Quartz 4

pathway_and_expression_markers_in_acc

18 min read

Pathway and Expression Markers in ACC

Immunohistochemistry and Molecular Pathology

Pathway and expression markers in adrenocortical carcinoma (ACC) are tissue-based molecular readouts that describe recurrent signaling, differentiation, metabolic, and microenvironmental programs within malignant adrenocortical tumors. Within ACC care, they belong to immunohistochemistry and molecular pathology and are used as adjuncts to conventional histopathologic assessment rather than as independent diagnostic systems. The most frequently studied groups involve Wnt/β-catenin signaling, IGF-associated growth pathways, developmental or steroidogenic regulators, immune checkpoint expression, and lipid-handling programs.123

In current practice, these markers are mainly relevant to three tasks: supporting the distinction between ACC and adenoma in difficult cortical neoplasms, refining biologic or prognostic stratification within confirmed ACC, and linking routine pathology to translational molecular subtypes.45 Their overall value remains limited by the rarity of ACC and by an evidence base dominated by retrospective single-center cohorts, exploratory immunohistochemical series, and correlative transcriptomic studies focused largely on resected primary tumors rather than metastatic or treated disease.67

Across marker classes, the most consistent finding is that ACC commonly shows dysregulation of developmental and growth-related pathways that are less prominent in benign adrenocortical tumors. Even the better-supported markers, however, do not establish malignancy on their own, and their incremental value beyond morphology, Weiss-based assessment, Ki67, stage, resection status, and hormone secretion is incompletely defined.152 As a result, pathway and expression markers are more established for biologic characterization than for routine treatment selection.

Diagnostic context

ACC diagnosis continues to rely primarily on integrated clinicopathologic assessment, including morphology, invasion, proliferation, and clinical or endocrine context. Pathway and expression markers therefore function as ancillary tools, especially in lesions with borderline or equivocal features, rather than replacing standard diagnostic frameworks.45

The literature supports two main diagnostic uses. First, marker expression may corroborate activation of pathways that are recurrent in ACC. Second, selected combinations may improve confidence in the adenoma-carcinoma distinction when routine criteria are indeterminate, as illustrated by paired approaches such as miR483-3p with low Smad4 expression.4 This adjunctive role is plausible and sometimes clinically useful, but no single marker has sufficient sensitivity, specificity, and standardization to reclassify an adrenal cortical tumor independently. Practically, reliable information comes from integrated panels and concordance with established pathology, not from isolated positive stains.

Major biologic marker groups

Wnt/β-catenin dysregulation is among the most reproducible molecular themes in ACC. Nuclear β-catenin accumulation correlates with CTNNB1 or APC pathway alteration, is more typical of ACC than benign cortical tumors, and has been associated with higher stage, higher Weiss score, necrosis, increased mitotic activity, and worse survival.15 LEF-1 appears to provide a related downstream readout of pathway activation and has also been linked to adverse outcome and suppression of immune-related transcriptional programs.3

Altered N-cadherin expression and abnormal β-catenin localization further suggest that adhesion loss and Wnt activation are interconnected features of malignant cortical progression.8 Among currently studied marker classes, this group has some of the strongest biologic and diagnostic support. Even so, reliability is highest for identifying a recurrent ACC-associated program, not for proving carcinoma in isolation; the practical implication is that Wnt-related staining may be most informative when pathology is otherwise suspicious but not definitive.

IGF-associated growth signaling

IGF-pathway activation is another recurring feature of ACC biology. IGF2 overexpression, along with FGFR4 and STAT3 upregulation in some series, supports a proliferative and angiogenesis-associated phenotype that is more common in carcinoma than adenoma.910 In adult cohorts, FGFR4 overexpression or amplification has also been associated with poorer survival, suggesting a link to clinically aggressive disease.9

Diagnostic support may be improved when IGF-related markers are combined with other assays rather than interpreted alone; miR483-3p, which correlates with IGF2 expression, is an example of a marker that may add specificity in borderline tumors when paired with reduced Smad4.4 The reliable conclusion is that IGF-associated signaling is common in ACC, but thresholds and standalone utility remain uncertain. Clinically, these findings mainly strengthen biologic suspicion of malignancy or aggressiveness rather than altering first-line management.

