Original Article

An In Silico Analysis Identified Members of the Pleckstrin Homology-Like Domain, Family B (PHLDB family) as Potential Prognostic and Predictive Biomarkers of Treatment Response in Breast Cancer Patients


  • Renan Gomes do Nascimento
  • Jéssica de Moraes
  • Danilo de Oliveira Cerqueira
  • Sandro Jorge Januário

Received Date: 23.03.2022 Accepted Date: 19.04.2022 Eur J Breast Health 2022;18(3):235-247 PMID: 35855195


Breast cancer is the leading cause of morbidity and mortality in women worldwide. This malignant neoplasm can be classified into four clinically relevant subtypes according to the expression of a number of biomarkers. However, these tumors show considerable intratumoral heterogeneity and multidrug resistance. Members of the pleckstrin homology-like domain, family B (PHLDB) play a critical role in the regulation of p53 and AKT signaling pathways, important for cancer and cellular metabolism. The present study was performed to evaluate the expression pattern of PHLDB family members in breast cancer and its potential prognostic and predictive value for therapeutic response using bioinformatics tools.

Materials and Methods:

This in silico analysis was performed using several online repositories, including UALCAN, GEPIA2, bc-GenExMiner, KM Plotter, PrognoScan and ROC Plotter.


PHLDB family genes were found to be differentially expressed in tumor samples when compared to healthy breast tissue samples. Furthermore, epigenetic regulation may be one of the regulatory mechanisms for the expression of these markers. The PHLDB family of genes proved to be potential markers for predicting the development of lymph node metastasis (p<0.0001) and poor clinical outcome. All members of the PHLDB family were significantly correlated with hormone receptors. High levels of PHLDBs expression were associated with worse overall survival and recurrence-free survival in breast cancer patients. Finally, our data demonstrate that members of the PHLDB family can be promising markers in the stratification of patients who may or may not respond to different available therapies.


Our cumulative results demonstrate that PHLDB family members may be promising biomarkers for predicting prognosis and therapeutic response in breast cancer patients.

Keywords: Breast cancer, PHLDB, in silico analysis, biomarkers

Key Points

• The pleckstrin homology-like domain, family b (PHLDB) family of genes are differentially expressed in tumor and normal breast tissues.

• Members of the PHLDB family are potential markers for predicting the development of lymph node metastasis and poor clinical outcome.

• Reduced expression of PHLDB 1, 2, and 3 mRNA was associated with decreased overall and recurrence-free survival rates in breast cancer patients.

• There is a possible relationship between PHLDB family member expression and response to endocrine therapy and to anti-HER2 antibodies.


Breast cancer is the malignant neoplasm with the highest rates of occurrence and mortality among women worldwide (1). Currently, it is known that breast cancer represents a phenotypically and biologically heterogeneous collection of diseases, culminating in different clinical patterns, prognosis and response to usual treatments (2).

Based on the expression of molecular biomarkers, breast cancer can be classified into four main subtypes widely accepted and used in clinical practice: Luminal A, Luminal B, human epidermal growth factor receptor 2 (HER2+) and triple negative breast cancer (TNBC) (3). The segregation of these molecular subtypes is due to genes responsible for the expression of hormone receptors for estrogen (ER) and progesterone (PR), HER2 and the cell proliferation marker, Ki-67 (4).

Although the sum of current clinical, pathological and molecular indicators favors a contribution in establishing the prognosis and predicting the therapeutic response of patients, the investigation of new, more robust, sensitive, specific and well-validated biomarkers is occurring, partially in response to the trend towards personalized medicine (5).

