether the cause of resistance involves PI3 K.³³

In triple negative breast cancer cell lines, everolimus was observed to inhibit the sub-set with a basal-like phenotype, in vitro and in mice xenografts in vivo.⁶⁹ In contrast the subset of triple negative breast cancer cell lines with epithelial to mesenchymal transition (EMT) phenotype, characterized by decreased E-cadherin expression and higher expression of the EMT core transcription factors Snail and Twist were less sensitive.⁶⁹

Perspective

Current evidence on biomarkers from randomized trials in ER positive, HER2 negative breast cancers, as described in the previous sections, do not support a role of the common PIK3CA mutations as predictive biomarkers of everolimus response.³³ In contrast, a role of protein expression of 4EBP1, pAKT, LKB1 and PI3 K as detected by immunohistochemistry is suggested by the TAMRAD study.³² PIK3CA mutations, PTEN expression loss, and PI3 K/ AKT/ mTOR pathway activation are suggested to have predictive value in HER2 positive cancers.³⁵ Anecdotal data from case series and case reports are hypothesis generating by their very nature. Molecular lesions in mTOR itself or the immediate regulators TSC1 and TSC2 could be markers of sensitivity but are rare in breast cancer, making prospective evaluation difficult or infeasible.

Clinical benefit observed with PI3 K/ AKT/ mTOR pathway inhibitors in breast cancer patients bearing concomitant FGF/ FGFR amplifications and PI3 K/ AKT/ mTOR pathway aberrations is intriguing. FGFR1 gene is located on the 8p11.23 amplicon that also includes the gene encoding for the target of mTOR kinase, 4E-BP1.⁷⁰ This mTOR target protein is less sensitive to rapalogs inhibition, as discussed above, but could be inhibited by active site inhibitors. Several genes encoding for FGF ligands are located on another breast cancer amplicon at 11q13 where the cyclin D1 gene, CCDN1 as well as another target protein of mTOR, p70S6K2 (RPS6KB2) both reside. As a result, these cancers harbor multiple targets of the PI3 K/ AKT/ mTOR pathway that may contribute to their sensitivity to pathway inhibition. Co-amplification of the two loci containing mTOR targets at 8p11.23 and 11q13 occur in breast cancers and have been proposed to synergize in carcinogenesis.^71,72 Sensitivity to everolimus of breast cancers with co-amplifications and comparison with the sensitivity of those cancers with individual amplification in either locus could be of interest from both a biomarker and a pathophysiologic point of view, especially given the endocrine resistance conferred by the amplifications.⁷³

Besides mTOR inhibitors, several other drugs inhibiting the PI3 K/ AKT/ mTOR pathway have entered the clinic or are in development, including PI3 K inhibitors, AKT inhibitors and dual PI3 K/ mTOR inhibitors.^7,74 However, despite inhibiting proteins in the same pathway, each class of drugs have unique effects stemming from acting at different levels of the pathway. As a result, they produce variable feedback and feed-forward loop regulations, that also depend on concomitant molecular alterations in related pathways. Thus, it is counterproductive to examine effects of different inhibitors of the PI3 K/ AKT/ mTOR pathway as a whole and studies need to focus on specific drugs rather than handling different inhibitors together based only on the fact that they affect targets in the same pathway. In addition, cases harboring aberrations in the specific target gene/ protein of a specific drug, for example mTOR for everolimus, should be prioritized as the target population. This strategy should result in a more homogeneous cohort of potentially responsive patients and could lead to a more precise definition of a sensitive patient population to be included in trials of each agent. In the other hand, this inclusion criterion would decrease the pool of potentially eligible patients and will require multicenter collaborations to complete trials in a timely fashion.