Developmental, differentiation, and proliferative programs

ACC also shows altered expression of developmental and steroidogenic regulators, including GATA-6, SF-1, SMAD-family proteins, and inhibin-related pathways, consistent with partial loss of normal cortical differentiation alongside re-expression of embryonal or dysregulated developmental programs.11 DLK1 overexpression fits this pattern and is increased in ACC relative to normal adrenal tissue and some benign adrenal lesions, although expression is heterogeneous and does not map cleanly onto all molecular subtypes.6

Additional exploratory markers such as STMN1 and ERFE suggest that more aggressive ACC may engage cell-cycle, cytoskeletal, extracellular matrix, and developmental networks.1213 These observations are biologically coherent, but validation is less mature than for Wnt-related markers. In practical terms, this group is more reliable for research classification and hypothesis generation than for routine diagnosis or prognosis, and indirect endocrine or cross-species marker data mainly provide context rather than clinically validated assays.14151617

Immune checkpoint and inflammatory markers

Attention has increasingly shifted from tumor-intrinsic pathways to the ACC microenvironment. Immune-marker studies generally describe ACC as having a limited or heterogeneous immune phenotype, with low or variable tumor-cell PD-L1 expression, somewhat more frequent PD-L2 expression in some cohorts, and inconsistent associations with stage or survival.18197

Broader checkpoint profiling indicates that PD-1, PD-L1, and CTLA-4 expression may correlate with tumor-infiltrating lymphocyte subsets, but prognostic signals remain modest and not consistently aligned across studies.7 Lower inflammatory infiltrate scores in ACC than in adenoma have also been reported, raising the possibility that immune-cell quantification could complement morphology in selected differential-diagnostic settings.20 The most reliable conclusion is that immune heterogeneity exists, but these markers are not dependable predictors of immunotherapy benefit or routine diagnostic classification. Their current clinical implication is primarily descriptive characterization of the tumor microenvironment.

Lipid metabolism and steroid-associated markers

Metabolic reprogramming is another recurrent theme, particularly in relation to steroidogenesis and cholesterol handling. SOAT1 has the clearest evidence within this category, with strong expression associated with cortisol secretion, advanced stage, higher Ki67, and worse overall and recurrence-free survival.2 This pattern suggests that lipid-storage and cholesterol-processing pathways may track with aggressive steroidogenic ACC phenotypes.

Other steroid-associated or lipid-related markers in the literature are less directly informative for human ACC and are supported largely by indirect, historical, or non-ACC evidence.21222324 Accordingly, what is reliable here is the broader principle that metabolic and steroidogenic programs are altered in ACC; what is not yet reliable is broad clinical deployment of most individual markers beyond SOAT1. The practical implication is that these findings mainly aid prognostic research and biologic subclassification.

Evidence quality and interpretive limits

The literature is constrained by small sample sizes, retrospective design, heterogeneous antibodies and scoring methods, and variable positivity thresholds.67 Many reported associations have not been tested against established prognostic variables in large multivariable models, and most data derive from primary untreated surgical specimens. This makes broad biologic themes more reproducible than any precise cutoff or single-test algorithm.

Another recurring limitation is that pathway expression does not by itself establish therapeutic actionability. A marker may indicate pathway engagement without proving sensitivity to targeted therapy or immunotherapy, and some cited literature is best interpreted as historical or cross-tumor context rather than ACC-specific validation.25262728 The practical consequence is that marker results should be interpreted within multidisciplinary pathology and oncology workflows and weighed against standard clinicopathologic assessment.

Role in management and research

Taken together, pathway and expression markers may supplement routine pathology in diagnostically difficult adrenal cortical lesions, enrich retrospective risk stratification, and connect histologic findings to emerging molecular subtypes of ACC.452 Their most credible current uses are as adjunctive tissue correlates of known ACC biology, especially for Wnt activation and selected metabolic or growth-associated programs. They are not established replacements for morphology, Ki67, staging, endocrine evaluation, or standard treatment planning.

Their larger present role is in translational research. By identifying recurrent biologic programs and possible vulnerabilities, these markers help define ACC heterogeneity in a rare malignancy where prospective validation is difficult.133 Future clinical utility will depend on external validation, assay standardization, and demonstration that marker-defined subsets provide actionable information beyond existing clinicopathologic models.