In this context, the Pleckstrin Homology-like Domain (PHLD) multifunctional protein class has been attracting interest for its role in the regulation of p53 and AKT signaling pathways, both of which are important for cancer and cellular metabolism (6). The PHLD protein class is organized into two separate families, PHLDA and PHLDB, each of which is composed of three members (6). All members of the PHLD families code for proteins that have a functional domain called PH (pleckstrin homology) (6). PH-like domains consist of 100 to 120 amino acid residues and are found in a wide range of proteins involved in intracellular signaling, and may also participate in cytoskeletal rearrangement and membrane trafficking (7). Furthermore, proteins with the PH domain have been well categorized as phosphatidylinositol-binding molecular modules located internally in the cell membrane, as well as other proteins with varying specificity (8, 9). The two PHLD protein families, A and B, differ from each other by the position of their PH domain in the N- or C-terminal region or in the length of the protein (6). Although identified nearly three decades ago, the PHLD class of proteins remains understudied in the oncological context, with members of the PHLDB family receiving the least attention in recent research.

Therefore, the present study was carried out to evaluate the expression pattern of PHLDB family members in breast cancer and its potential prognostic and predictive value for therapeutic response, through public datasets deposited in online repositories.

Materials and Methods

UALCAN and GEPIA2: UALCAN (http://ualcan.path.uab.edu/) is a free online platform to access and assess the expression profile of biomarkers in different types of cancers (10). UALCAN was used to investigate gene expression levels of PHLDB family members in normal and tumor samples from the breast, as well as in tumor subgroups and at different clinical stages. The level of methylation of the promoter region of the PHLDB family in breast cancer samples and normal tissues was also investigated using this same platform. Additionally, GEPIA2 (http://gepia2.cancer-pku.cn/) was accessed. GEPIA2 is a new improved web server to analyze RNA sequencing expression data from 9,736 tumors and 8,587 normal samples from the TCGA project (The Cancer Genome Atlas) and Genotype-Tissue Expression (GTEx) (11).

bc-GenExMiner: The Breast Cancer Gene-Expression Miner v4.5 (http://bcgenex.centregauducheau.fr/) is an online mining tool for properly annotated breast cancer transcriptomic data (12). For this study, we considered only the microarray data to analyze the expression of the PHLDB family with clinic pathological parameters, regarding the classic breast cancer biomarkers and the different molecular subtypes. The median expression was used as the cut-off point.

KM Plotter: The Kaplan–Meier Plotter (https://kmplot.com/analysis/) is a practical, easy-to-use survival analysis platform that hosts data from 21 different types of cancers (13). We investigated the expression of PHLDB family members according to overall survival (OS) and recurrence-free survival (RFS). The dataset included cDNA microarrays from the TCGA available in the KM Plotter online database. The validated probes were chosen according to the best automatic cut selection criteria. Follow-up time was adjusted to 120 months. Log-rank p-values and hazard ratio (HR) with 95% confidence interval (CI) were automatically determined.

PrognoScan: The PrognoScan online database (http://www.prognoscan.org/) provides a powerful platform to assess biological relationships between gene expression and cancer patient prognosis information, including overall survival (OS), relapse-free survival (RFS), distant metastasis-free survival (DMFS), and disease-specific survival (DSS) (14). PrognoScan includes public cDNA microarray datasets with clinical annotations of gene expression and prognosis from Gene Expression Omnibus (GEO) and ArrayExpress, for example. Cox p-values and hazard ratio (HR) with 95% confidence intervals (CI) were calculated automatically.

ROC Plotter: The ROC Plotter (http://www.rocplot.org/) is an interactive and user-friendly online tool (15). With transcriptomic data from 3104 breast cancer patients treated and not treated with endocrine therapy, anti-HER2 therapy, or chemotherapy. Here, we quickly evaluated the expression pattern of PHLDB family genes in the face of the treatment received by the patient.