The importance of the individual cancer molecular environment in determining sensitivity and eventual efficacy of everolimus is further stressed by the fact that several processes intertwined with mTOR regulation and functions are cancer stage specific. For example, kinase AMPK has dual effects according to the stage of carcinogenesis.⁷⁵ In normal cells and early stages of neoplastic transformation, activation of AMPK by metformin protects from neoplastic transformation or cancer progression. However, in established cancers, AMPK activation may promote cell survival by signaling down-stream of DNA damage and causing a cell cycle arrest leading to repair of damaged DNA.⁷⁵ It is plausible that mTOR as a major effector of AMPK has a similar stage dependent effect in neoplasia. Indeed, besides functioning in protein translation and inhibition of autophagy, mTOR promotes pyroptosis (inflammation associated apoptosis) through mitochondrial Reactive Oxygen Species (ROS) production and gasdermin D membrane polymerization.⁷⁶ Pyroptosis promotion is mediated through interaction of mTORC1 with the Rag/ Ragulator complex at lysosomal membranes and requires the previous localization of the aminoterminal part of gasdermin D cleaved by inflammatory caspases to the cell membrane, where ROS promote gasdermin D polymerization and membrane pore formation.⁷⁷ Oxidative stress is a process with dual effect in cancer pathogenesis, whence in normal cells and early cancers, mechanisms protecting against reactive species detoxification such as the master transcription factor regulators of oxidative stress NRF1 and NRF2, become cancer promoters in later phase of cancers by protecting cancer cells.⁷⁸ Autophagy is another process regulated by mTOR with dual effects in carcinogenesis, that depend on cellular context.⁷⁹ Activation of autophagy under conditions of macronutrients depletion protects cells by providing fuels for survival and building blocks for damaged cellular organelles rejuvenation. However, cancer cells under stress may also use autophagy for energy and macromolecules supply.⁸⁰ Thus, mTOR inhibition may have inadvertent pro-carcinogenic effects in established cancers through de-repression of autophagy and suppression of ROS production. The key role of mTOR in integrating multiple signals suggests that its inhibition will be beneficial only in carefully selected cancers with specific molecular lesions that remain to be fully characterized.

Feedback loops leading to activation of the PI3 K/ AKT/ mTOR pathway may result not only in restoration of mTOR activity despite continuous everolimus treatment but also to activation of other pro-carcinogenic functions of AKT kinase stemming from inhibition of other targets, such as kinase GSK3 and the FOXO family transcription factors.^81,82 Activation of IRS-1 (Insulin Receptor Substrate 1) follows inhibition of p70S6 K that is a negative regulator of IRS-1 and alleviates the physiologic feedback inhibition of the pathway in cancers with mTOR activation treated with everolimus.⁸³ In addition, treatment with an inhibitor of IGF-IR (Insulin like Growth Factor I Receptor) or an antibody against the receptor abrogated feedback AKT activation following everolimus exposure. GSK3 is constitutively active in cells at baseline conditions without growth factor stimulation and inhibits cell survival by neutralization of anti-apoptotic BCL2 family member MCL1. It also inhibits oncogene c-Myc, NRF2, the master regulator of oxidative stress and β-catenin.⁸⁴ GSK3 is inhibited by AKT after growth factor stimulation.⁸² MCF-7 breast cancer cells with a kinase-dead GSK3 are sensitive to everolimus.⁸⁵ The presence of the IRS-1 feedback loop suggests that treatment combinations of everolimus with inhibitors of IGFR-I or substituting mTOR kinase inhibitors for rapalogs could be effective strategies to avoid AKT activity induction.⁷⁴ In addition, IRS-1 levels could be used as biomarkers of sensitivity to rapalogs. Neuroendocrine tumor cells that are resistant to everolimus after exposure to the drug in vitro display upregulation of IRS-1, compared with parental cells that are everolimus sensitive.⁸⁶

Immunotherapy with immune checkpoint inhibitors is one of the newest weapons in the drug armamentarium of clinical oncology and inhibitors of the PD-L1/ PD-1 ligand/ receptor pair are the most successfully used drugs.⁶ Although immunotherapy for breast cancer has currently limited use, mostly in triple negative cancers, due to low immunogenicity of the disease, the observation that mTOR inhibition up-regulates ligand PD-L1 suggests a possible role in combination with rapalogs.⁸⁷ PD-L1 is up-regulated following everolimus exposure of various cancer cells, and this is associated with p70S6 K activity suppression and suppression of PD-L1 proteasome degradation through ligase β-TrCP.⁸⁸ Given that phosphorylation by GSK3 is a usual prerequisite for ubiquitination, up-regulation of PD-L1 may be caused by the IRS-1 feedback loop up-regulation of AKT that suppresses GSK3. PD-L1 expression may, thus, be a marker of everolimus resistance and an indication that combination treatment with the addition of a checkpoint inhibitor may restore everolimus sensitivity.