Included Articles

  • PMID 21088256: In resected primary ACC, nuclear beta-catenin staining by immunohistochemistry marks Wnt pathway activation and correlates with CTNNB1 or APC alterations, higher stage, higher Weiss score, necrosis, and increased mitotic activity. Across two cohorts, beta-catenin nuclear positivity was independently associated with shorter overall and disease-free survival.1
  • PMID 22241720: In pediatric and adult adrenocortical tumors, FGFR4 overexpression was frequent and correlated strongly with IGF2 expression, while FGFR4 gene amplification was present in a subset of cases. In adults, both FGFR4 overexpression and amplification were associated with poorer overall survival, supporting their relevance as molecular prognostic markers.9
  • PMID 23866946: In a small human adrenocortical tumor series, immunohistochemistry and RT-PCR linked ACC aggressiveness to reduced SMAD3, GATA-6, SF-1, LRP5/LRP6, and INHA-related signaling, while SMAD2 increased with higher Weiss score. In ACCs, diminished GATA-6 and SF-1 expression correlated with poor outcome.11
  • PMID 24102952: This retrospective study found that miR483-3p overexpression and low or negative Smad4 expression can complement routine pathology when distinguishing adrenocortical carcinoma from adenoma, especially in borderline tumors with equivocal Weiss scores. miR483-3p correlated with IGF2 expression, and the combined miR483-3p/Smad4 pattern improved diagnostic specificity.4
  • PMID 24178245: In a small immunohistochemical series of sporadic adrenocortical tumors, ACC showed more frequent STAT3 and IGF2 expression than adenoma, and both markers correlated positively with CD34-based microvessel density. These findings support angiogenesis-related molecular differences in ACC and suggest possible investigational targets.10
  • PMID 25767716: An exploratory immunohistochemical study found PD-L1 expression in ACC on tumor cell membranes in 10.7% of cases and in tumor-infiltrating mononuclear cells in 70.4%. In this small retrospective cohort, PD-L1 positivity was not associated with stage, grade, hormone secretion, or 5-year overall survival.18
  • PMID 26317117: In adrenal cortical neoplasms, CTNNB1 mutations were found in 39% of ACCs, and strong nuclear beta-catenin staining correlated with mutation status with 83% sensitivity and 94% specificity. The study suggests beta-catenin immunolocalization may help evaluate diagnostically uncertain adrenal cortical lesions.5
  • PMID 27468715: This comparative study of adrenocortical tumors found frequent N-cadherin downregulation in ACC and recurrent cytoplasmic or nuclear beta-catenin accumulation, with CTNNB1 alterations also identified. The findings support cadherin/Wnt-pathway dysregulation as a molecular feature of ACC and a potential source of diagnostic or prognostic biomarkers.8
  • PMID 30413322: An exploratory immunohistochemistry study found PD-L2 expression in 44% of ACC tumor samples and 53% when tumor or stromal positivity was considered, whereas PD-L1 was detected in only one ACC. PD-L2 expression was not associated with CD8 infiltration, hormone secretion, metastases at diagnosis, age, or sex.19
  • PMID 31265901: This study identifies marked DLK1 overexpression in adrenocortical carcinoma compared with normal adrenal tissue and aldosterone-producing adenomas, supported by immunohistochemistry, western blotting, and transcriptomic datasets. DLK1 expression was heterogeneous across ACC and did not differ between ENSAT C1A and C1B molecular subtypes in the analyzed cohort.6
  • PMID 31963898: In a tertiary-center cohort of 112 adult ACCs, strong SOAT1 protein expression by immunohistochemistry was heterogeneous but associated with cortisol secretion, advanced ENSAT stage, higher Ki67, and independently worse overall survival and recurrence-free survival. The study positions SOAT1 as a prognostic tissue biomarker with possible therapeutic relevance.2
  • PMID 32825943: This immunohistochemical study of adrenal neoplasms found lower inflammatory infiltrate scores in adrenocortical carcinomas than in adenomas, using CD45, CD3, and CD20 staining. The authors propose inflammatory immunoscoring as a potentially more objective adjunct to morphology in difficult adrenal cortical differential diagnosis, with limited supportive data on PD-L1.20
  • PMID 36675239: Pan-cancer in silico analysis identified ERFE mRNA overexpression as an independent adverse prognostic signal in ACC, associated with higher tumor mutational burden and enrichment of cell-cycle, extracellular matrix, NOTCH, WNT, and PI3K-AKT related programs. In ACC, combined ERFE-high and TMB-high status marked particularly poor survival metrics.13
  • PMID 38415841: In a large ACC immunohistochemistry series, PD-1, PD-L1, and CTLA-4 showed heterogeneous expression, with PD-1 and PD-L1 correlating with tumor-infiltrating T-cell subsets. PD-1 positivity was independently associated with longer progression-free survival, whereas PD-L1 and CTLA-4 were not linked to outcome.7
  • PMID 40130164: In ACC tissue, nuclear LEF-1 correlated with CTNNB1 expression and was proposed as a practical biomarker of activated Wnt/beta-catenin signaling. High LEF-1 expression was associated with glucocorticoid-secreting tumors, downregulation of multiple immune-related transcripts including antigen-presentation genes, and worse overall survival.3