PHLDB Family Expression and Methylation Status in Samples From Breast Cancer Patients

Using TCGA data analyzed by the UALCAN platform, it was found that PHLDB1 and PHLDB2 had reduced expression in breast cancer tumor tissues when compared to adjacent normal tissues (Figures 1a and 2a; p<0.0001, respectively) and in a larger cohort the same pattern was observed (Supplementary Figures 1a and 1b; p = 0.01, respectively). Furthermore, hyper-methylation of the PHLDB1 promoter region was observed in breast cancer tissues in relation to healthy tissues (Figure 1b; p<0.0001), indicating a possible direct relationship of this epigenetic regulatory mechanism with the reduction of expression in samples of breast cancer. Meanwhile, the highest level of methylation of the PHLDB2 promoter region was observed in healthy breast tissues compared to tumor tissues (Figure 2b; p<0.0001). Contrary to what was observed for PHLDB1 and PHLDB2, PHLDB3 gene expression was higher in breast tumor samples when compared to healthy tissue samples (Figure 3a; p<0.0001) and, again, the same pattern was observed in a larger cohort, although this was not statistically significant (Supplementary Figure 1C). The highest level of methylation of the PHLDB3 promoter region was observed in breast cancer tissues compared to normal tissues (Figure 3b; p<0.0001).

Additionally, the expression patterns of PHLDB family members in relation to molecular classification was investigated. The Luminal type exhibited an increased transcriptional distribution in relation to the TNBC and HER2+ subtypes (Figures 1c, 2c and 3c; p<0.0001, respectively). Furthermore, patients with the most advanced clinical stage of breast cancer tended to express lower levels of PHLDB1, although this was not statistically significant when compared to the other stages of the disease (Figure 1d). However, there was no association between the differential expression of PHLDB2 and PHLDB3 with the different clinical stages of patients with breast tumors (Figures 2d and 3d, respectively).

Association of the Expression of PHLDB Family Members With Clinical-Pathological Characteristics

The open-source tool, bc-GenExMiner, was used for this analysis. The sample evidence suggested that there was a statistically significant association between the status of PHLDB1 expression with all variables tested (Table 1). Significant associations were observed between PHLDB2 expression and nodal status (p = 0.0228), Scarff-Bloom-Richardson (SBR) classification (p<0.0001), Nottingham Prognostic Index (NPI) (p = 0.0002), the statuses of ER (p<0.0001), PR (p = 0.0061), HER2 (p = 0.0147), and TP53 (p<0.0001) and molecular classification (p<0.0001) (Table 1). For the last member of the PHLDB family, statistically significant associations were found between differential expression of PHLDB3 and patient age (p<0.0001), SBR classification (p<0.0001), NPI (p<0.0001), TP53 mutational status (p<0.0001), the expression of ER (p<0.0001), PR (p<0.0001), and HER2 (p<0.0001) and molecular subtype (p<0.0001) (Table 1).

Expression of PHLDB Family Members and Prognostic Value in Breast Cancer Patients

Next, the prognostic value of PHLDB family genes using the KM Plotter platform was investigated. Most notably, reduced levels of mRNA expression of PHLDB family members were significantly correlated with poor prognosis for overall survival (PHLDB1 p = 0.0044; PHLDB2 p = 0.0040 and PHLDB3 p = 0.0046) (Figures 4a, 4b and 4c, respectively) and recurrence-free survival (PHLDB1 p<0.0001; PHLDB2 p = 0.0013 and PHLDB3 p<0.0001) (Figures 4d, 4e and 4f, respectively). Additionally, the PrognoScan database showed that down-regulation of PHLDB family expression was significantly associated with reduction in cumulative rates of overall survival (OS), recurrence-free survival (RFS), distant metastasis-free survival (DMFS) and disease-free survival (DFS) (Table 2).