Everolimus is a substrate of P-glycoprotein (ABCB1, ATP Binding Cassette B1), a transporter protein associated with cell protection through cell expulsion of toxins and drugs.⁸⁹ Thus, cancer cells that express the protein may become resistant to everolimus by prevention of intracellular accumulation.⁹⁰ Resistance of intracranial disease to everolimus is also mediated by P-glycoprotein.⁹¹ P-glycoprotein expression in tumors and other proteins that affect intratumoral availability of everolimus, such as metabolizing enzyme CYP3A, could be envisioned as clinical biomarkers of response.

Data discussed in this article confirm that everolimus has the potential to benefit a sub-set of carefully selected patients with breast cancer. NGS studies allowing for parallel examination of a greater number of genes and their alterations have made clear that multiple molecular aberrations are present in most cases of cancer. As a result, it is unlikely that inhibiting a single alteration or pathway will have a significant effect in outcomes and the next step is to define combinations of therapies that display significant synergy in specific cancers. In fact, the partial success of everolimus in ER positive breast cancers is based in combinations with hormonal agents which continue to be relevant as partners in the therapy of this subset of breast cancers even after development of hormonal resistance. Moreover, ER is a down-stream target of phosphorylation by kinase p70S6 K and thus inhibition of mTOR blocks one of the mechanisms of resistance to hormone therapies.⁹² Other partners for combination therapies including combinations with inhibitors of receptor tyrosine kinases await discovery and development.⁹³ However, as it is often the case with targeted combination therapies toxicity may become a barrier that interferes with feasibility of full monotherapy doses of the used agents. Thus, even mechanistically supported, and effective combinations will have to pass the safety test before entering the clinic. Related to this discussion, initial enthusiasm for the clinical use of combinations of hormonal therapies with everolimus was somewhat blunted by the oral toxicity of everolimus which produces painful mouth ulcerations, inconsistent with long-term use. Improvement in the management of this toxicity with the use of steroid mouth wash abrogates oral toxicity and allows continuation of treatment.⁹⁴ Improved toxicity management has the potential to benefit everolimus use in combination with other drugs if sensitive subsets of breast cancers are better defined.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

ORCID iD

Ioannis A. Voutsadakis

References

Siegel

Miller

Fuchs

, et al. Cancer statistics, 2021. CA Cancer J Clin 2021; 71: 7–33. PMID: 33433946.

Schettini

Prat

. Dissecting the biological heterogeneity of HER2-positive breast cancer. Breast 2021; 59: 339–350. PMID: 34392185; PMCID: PMC8374722.

Finn

Martin

Rugo

, et al. Palbociclib and letrozole in advanced breast cancer. N Engl J Med 2016; 375: 1925–1936. PMID: 27959613.

Robson

Senkus

, et al. Olaparib for metastatic breast cancer in patients with a germline BRCA mutation. N Engl J Med 2017; 377: 523–533. PMID: 28578601.

Tutt

ANJ

Garber

Kaufman

, et al. Geyer CE Jr; Olympia clinical trial steering committee and investigators. Adjuvant Olaparib for Patients With BRCA1– or BRCA2-mutated breast cancer. N Engl J Med 2021; 384: 2394–2405. PMID: 34081848.

Voutsadakis

. Immune blockade inhibition in breast cancer. Anticancer Res 2016; 36: 5607–5622. PMID: 27793883.

Stravodimou

Voutsadakis

. The future of ER + /HER2- metastatic breast cancer therapy: beyond PI3 K inhibitors. Anticancer Res 2020; 40: 4829–4841. PMID: 32878771.

Baselga

Campone

Piccart

, et al. Everolimus in postmenopausal hormone-receptor-positive advanced breast cancer. N Engl J Med 2012; 366: 520–529. PMID: 22149876; PMCID: PMC5705195.

André

Ciruelos

Rubovszky

, et al. N Engl J Med 2019; 380: 1929–1940. doi: 10.1056/NEJMoa1813904. PMID: 31091374.

10.

Gong

, et al. A multicenter analysis of treatment patterns and clinical outcomes of subsequent therapies after progression on palbociclib in HR + /HER2- metastatic breast cancer. Ther Adv Med Oncol 2021; 13: 17588359211022890. PMID: 34178122; PMCID: PMC8202336.

11.

Oza

Thomas

, et al. Retrospective analysis of treatment patterns and effectiveness of palbociclib and subsequent regimens in metastatic breast cancer. J Natl Compr Canc Netw 2019; 17: 141–147. PMID: 30787127; PMCID: PMC6752198.

12.