  • PMID 14611675: An ovarian carcinoma study identified DAX-1 as an independent adverse prognostic marker and reviewed its established expression in steroidogenic tissues and adrenocortical tumors. For ACC, the article is only indirectly informative, but it supports inclusion of DAX-1 as a steroidogenesis-related contextual marker rather than a validated ACC biomarker.21
  • PMID 3860843: A 1985 rat study described AUT-PK 500, a serine kinase originally purified from rat adrenocortical carcinoma, and found markedly increased expression in that model. However, the kinase was also elevated in multiple other rat tumors and premalignant lesions, making its relevance to human ACC indirect and primarily historical.25
  • PMID 40814374: A colorectal cancer study identified CCDC78 as a prognostic biomarker and noted pan-cancer upregulation including in ACC, with enrichment of WNT, JAK-STAT, and immune-related pathways. For ACC, the finding is indirect and mainly highlights a possible cross-tumor transcriptomic signal rather than a validated ACC marker.26
  • PMID 8355444: A 1993 biochemical study found specific nuclear progesterone-binding activity in cortisol-secreting adrenocortical adenomas but not in nonfunctioning or aldosterone-producing adenomas by standard assay, with possible very low-level binding in normal adrenal tissue detectable by thin-layer chromatography. The article is not an ACC study, but it provides historical context for steroid-associated adrenal marker literature and its limited tumor specificity.24
  • PMID 15232151: A ferret adrenocortical tumor study found GATA-4 expression in most carcinomas, especially anaplastic and myxoid areas, with no expression in normal cortex or nodular hyperplasia. Its relevance to human ACC is indirect, but it supports the historical nonhuman literature linking developmental transcription factors to aggressive adrenocortical tumor phenotypes.16
  • PMID 17992631: A hepatocyte-focused study identified AOX1 as an ABCA1-interacting protein and noted coexpression of both proteins in normal adrenocortical cells. Because the tumor analyses were performed in hepatocellular carcinoma rather than ACC, its relevance to ACC is indirect and mainly supports broader lipid-metabolism and differentiation-related context.22
  • PMID 18246525: A 2008 study found ghrelin expression in normal adrenal gland and multiple benign adrenal tumor types, while GHS-R1a protein was usually absent. The work did not include ACC, so it mainly provides indirect context that peptide-expression signals in adrenal tissue are not necessarily ACC-specific.23
  • PMID 18827407: A prostate cancer immunohistochemistry study found nuclear DAX1 expression in 53% of tumors, with lower expression associated with higher Gleason score and no significant correlation with AR, ERβ, or PR. For ACC, the relevance is indirect and mainly supports DAX-1 as a broader steroidogenic regulatory marker rather than an ACC-specific biomarker.29
  • PMID 25010021: A veterinary study in dogs found nuclear SF-1 expression in normal adrenal cortex and cortisol-secreting adrenocortical tumors, with no overall mRNA difference between normal tissue, adenomas, and carcinomas. Higher SF-1 mRNA in a small carcinoma follow-up subset was associated with earlier recurrence, providing indirect comparative support for SF-1 as a prognostic rather than purely diagnostic marker context in adrenocortical neoplasia.17
  • PMID 26143682: A canine study of cortisol-secreting adrenocortical tumors found reduced SSTR1 in adenomas and carcinomas, reduced DRD2 in carcinomas, and increased DRD1 expression in recurrent carcinomas, without clear overexpression of progesterone or growth hormone receptor transcripts. For human ACC, this is indirect comparative evidence that broadens the historical endocrine-marker context rather than defining a validated marker.28
  • PMID 29214451: A 2018 study identified STMN1 as markedly overexpressed in ACC relative to benign adrenal tissue across RNA, protein, and immunohistochemical analyses. Functional knockdown in NCI-H295R cells reduced multiple in vitro features of aggressiveness, suggesting biologic relevance but not yet establishing clinical utility.12
  • PMID 19849700: MicroRNA profiling in massive macronodular adrenocortical disease identified dysregulated adrenal regulatory RNAs and showed direct miR-200b targeting of MATR3 in the H295R ACC cell line. The study is indirect for ACC but supports a post-transcriptional regulatory layer relevant to developmental and cytoskeletal programs in adrenocortical tumorigenesis.14
  • PMID 25364424: A breast cancer study found HURP overexpression associated with more advanced stage and in vitro proliferative effects after knockdown. Its relevance to ACC is indirect, serving mainly as contextual support for proliferation-related spindle programs rather than validated ACC marker use.30
  • PMID 26793165: A review of endocrine tumors describes PI3K/Akt signaling as relevant to ACC and notes phospho-Akt immunohistochemistry as a potential pathway readout, but emphasizes variable evidence quality, fixation-related assay limitations, and uncertain prognostic standardization for ACC-specific use.15
  • PMID 39130148: A pancreatic cancer study found CRHR1 expression associated with poorer survival on univariate analysis, but not multivariate analysis, and cited prior CRHR1 immunoreactivity in adrenal cortical tumors. For ACC, this serves only as indirect cross-tumor context supporting cautious interpretation of endocrine pathway markers rather than disease-specific evidence.27