We also investigated the prognostic role of PHLDB family members in different intrinsic molecular subtypes. Kaplan-Meier curves indicated that high PHLDB1 level was significantly associated with lower cumulative rates of RFS in the TNBC subtype (p = 0.0330) and OS in the TNBC (p = 0.0330) and HER2 subtypes (p = 0.0067) (Supplementary Figures 2D, 2A and 5A, respectively). Meanwhile, reduced levels of PHLDB1 showed lower RFS in Luminal A (p < 0.0001) and Luminal B (p < 0.0001) and OS in Luminal A (p = 0.0008) subtypes in breast cancer patients (Supplementary Figures 3D, 4D and 3A, respectively). We found that upregulation of PHLDB2 expression was significantly correlated with worse rates of RFS in the TNBC (p = 0.0014) and HER2 (p = 0.0210) subtypes and OS in the HER2 subtype (p = 0.0240) (Supplementary Figures 2E, 5E and 5B, respectively). In contrast, reduced levels of PHLDB2 mRNA expression were significantly correlated with reduced RFS in Luminal A (p = 0.0002) and Luminal B (p = 0.0170) and OS in Luminal A subtype (p = 0.0025) (Supplementary Figures 3E, 4E and 3B, respectively). Finally, the Kaplan-Meier curves indicated that the highest level of PHLDB3 correlated with preferable RFS in TNBC (p = 0.0260), Luminal A (p < 0.0001) and Luminal B (p = 0.0009) and OS subtypes in the Luminal A subtype (p = 0.0260) (Supplementary Figures 2F, 3F, 4F and 3C, respectively). Meanwhile, high PHLDB3 level was significantly associated with lower cumulative OS rates in the HER2 subtype (p = 0.0150) (Supplementary Figure 5C).

Predictive Value of PHLDB Family Members for Treatment Response

Considering the reports of some previous studies indicating PHLD family members as potential biomarkers for response to different treatments (16, 17), we conducted an analysis with the ROC Plotter web tool. Our results showed that among patients with hormone-dependent tumors, those who did not respond to hormone treatment had significantly reduced expression of PHLDB1 in cases classified as Luminal A (p = 0.040) (Figure 5a), but there was no relationship in Luminal B tumors (p = 0.054) (Figure 5b). In the evaluation of PHLDB2 related to response rates to endocrine treatment, there was no statistical association in cases subtyped as Luminal A (p = 0.300) and Luminal B (p = 0.054) (Figures 6a and 6b, respectively). Finally, for the last family member, a significant relationship was found between high levels of PHLDB3 for patients who responded to endocrine treatment with tamoxifen or anastrozole (Luminal A, p = 0.047 and Luminal B, p = 0.012) (Figures 7a and 7b, respectively). Furthermore, reduced PHLDB3 expression in HER2+ tumors was correlated with low response rates to anti-HER2 treatment (p = 0.029) (Figure 7c). However, PHLDB1 and PHLDB2 showed no relationship in the response rates of patients with tumors that overexpress HER2 when treated with monoclonal antibodies targeting this receptor (p = 0.710 and p = 0.320, respectively) (Figures 5c and 6c, respectively). Contrary to the effect observed for hormone-dependent and HER2-overexpressing tumors, patients with TNBC-type tumors that did not respond to chemotherapy had significantly increased rates of PHLDB1 (p = 0.009) (Figure 5d) and PHLDB2 (p = 0.034) (Figure 6d), in this particularly more aggressive form of breast cancer. However, for the third family member, no relationship between PHLDB3 differential expression with response to chemotherapeutic treatments was observed in TNBC cases (p = 0.730) (Figure 7d).

Discussion and Conclusion

Despite great advances in the diagnosis, prognosis, prevention and treatment of breast cancer, this type of malignant tumor remains the most prevalent and lethal in women globally (3). In this context, hundreds of other biomarker candidates are being studied for potential implications for improving diagnosis and personalized therapy. In view of this, our study aimed to investigate the expression profile of members of the PHLDB family and the potential prognostic and clinically useful value in breast cancer using bioinformatics tools, taking into account the limitation of studies of members of the PHLDB family in the context of breast oncology and the attractive relationship of these markers as direct and indirect targets of p53 at its transcriptional levels and as competitive modulators of AKT activity by directly interfering in the binding of this oncoprotein to phosphatidylinositol (6).