Rozenblit

Mun

Soulos

, et al. Patterns of treatment with everolimus exemestane in hormone receptor-positive HER2-negative metastatic breast cancer in the era of targeted therapy. Breast Cancer Res 2021; 23: 14. PMID: 33514405; PMCID: PMC7844919.

13.

Fruman

Chiu

Hopkins

, et al. The PI3 K pathway in human disease. Cell 2017; 170: 605–635. PMID: 28802037; PMCID: PMC5726441.

14.

Saxton

Sabatini

. mTOR signaling in growth, metabolism, and disease. Cell 2017; 168: 960–976. PMID: 28283069; PMCID: PMC5394987.

15.

Yang

Rudge

Koos

, et al. mTOR kinase structure, mechanism and regulation. Nature 2013; 497: 217–223. PMID: 23636326; PMCID: PMC4512754.

16.

González

Hall

Lin

, et al. AMPK And TOR: the Yin and yang of cellular nutrient sensing and growth control. Cell Metab 2020; 31: 472–492. PMID: 32130880.

17.

Karlsson

Magić

Bostner

, et al. Revealing different roles of the mTOR-targets S6K1 and S6K2 in breast cancer by expression profiling and structural analysis. PLoS One 2015; 10: e0145013. PMID: 26698305; PMCID: PMC4689523.

18.

Dorrello

Peschiaroli

Guardavaccaro

, et al. S6K1- and betaTRCP-mediated degradation of PDCD4 promotes protein translation and cell growth. Science 2006; 314: 467–471. PMID: 17053147.

19.

Ben-Sahra

Howell

Asara

, et al. Stimulation of de novo pyrimidine synthesis by growth signaling through mTOR and S6K1. Science 2013; 339: 1323–1328. PMID: 23429703; PMCID: PMC3753690.

20.

Gallagher

Williamson

Chan

. Advances in autophagy regulatory mechanisms. Cells 2016; 5: 24. PMID: 27187479; PMCID: PMC4931673.

21.

Kolos

Voll

Bauder

, et al.

FKBP Ligands-Where We Are and where to Go?

Front Pharmacol 2018; 9: 1425. PMID: 30568592; PMCID: PMC6290070.

22.

Annett

Moore

Robson

. FK506 Binding proteins and inflammation related signalling pathways; basic biology, current status and future prospects for pharmacological intervention. Pharmacol Ther 2020; 215: 107623. PMID: 32622856.

23.

März

Fabian

Kozany

, et al. Large FK506-binding proteins shape the pharmacology of rapamycin. Mol Cell Biol 2013; 33: 1357–1367. PMID: 23358420; PMCID: PMC3624267.

24.

Kang

Pacold

Cervantes

, et al. mTORC1 phosphorylation sites encode their sensitivity to starvation and rapamycin. Science 2013; 341: 1236566. PMID: 23888043; PMCID: PMC3771538.

25.

Choo

Yoon

Kim

, et al. Rapamycin differentially inhibits S6Ks and 4E-BP1 to mediate cell-type-specific repression of mRNA translation. Proc Natl Acad Sci U S A 2008; 105: 17414–17419. 18955708; PMCID: PMC2582304.

26.

She

Halilovic

, et al. 4E-BP1 Is a key effector of the oncogenic activation of the AKT and ERK signaling pathways that integrates their function in tumors. Cancer Cell 2010; 18: 39–51. PMID: 20609351; PMCID: PMC3286650.

27.

Alain

Morita

Fonseca

, et al. eIF4E/4E-BP ratio predicts the efficacy of mTOR targeted therapies. Cancer Res 2012; 72: 6468–6476. PMID: 23100465.

28.

Cancer Genome Atlas Network. Comprehensive molecular portraits of human breast tumours. Nature 2012; 490: 61–70. PMID: 23000897; PMCID: PMC3465532.

29.

Gao

Aksoy

Dogrusoz

, et al. Integrative analysis of complex cancer genomics and clinical profiles using the cBioPortal. Sci Signal 2013; 6: l1. PMID: 23550210; PMCID: PMC4160307.

30.

. Biomarkers of everolimus sensitivity in hormone receptor-positive breast cancer. J Breast Cancer 2017; 20: 321–326. PMID: 29285035; PMCID: PMC5743990.

31.