References

Footnotes

  1. β-catenin activation is associated with specific clinical and pathologic characteristics and a poor outcome in adrenocortical carcinoma.. Clin Cancer Res. 2011. PMID: 21088256. Local full text: 21088256.md ↩ ↩2 ↩3 ↩4

  2. Sterol O-Acyl Transferase 1 as a Prognostic Marker of Adrenocortical Carcinoma.. Cancers (Basel). 2020. PMID: 31963898. Local full text: 31963898.md ↩ ↩2 ↩3 ↩4 ↩5

  3. Wnt/β-catenin pathway activation is associated with glucocorticoid secretion in adrenocortical carcinoma, but not directly with immune cell infiltration.. Front Endocrinol (Lausanne). 2025. PMID: 40130164. Local full text: 40130164.md ↩ ↩2 ↩3 ↩4

  4. Distinguishing adrenal cortical carcinomas and adenomas: a study of clinicopathological features and biomarkers.. Histopathology. 2014. PMID: 24102952. Local full text: 24102952.md ↩ ↩2 ↩3 ↩4 ↩5 ↩6

  5. Genomic and immunohistochemical analysis in human adrenal cortical neoplasia reveal beta-catenin mutations as potential prognostic biomarker.. Discoveries (Craiova). 2015. PMID: 26317117. Local full text: 26317117.md ↩ ↩2 ↩3 ↩4 ↩5 ↩6

  6. DLK1/PREF1 marks a novel cell population in the human adrenal cortex.. J Steroid Biochem Mol Biol. 2019. PMID: 31265901. Local full text: 31265901.md ↩ ↩2 ↩3 ↩4

  7. Expression and Prognostic Relevance of PD-1, PD-L1, and CTLA-4 Immune Checkpoints in Adrenocortical Carcinoma.. J Clin Endocrinol Metab. 2024. PMID: 38415841. Local full text: 38415841.md ↩ ↩2 ↩3 ↩4 ↩5

  8. Investigation of N-cadherin/β-catenin expression in adrenocortical tumors.. Tumour Biol. 2016. PMID: 27468715. Local full text: 27468715.md ↩ ↩2

  9. The role of fibroblast growth factor receptor 4 overexpression and gene amplification as prognostic markers in pediatric and adult adrenocortical tumors.. Endocr Relat Cancer. 2012. PMID: 22241720. Local full text: 22241720.md ↩ ↩2 ↩3

  10. Expression of STAT3 and IGF2 in adrenocortical carcinoma and its relationship with angiogenesis.. Clin Transl Oncol. 2014. PMID: 24178245. Local full text: 24178245.md ↩ ↩2

  11. Expression of Wnt and TGF-β pathway components and key adrenal transcription factors in adrenocortical tumors: association to carcinoma aggressiveness.. Pathol Res Pract. 2013. PMID: 23866946. Local full text: 23866946.md ↩ ↩2