The PH domain shared by all members of the PHLD family has the ability to anchor itself transiently on the surface of the intracellular membrane and participate in multiple signal transduction processes, being the subject a number of studies (9, 18). To date, the expression pattern in patient samples and the potential prognostic and predictive value of response to different accepted therapies provided by investigating PHLDB family members remain unclear in breast cancer.

Initially, we analyzed the expression profile of the members of the PHLDB family using the UALCAN and GEPIA2 databases. PHLDB1 and PHLDB2 were expressed less in breast tumor samples when compared to healthy tissue. Meanwhile, PHLDB3 was expressed more highly in breast cancer samples. To date, no study has investigated the gene expression profile of the PHLDB family in healthy and tumor samples from the breast and therefore the current study is a pioneer in this sense. Furthermore, our results indicate that the methylation process can serve to repress or activate PHLDB family gene expression in breast tumor samples. It is known that the loss of balance in the methylation of specific regions of DNA can lead to increased predisposition to various diseases and abnormalities, including cancer (19). Another study identified PHLDB2 mRNA as differentially expressed, driven by methylation in uterine corpus endometrial carcinoma (UCEC) samples (20). Together, these observations may indicate that DNA methylation may be an important mechanism of epigenetic regulation of the PHLDB family in breast cancer, requiring further investigation.

Next, the relevance of the expression of PHLDB family members to different clinic pathological characteristics of breast cancer patients was analyzed. It was found that increased expression of the three members of the PHLDB family was significantly correlated with several variables, including lower rates of lymph node involvement and with the lowest degree of SBR and NPI. Routinely in clinical practice, the presence and extent of lymph node metastases are indicators of an aggressive phenotype, generally with an inverse relationship with prognosis (21). Thus, the genes of the PHLDB family, based on this in silico study, are shown to be potential markers for predicting the development of lymph node metastasis and unsatisfactory clinical outcome.

Additionally, our work showed a statistically significant correlation between the increased expression of PHLDB1, 2 and 3 with wild-type TP53 and hormone receptor positivity (ER and PR) and, inevitably, with Luminal subtype tumors. In addition, PHLDB2 and 3 were more highly expressed in tumors with positive HER2 receptor tyrosine kinase classification, while PHLDB1 was inversely correlated compared to its paralogs. Interestingly, in addition to our findings, in previous studies it was observed that MCF-7 malignant breast cells treated with E2 (17β-estradiol) showed a large increase in the expression of PHLDA1 transcripts compared to untreated cells (22) and that ER and NF-κB act synergistically for the direct transcriptional activation of PHLDA1 (23). As for HER2, the picture remains unclear between the relationship between the PHLD family and this tyrosine kinase. However, previous work has already identified that PHLDA2 expression is reduced at transcriptional and protein levels immediately and significantly by suppression of EGFR/HER2 oncogenic signaling in multiple HER2+ breast cancer cell lines (24, 25). These data indicate that members of the PHLD family can act as downstream targets of the EGFR/HER2 oncogenic signaling pathway. Finally, PHLD class proteins have been suggested as direct and indirect targets of p53 at its transcriptional levels by different studies (26, 27), demonstrating a potential critical role in tumorigenesis.