Bachelot

Bourgier

Cropet

, et al. Randomized phase II trial of everolimus in combination with tamoxifen in patients with hormone receptor-positive, human epidermal growth factor receptor 2-negative metastatic breast cancer with prior exposure to aromatase inhibitors: a GINECO study. J Clin Oncol 2012; 30: 2718–2724. PMID: 22565002.

32.

Treilleux

Arnedos

Cropet

, et al. Translational studies within the TAMRAD randomized GINECO trial: evidence for mTORC1 activation marker as a predictive factor for everolimus efficacy in advanced breast cancer. Ann Oncol 2015; 26: 120–125. PMID: 25361980.

33.

Hortobagyi

Chen

Piccart

, et al. Correlative analysis of genetic alterations and everolimus benefit in hormone receptor-positive, human epidermal growth factor receptor 2-negative advanced breast cancer: results From BOLERO-2. J Clin Oncol 2016; 34: 419–426. PMID: 26503204; PMCID: PMC5070556.

34.

Kruger

Opdam

van der Noort

, et al. PI3 K Pathway protein analyses in metastatic breast cancer patients receiving standard everolimus and exemestane. J Cancer Res Clin Oncol 2020; 146: 3013–3023. PMID: 32566979; PMCID: PMC7519923.

35.

André

Hurvitz

Fasolo

, et al. Molecular alterations and everolimus efficacy in human epidermal growth factor receptor 2-overexpressing metastatic breast cancers: combined exploratory biomarker analysis From BOLERO-1 and BOLERO-3. J Clin Oncol 2016; 34: 2115–2124. PMID: 27091708.

36.

Jovanović

Mayer

, et al. A randomized phase II neoadjuvant study of cisplatin, paclitaxel With or without everolimus in patients with stage II/III triple-negative breast cancer (TNBC): responses and long-term outcome correlated with increased frequency of DNA damage response gene mutations, TNBC subtype, AR Status, and Ki67. Clin Cancer Res 2017; 23: 4035–4045. PMID: 28270498; PMCID: PMC5540799.

37.

Piccart

Hortobagyi

Campone

, et al. Everolimus plus exemestane for hormone-receptor-positive, human epidermal growth factor receptor-2–negative advanced breast cancer: overall survival results from BOLERO-2. Ann Oncol 2014; 25: 2357–2362. PMID: 25231953; PMCID: PMC6267855.

38.

Moynahan

Chen

, et al. Correlation between PIK3CA mutations in cell-free DNA and everolimus efficacy in HR⁺, HER2^– advanced breast cancer: results from BOLERO-2. Br J Cancer 2017; 116: 726–730. PMID: 28183140; PMCID: PMC5355930.

39.

Iyer

Hanrahan

Milowsky

, et al. Genome sequencing identifies a basis for everolimus sensitivity. Science 2012; 338: 21. PMID: 22923433; PMCID: PMC3633467.

40.

Voss

Hakimi

Pham

, et al. Tumor genetic analyses of patients with metastatic renal cell carcinoma and extended benefit from mTOR inhibitor therapy. Clin Cancer Res 2014; 20: 1955–1964. PMID: 24622468; PMCID: PMC4140619.

41.

Lim

Park

Kim

, et al. Next-generation sequencing reveals somatic mutations that confer exceptional response to everolimus. Oncotarget 2016; 7: 10547–10556. PMID: 26859683; PMCID: PMC4891139.

42.

Wagle

Grabiner

Van Allen

, et al. Activating mTOR mutations in a patient with an extraordinary response on a phase I trial of everolimus and pazopanib. Cancer Discov 2014; 4: 546–553. PMID: 24625776; PMCID: PMC4122326.

43.

Krasniqi

Barchiesi

Mazzotta

, et al. Case report: 5-year progression free survival and complete liver response in a patient with metastatic breast cancer treated with everolimus plus exemestane. Medicine (Baltimore) 2020; 99: e21211. PMID: 32756099; PMCID: PMC7402792.

44.

Bajpai

Ramaswamy

Chandrasekharan

, et al. Activation of phosphoinositide 3-kinase/Akt/mechanistic target of rapamycin pathway and response to everolimus in endocrine receptor-positive metastatic breast cancer - A retrospective pilot analysis and viewpoint. South Asian J Cancer 2017; 6: 102–105. PMID: 28975114; PMCID: PMC5615875.

45.

Wheler

Atkins

Janku

, et al. Multiple gene aberrations and breast cancer: lessons from super-responders. BMC Cancer 2015; 15: 42. PMID: 26021831; PMCID: PMC4446801.