  12. STMN1 is Overexpressed in Adrenocortical Carcinoma and Promotes a More Aggressive Phenotype In Vitro.. Ann Surg Oncol. 2018. PMID: 29214451. Local full text: 29214451.md ↩ ↩2

  13. In Silico Pan-Cancer Analysis Reveals Prognostic Role of the Erythroferrone (ERFE) Gene in Human Malignancies.. Int J Mol Sci. 2023. PMID: 36675239. Local full text: 36675239.md ↩ ↩2 ↩3

  14. MicroRNA signature in massive macronodular adrenocortical disease and implications for adrenocortical tumourigenesis.. Clin Endocrinol (Oxf). 2010. PMID: 19849700. Local full text: 19849700.md ↩ ↩2

  15. The PI3K/Akt Pathway in Tumors of Endocrine Tissues.. Front Endocrinol (Lausanne). 2015. PMID: 26793165. Local full text: 26793165.md ↩ ↩2

  16. Transcription factor GATA-4 is a marker of anaplasia in adrenocortical neoplasms of the domestic ferret (Mustela putorius furo).. Vet Pathol. 2004. PMID: 15232151. Local full text: 15232151.md ↩ ↩2

  17. Expression of steroidogenic factor 1 in canine cortisol-secreting adrenocortical tumors and normal adrenals.. Domest Anim Endocrinol. 2014. PMID: 25010021. Local full text: 25010021.md ↩ ↩2

  18. Programmed death ligand-1 expression in adrenocortical carcinoma: an exploratory biomarker study.. J Immunother Cancer. 2015. PMID: 25767716. Local full text: 25767716.md ↩ ↩2

  19. Expression of programmed death ligand 1 and 2 in adrenocortical cancer tissues: An exploratory study.. Surgery. 2019. PMID: 30413322. Local full text: 30413322.md ↩ ↩2

  20. Inflammation in the neoplasms of the adrenal gland: Is there a prognostic role? An immunohistochemical study.. Pathol Res Pract. 2020. PMID: 32825943. Local full text: 32825943.md ↩ ↩2

  21. Nuclear receptor DAX-1 in human common epithelial ovarian carcinoma: an independent prognostic factor of clinical outcome.. Cancer Sci. 2003. PMID: 14611675. Local full text: 14611675.md ↩ ↩2

  22. Human aldehyde oxidase 1 interacts with ATP-binding cassette transporter-1 and modulates its activity in hepatocytes.. Horm Metab Res. 2007. PMID: 17992631. Local full text: 17992631.md ↩ ↩2

  23. Differential expression of ghrelin and its receptor (GHS-R1a) in various adrenal tumors and normal adrenal gland.. Horm Metab Res. 2008. PMID: 18246525. Local full text: 18246525.md ↩ ↩2

  24. [Expression of progesterone-binding protein in normal and neoplastic human adrenals].. Nihon Hinyokika Gakkai Zasshi. 1993. PMID: 8355444. Local full text: 8355444.md ↩ ↩2

  25. Expression of autophosphorylating protein kinase 500 in normal and neoplastic rat cells.. Proc Natl Acad Sci U S A. 1985. PMID: 3860843. Local full text: 3860843.md ↩ ↩2

  26. Transcriptomic validation and clinical translation of CCDC78 as a prognostic biomarker in colorectal cancer.. Am J Cancer Res. 2025. PMID: 40814374. Local full text: 40814374.md ↩ ↩2

  27. Expression of Corticotropin-Releasing Hormone and Its Receptors May Be Associated With Survival Rate in Pancreatic Cancer.. Gastro Hep Adv. 2023. PMID: 39130148. Local full text: 39130148.md ↩ ↩2

  28. Expression of somatostatin, dopamine, progesterone and growth hormone receptor mRNA in canine cortisol-secreting adrenocortical tumours.. Vet J. 2015. PMID: 26143682. Local full text: 26143682.md ↩ ↩2

  29. Nuclear receptor DAX1 in human prostate cancer: a novel independent biological modulator.. Endocr J. 2009. PMID: 18827407. Local full text: 18827407.md ↩

  30. Hepatoma upregulated protein expression is involved in the pathogenesis of human breast carcinogenesis.. Oncol Lett. 2014. PMID: 25364424. Local full text: 25364424.md ↩

Graph View