Subsequently, the prognostic significance of PHLDB family members in breast cancer was investigated using the public Kaplan–Meier Plotter and PrognoScan databases. It was found that reduced expression of PHLDB1, 2 and 3 mRNA was associated with decreased rates of OS and RFS in breast cancer patients. Supporting our previous data, the reduced expression of PHLDB family members was identified as critical for OS, RFS, DMFS and DFS reduction by the meta-analysis performed with the PrognoScan online repository. No study to date has evaluated the possible prognostic role of the PHLDB family in breast cancer. However, other works have already convincingly demonstrated that among the paralogs of the PHLDB family, members of the PHLDA family have a possible tumor suppressor role in breast cancer (28, 29, 30). Regarding the prognostic impact on different molecular subtypes, we identified that the reduced expression of PHLDB family members was associated with significantly reduced rates of OS and RFS in patients with Luminal-type tumors. For TNBC subtype tumors, an inverse role was observed, where the increased expression of PHLDB1 and 2 seems to favor a worse prognosis. Finally, among patients with tumors classified as HER2+, increased expression of PHLDB1 and 3 was responsible for worse OS. However, when evaluating these data, we have to take into account that the curves generated for OS and RFS of patients with breast cancer of molecular subtypes TNBC and HER2+ was based on smaller data sets when compared to Luminal-type tumors. Furthermore, we already know that many members of the PHLD family have a pleiotropic mechanism that will depend on the cell, tissue and molecular type and context. These findings provide evidence that PHLDB family members can serve as predictive markers for breast cancer prognosis.

Finally, our results for predicting therapeutic response showed that among patients with tumors classified as hormone-dependent and who were not responsive to endocrine treatment, these cases had lower gene expression for PHLDB1 and PHLDB3. For HER2+ cases, reduced expression of PHLDB3 was observed in samples from patients who did not respond to anti-HER2 antibody therapy. Finally, for the TNBC subtype, high expression of PHLDB1 and PHLDB2 was identified in samples from patients who did not respond to chemotherapeutic agents. So far, we do not know how these markers may be acting in TNBC cases, and in vitro studies are needed to confirm the relationship between PHLDB1 and 2 in the rates of patients’ responses to chemotherapy.

Whereas, the PI3K/AKT/mTOR signaling pathway has been consistently implicated in resistance to several therapies in breast cancer (31) and that proteins with the PH domain can bind to phosphatidylinositol coupled to the surface of the intracellular membrane for suppression of this important oncogenic signaling pathway (9), we can hypothesize that PHLDB1 and 3 appear to be promising molecules to stratify patients who may or may not respond to hormone therapy and anti-HER2 agents. In addition to our findings, other studies have already demonstrated a possible relationship between the members of the PHLD family for therapeutic response in cases of Luminal and HER2+ breast cancer (16, 17, 24, 32).

In summary, this pioneering research revealed that members of the PHLDB family may be promising biomarkers for predicting prognosis and therapeutic response in breast cancer patients. It is important to highlight that in silico and data mining analyzes may have certain limitations, such as the extent and quality of information in publicly available databases, non-pairing of samples and, sometimes, small cohort size. However, our research was able to provide a stimulus, we hope, for possible further in vitro and in vivo studies, necessary for an application in the context of translational medicine in oncology.

Ethics Committee Approval: For this type of project, research ethics committee approval is not required.

Informed Consent: Informed consent was not required for this study.

Peer-review: Externally peer-reviewed.

Authorship Contributions

Concept: R.G.N, J.M., D.O.C.; Design: R.G.N, J.M., D.O.C.; Data Collection and/or Processing: R.G.N, J.M.; Analysis and/or Interpretation: R.G.N, D.O.C., S.J.J.; Literature Search: R.G.N, J.M., D.O.C., S.J.J.; Writing: R.G.N, J.M., D.O.C., S.J.J.

Conflict of Interest: No conflict of interest was declared by the authors.

Financial Disclosure: This study was funded in part by the Institute of Research and Education in Health of São Paulo, the University Center of the National Service for Commercial Learning and the Anhanguera University.