46.

Voutsadakis

. 8p11.23 Amplification in breast cancer: molecular characteristics, prognosis and targeted therapy. J Clin Med 2020; 9: 3079. PMID: 32987805; PMCID: PMC7598661.

47.

Wheler

Moulder

Naing

, et al. Anastrozole and everolimus in advanced gynecologic and breast malignancies: activity and molecular alterations in the PI3 K/AKT/mTOR pathway. Oncotarget 2014; 5: 3029–3038. PMID: 24912489; PMCID: PMC4102789.

48.

Parachoniak

Rankin

Gaffney

, et al. Exceptional durable response to everolimus in a patient with biphenotypic breast cancer harboring an STK11 variant. Cold Spring Harb Mol Case Stud 2017; 3: a000778. PMID: 28550065; PMCID: PMC5593157.

49.

Klümpen

Queiroz

Spek

, et al. mTOR inhibitor treatment of pancreatic cancer in a patient With peutz-jeghers syndrome. J Clin Oncol 2011; 29: e150–e153. PMID: 21189378; PMCID: PMC4856022.

50.

de Brabander

Eskens

FALM

Korsse

, et al. Chemoprevention in patients with peutz-jeghers syndrome: lessons learned. Oncologist 2018; 23: 399–e33. PMID: 29371475; PMCID: PMC5896716.

51.

Estevez-Diz

Bonadio

Carvalho

, et al. Everolimus plus anastrozole for female adnexal tumor of probable wolffian origin (FATWO) with STK11 mutation. Gynecol Oncol Rep 2021; 37: 100838. PMID: 34386569; PMCID: PMC8346534.

52.

Wheler

Atkins

Janku

, et al. Presence of both alterations in FGFR/FGF and PI3 K/AKT/mTOR confer improved outcomes for patients with metastatic breast cancer treated with PI3 K/AKT/mTOR inhibitors. Oncoscience 2016; 3: 164–172. PMID: 27489863; PMCID: PMC4965259.

53.

Mimoto

Nihira

Hirooka

, et al. Diminished DYRK2 sensitizes hormone receptor-positive breast cancer to everolimus by the escape from degrading mTOR. Cancer Lett 2017; 384: 27–38. PMID: 27746162.

54.

Hurvitz

Andre

Jiang

, et al. Combination of everolimus with trastuzumab plus paclitaxel as first-line treatment for patients with HER2-positive advanced breast cancer (BOLERO-1): a phase 3, randomised, double-blind, multicentre trial. Lancet Oncol 2015; 16: 816–829. PMID: 26092818.

55.

André

O’Regan

Ozguroglu

, et al. Everolimus for women with trastuzumab-resistant, HER2–positive, advanced breast cancer (BOLERO-3): a randomised, double-blind, placebo-controlled phase 3 trial. Lancet Oncol 2014; 15: 580–591. PMID: 24742739.

56.

Okazaki

Horimoto

Tanabe

, et al. Predictive markers for efficacy of everolimus plus exemestane in patients with luminal HER2-negative metastatic breast cancer. Med Oncol 2018 Mar 8; 35: 48. PMID: 29520681.

57.

Van den Bossche

Jadot

Grisay

, et al. c-MET as a potential resistance mechanism to everolimus in breast cancer: from a case report to patient cohort analysis. Target Oncol 2020 Feb; 15: 139–146. PMID: 32020516.

58.

Chandarlapaty

Chen

, et al. Prevalence of ESR1 mutations in cell-free DNA and outcomes in metastatic breast cancer: a secondary analysis of the BOLERO-2 clinical trial. JAMA Oncol 2016; 2: 1310–1315. PMID: 27532364; PMCID: PMC5063698.

59.

Prat

Brase

Cheng

, et al. Everolimus plus exemestane for hormone receptor-positive advanced breast cancer: a PAM50 intrinsic subtype analysis of BOLERO-2. Oncologist 2019; 24: 893–900. PMID: 30679318; PMCID: PMC6656445.

60.

Pizzuti

Marchetti

Natoli

, et al. Fasting glucose and body mass index as predictors of activity in breast cancer patients treated with everolimus-exemestane: the EverExt study. Sci Rep 2017; 7: 10597. PMID: 28878375; PMCID: PMC5587713.

61.

Corona

Giudici

Jerusalem

, et al. Impact of BMI on the outcome of metastatic breast cancer patients treated with everolimus: a retrospective exploratory analysis of the BALLET study. Oncotarget 2020; 11: 2172–2181. PMID: 32577163; PMCID: PMC7289535.

62.

Schettini

Sobhani

Ianza

, et al. Immune system and angiogenesis-related potential surrogate biomarkers of response to everolimus-based treatment in hormone receptor-positive breast cancer: an exploratory study. Breast Cancer Res Treat 2020; 184: 421–431. PMID: 32770287; PMCID: PMC7599144.

63.

Jerusalem

Mariani

Ciruelos

, et al. Safety of everolimus plus exemestane in patients with hormone-receptor-positive, HER2–negative locally advanced or metastatic breast cancer progressing on prior non-steroidal aromatase inhibitors: primary results of a phase IIIb, open-label, single-arm, expanded-access multicenter trial (BALLET). Ann Oncol 2016; 27: 1719–1725. PMID: 27358383.

64.

Sirico

Bernocchi

Sobhani

, et al. Early changes of the standardized uptake values (SUV_max) predict the efficacy of everolimus-exemestane in patients with hormone receptor-positive metastatic breast cancer. Cancers (Basel) 2020; 12: 3314. PMID: 33182575; PMCID: PMC7697290.

65.

Hurvitz

Kalous

Conklin

, et al. In vitro activity of the mTOR inhibitor everolimus, in a large panel of breast cancer cell lines and analysis for predictors of response. Breast Cancer Res Treat 2015; 149: 669–680. PMID: 25663547.

66.

Grabiner

Nardi

Birsoy

, et al. A diverse array of cancer-associated MTOR mutations are hyperactivating and can predict rapamycin sensitivity. Cancer Discov 2014; 4: 554–563. PMID: 24631838; PMCID: PMC4012430.

67.

Satheesha

Cookson

Coleman

, et al. Response to mTOR inhibition: activity of eIF4E predicts sensitivity in cell lines and acquired changes in eIF4E regulation in breast cancer. Mol Cancer 2011; 10: 19. PMID: 21320304; PMCID: PMC3055230.

68.

Di Nicolantonio

Arena

Tabernero

, et al. Deregulation of the PI3 K and KRAS signaling pathways in human cancer cells determines their response to everolimus. J Clin Invest 2010; 120: 2858–2866. PMID: 20664172; PMCID: PMC2912177.

69.

Yunokawa

Koizumi

Kitamura

, et al. Efficacy of everolimus, a novel mTOR inhibitor, against basal-like triple-negative breast cancer cells. Cancer Sci 2012; 103: 1665–1671. PMID: 22703543; PMCID: PMC7659327.

70.

Voutsadakis

. Amplification of 8p11.23 in cancers and the role of amplicon genes. Life Sci 2021; 264: 118729. PMID: 33166592.

71.

Karlsson

Waltersson

Bostner

, et al. High-resolution genomic analysis of the 11q13 amplicon in breast cancers identifies synergy with 8p12 amplification, involving the mTOR targets S6K2 and 4EBP1. Genes Chromosomes Cancer 2011; 50: 775–787. PMID: 21748818.

72.

Kwek

Roy

Zhou

, et al. Co-amplified genes at 8p12 and 11q13 in breast tumors cooperate with two major pathways in oncogenesis. Oncogene 2009; 28: 1892–1903. PMID: 19330026; PMCID: PMC2722962.

73.

Turner

Pearson

Sharpe

, et al. FGFR1 Amplification drives endocrine therapy resistance and is a therapeutic target in breast cancer. Cancer Res 2010; 70: 2085–2094. PMID: 20179196; PMCID: PMC2832818.

74.

Miskimins

Dey

, et al. Promise of rapalogues versus mTOR kinase inhibitors in subset specific breast cancer: old targets new hope. Cancer Treat Rev 2013; 39: 403–412. PMID: 23352077.

75.

Russell

Hardie

. AMP-Activated Protein kinase: do we need activators or inhibitors to treat or prevent cancer? Int J Mol Sci 2020; 22: 86. PMID: 33375416; PMCID: PMC7795930.

76.

Evavold

Hafner-Bratkovič

Devant

, et al. Control of gasdermin D oligomerization and pyroptosis by the ragulator-Rag-mTORC1 pathway. Cell 2021; 184: 4495–4511.e19. PMID: 34289345; PMCID: PMC8380731.

77.