  1. Lei S, Zheng R, Zhang S, Wang S, Chen R, Sun K, et al. Global patterns of breast cancer incidence and mortality: A population-based cancer registry data analysis from 2000 to 2020. Cancer Communications 2021; 41: 1183-1194. (PMID: 34399040)
  2. Nascimento RG, Otoni KM. Histological and molecular classification of breast cancer: what do we know? Mastology 2020; 30: 1-8. (PMID: 31123102)
  3. Harbeck N, Penault-Llorca F, Cortes J, Gnant M, Houssami N, Poortmans P, et al. Breast cancer. Nature Reviews Disease Primers 2019; 5: 1-31. (PMID: 31548545)
  4. Tsang JYS, Tse GM. Molecular Classification of Breast Cancer. Advances in Anatomic Pathology 2020; 27: 1-9. (PMID: 31045583)
  5. Kalia M. Biomarkers for personalized oncology: Recent advances and future challenges. Metabolism Clinical and Experimental 2015; 64: 16-21. (PMID: 25468140)
  6. Fuselier TT, Lu H. PHLD class proteins: A family of new players in the P53 network. International Journal of Molecular Sciences. 2020; 21: 1-10. (PMID: 32429563)
  7. Lemmon MA. Pleckstrin homology domains: Not just for phosphoinositides. Biochemical Society Transactions 2004; 32: 707-711. (PMID: 15493994)
  8. Lemmon MA. Membrane recognition by phospholipid-binding domains. Nature Reviews Molecular Cell Biology 2008; 9: 99-111. (PMID: 16689643)
  9. Jiang Z, Liang Z, Shen B, Hu G. Computational analysis of the binding specificities of PH domains. BioMed Research International 2015; 1: 1-12. (PMID: 26881206)
  10. Chandrashekar DS, Bashel B, Balasubramanya SAH, Creighton CJ, Ponce-Rodriguez I, Chakravarthi BVSK, et al. UALCAN: A Portal for Facilitating Tumor Subgroup Gene Expression and Survival Analyses. Neoplasia. 2017; 19: 649-658. (PMID: 28732212)
  11. Tang Z, Kang B, Li C, Chen T, Zhang Z. GEPIA2: an enhanced web server for large-scale expression profiling and interactive analysis. Nucleic Acids Research 2019; 47: 556-560. (PMID: 31114875)
  12. Jézéquel P, Campone M, Gouraud W, Guérin-Charbonnel C, Leux C, Ricolleau G, et al. Bc-GenExMiner: An easy-to-use online platform for gene prognostic analyses in breast cancer. Breast Cancer Research and Treatment 2012; 131: 765-775. (PMID: 21452023)
  13. Győrffy B. Survival analysis across the entire transcriptome identifies biomarkers with the highest prognostic power in breast cancer. Computational and Structural Biotechnology Journal 2021; 19: 4101-4109. (PMID: 34527184)
  14. Mizuno H, Kitada K, Nakai K, Sarai A. PrognoScan: A new database for meta-analysis of the prognostic value of genes. BMC Medical Genomics 2009; 2: 1-11. (PMID: 19393097)
  15. Fekete JT, Győrffy B. ROCplot.org: Validating predictive biomarkers of chemotherapy/hormonal therapy/anti-HER2 therapy using transcriptomic data of 3,104 breast cancer patients. International Journal of Cancer 2019; 145: 3140-3151. (PMID: 31020993)
  16. Fearon AE, Carter EP, Clayton NS, Wilkes EH, Baker AM, Kapitonova E, et al. PHLDA1 Mediates Drug Resistance in Receptor Tyrosine Kinase-Driven Cancer. Cell Reports 2018; 22: 2469-2481. (PMID: 29490281)
  17. Mangone FR, Valoyes MAV, Nascimento RG, Conceição MPF, Bastos DR, Pavanelli AC, et al. Prognostic and predictive value of Pleckstrin homology-like domain, family A family members in breast cancer. Biomarkers in Medicine 2020; 14: 1537-1552. (PMID: 33179538)