Sborgi

Rühl

Mulvihill

, et al. GSDMD Membrane pore formation constitutes the mechanism of pyroptotic cell death. EMBO J 2016; 35: 1766–1778. PMID: 27418190; PMCID: PMC5010048.

78.

Voutsadakis

. Proteasome expression and activity in cancer and cancer stem cells. Tumour Biol 2017; 39: 1010428317692248. PMID: 28345458.

79.

Alvarez-Meythaler

Garcia-Mayea

Mir

, et al. Autophagy takes center stage as a possible cancer hallmark. Front Oncol 2020; 10: 586069. PMID: 33194736; PMCID: PMC7643020.

80.

Voutsadakis

. Mutations of p53 associated with pancreatic cancer and therapeutic implications. Ann Hepatobiliary Pancreat Surg 2021; 25: 315–327. PMID: 34402431; PMCID: PMC8382872.

81.

Sun

. mTOR-targeted cancer therapy: great target but disappointing clinical outcomes, why? Front Med 2021; 15: 221–231. PMID: 33165737.

82.

Manning

Toker

. AKT/PKB signaling: navigating the network. Cell 2017; 169: 381–405. PMID: 28431241; PMCID: PMC5546324.

83.

O’Reilly

Rojo

She

, et al. mTOR inhibition induces upstream receptor tyrosine kinase signaling and activates Akt. Cancer Res 2006; 66: 1500–1508. PMID: 16452206; PMCID: PMC3193604.

84.

Kaidanovich-Beilin

Woodgett

. GSK-3: functional insights from cell biology and animal models. Front Mol Neurosci 2011; 4: 40. PMID: 22110425; PMCID: PMC3217193.

85.

Sokolosky

Chappell

Stadelman

, et al. Inhibition of GSK-3β activity can result in drug and hormonal resistance and alter sensitivity to targeted therapy in MCF-7 breast cancer cells. Cell Cycle 2014; 13: 820–833. PMID: 24407515; PMCID: PMC3979918.

86.

Aristizabal Prada

Spöttl

Maurer

, et al. The role of GSK3 and its reversal with GSK3 antagonism in everolimus resistance. Endocr Relat Cancer 2018; 25: 893–908. PMID: 29895527; PMCID: PMC7439002.

87.

Hirayama

Yamano

, et al. Anti-PD-L1 treatment enhances antitumor effect of everolimus in a mouse model of renal cell carcinoma. Cancer Sci 2016; 107: 1736–1744. PMID: 27712020; PMCID: PMC5198964.

88.

Deng

Qian

Zhang

, et al. Inhibition of mTOR complex 1/p70 S6 kinase signaling elevates PD-L1 levels in human cancer cells through enhancing protein stabilization accompanied with enhanced β-TrCP degradation. Oncogene 2019; 38: 6270–6282. PMID: 31316145.

89.

Hua

Kong

Zhang

, et al. Targeting mTOR for cancer therapy. J Hematol Oncol 2019; 12: 71. PMID: 31277692; PMCID: PMC6612215.

90.

Ravaud

Urva

Grosch

, et al. Relationship between everolimus exposure and safety and efficacy: meta-analysis of clinical trials in oncology. Eur J Cancer 2014; 50: 486–495. PMID: 24332451.

91.

Tang

Sparidans

Cheung

, et al. P-glycoprotein, CYP3A, and plasma carboxylesterase determine brain and blood disposition of the mTOR inhibitor everolimus (afinitor) in mice. Clin Cancer Res 2014; 20: 3133–3145. PMID: 24727322.

92.

Bostner

Karlsson

Eding

, et al. S6 kinase signaling: tamoxifen response and prognostic indication in two breast cancer cohorts. Endocr Relat Cancer 2015; 22: 331–343. PMID: 25972244.

93.

Fox

Kuba

Miller

, et al. Autocrine IGF-I/insulin receptor axis compensates for inhibition of AKT in ER-positive breast cancer cells with resistance to estrogen deprivation. Breast Cancer Res 2013; 15: 55. PMID: 23844554; PMCID: PMC3979036.

94.

Rugo

Seneviratne

Beck

, et al. Prevention of everolimus-related stomatitis in women with hormone receptor-positive, HER2–negative metastatic breast cancer using dexamethasone mouthwash (SWISH): a single-arm, phase 2 trial. Lancet Oncol 2017; 18: 654–662. PMID: 28314691.