  18. Scheffzek K, Welti S. Pleckstrin homology (PH) like domains - Versatile modules in protein-protein interaction platforms. FEBS Letters 2012; 586: 2662-2673. (PMID: 22728242)
  19. Dhar GA, Saha S, Mitra P, Nag Chaudhuri R. DNA methylation and regulation of gene expression: Guardian of our health. Nucleus 2021; 64: 259-270. (PMID: 34421129)
  20. Zeng Z, Cheng J, Ye Q, Zhang Y, Shen X, Cai J, et al. A 14-Methylation-Driven Differentially Expressed RNA as a Signature for Overall Survival Prediction in Patients with Uterine Corpus Endometrial Carcinoma. DNA and Cell Biology 2020; 39: 975-991. (PMID: 34421129)
  21. Bakkour AM, Surriah MH, Al-Imari ANK, Al-Asadi RRJ. The predictors and the prognostic significance of axillary lymph nodes involvement in breast cancer. International Surgery Journal 2019; 6: 1-5. (PMID: 15812825)
  22. Marchiori AC, Casolari DA, Nagai MA. Transcriptional up-regulation of PHLDA1 by 17beta-estradiol in MCF-7 breast cancer cells. Brazilian Journal of Medical and Biological Research 2008; 41: 579-582. (PMID: 18641796)
  23. Kastrati I, Canestrari E, Frasor J. PHLDA1 Expression is Controlled by an Estrogen Receptor (ER)- NFkB-miR-181 Regulatory Loop and is Essential for Formation of ER+ Mammospheres. Oncogene 2015; 34: 2309-2316. (PMID: 24954507)
  24. Li G, Wang X, Hibshoosh H, Jin C, Halmos B. Modulation of ErbB2 blockade in ErbB2-positive cancers: The role of ErbB2 mutations and PHLDA1. PLoS ONE 2014; 9: 1-13. (PMID: 25238247)
  25. Wang X, Li G, Koul S, Ohki R, Maurer M, Borczuk A, et al. PHLDA2 is a key oncogene-induced negative feedback inhibitor of EGFR/ErbB2 signaling via interference with AKT signaling. Oncotarget 2018; 9: 24914-24926. (PMID: 29861842)
  26. Chen Y, Takikawa M, Tsutsumi S, Yamaguchi Y, Okabe A, Shimada M, et al. PHLDA1, another PHLDA family protein that inhibits Akt. Cancer Science. 2018; 109: 3532-3542. (PMID: 30207029)
  27. Kawase T, Ohki R, Shibata T, Tsutsumi S, Kamimura N, Inazawa J, et al. PH Domain-Only Protein PHLDA3 Is a p53-Regulated Repressor of Akt. Cell 2009; 136: 535-550. (PMID: 19203586)
  28. Nagai MA, Fregnani JHTG, Netto MM, Brentani MM, Soares FA. Down-regulation of PHLDA1 gene expression is associated with breast cancer progression. Breast Cancer Research and Treatment 2007; 106: 49-56. (PMID: 17211533)
  29. Moon HG, Oh K, Lee J, Lee M, Kim JY, Yoo TK, et al. Prognostic and functional importance of the engraftment-associated genes in the patient-derived xenograft models of triple-negative breast cancers. Breast Cancer Research and Treatment 2015; 154: 13-22. (PMID: 26438141)
  30. Christgen M, Noskowicz M, Heil C, Schipper E, Christgen H, Geffers R, et al. IPH-926 lobular breast cancer cells harbor a p53 mutant with temperature-sensitive functional activity and allow for profiling of p53-responsive genes. Laboratory Investigation 2012; 92: 1635-1647. (PMID: 22945757)
  31. Miricescu D, Totan A, Stanescu-Spinu II, Badoiu SC, Stefani C, Greabu M. PI3K/AKT/mTOR signaling pathway in breast cancer: From molecular landscape to clinical aspects. International Journal of Molecular Sciences 2021; 22: 1-24. (PMID: 33375317)
  32. Magi S, Iwamoto K, Yumoto N, Hiroshima M, Nagashima T, Ohki R, et al. Transcriptionally inducible pleckstrin homology-like domain, family a, member 1, attenuates ERBB receptor activity by inhibiting receptor oligomerization. Journal of Biological Chemistry 2018; 293: 2206-2218. (PMID: 30778399)