Review Article

Author: Mariaelena Pierobon, MD; Leena Gandhi, MD; Massimo Cristofanilli, MD; Debu Tripathy, MD; Emanuel F Petricoin III, PhD


Molecular studies mapping malignant lesions of similar anatomical origin have led to the understanding that tumors originating from the same organ can be extremely heterogeneous at the molecular level.1 This underpinning molecular heterogeneity is considered one of the underlying causes of high discrepancy in treatment success rates across patients with lesions that have similar histopathologic characteristics.2,3 At the same time, broad-scale molecular profiling efforts such as The Cancer Genome Atlas (TCGA) have revealed that tumors originating from different organs also share common molecular alterations.4 For these reasons, drugs such as trastuzumab (Herceptin®), a monoclonal antibody targeting the HER2 protein, is FDA-approved for treating patients with breast or gastric cancer overexpressing the HER2 protein. This approval may be broadened in the near future to other HER2 overexpressing solid tumors and to a subpopulation of patients presenting with HER2 mutations.5-7 Because these molecular characteristics play such a central role in determining the most appropriate therapeutic approach for patients with cancer, upfront molecular profiling is becoming a de facto part of the standard of care work-up for classifying and describing malignant lesions, and true oncogenic drivers are being used as key targets for effective therapy.
Although a number of DNA mutations and chromosomal amplifications/translocations are recognized molecular ‘drivers’ of many cancers, it is impossible to distinguish in any given tumor which alterations are drivers of the disease and which adjustments were necessary in the early stage of the malignancy, but are not any more necessary or sufficient to sustain tumor progression. Despite the development of high-throughput genomic technologies that have opened new opportunities for precision medicine, the results of these analyses can generate unclear data for the treating physician because multiple genomic abnormalities are often identified within the same lesion.8 Moreover, different organ sites, with ostensibly the same driving genomic event, often show diverse response rates to the same targeted compound.9 Therefore, the identification of genomic changes in isolation may be only partially sufficient for stratifying patients to the most appropriate line of treatment.
Finally, while genomic alterations play a central role in cancer progression, the overall cellular signaling architecture is comprised of proteins, and genetic aberrations ultimately exhibit their phenotypic consequences through the action of activated proteins. Moreover, the activation of specific signaling pathways in tumor cells via feedback mechanisms or the activation of cellular receptor by their ligands are often independent from genomic alterations. For these reasons, kinase-driven activation via phosphorylation represents the most direct means for measuring pathway activation in tumor cells.10,11 Because derangements in cellular signaling processes are most often the causal aspect of the disease and thus targets for new therapies, molecular analysis of the proteome and phosphoproteome are becoming centrally important for identifying new predictive biomarkers as well as new therapeutic targets.12 Therefore, there is an urgent need to find more effective tools for exploring the impact of genetic changes on the function of their protein products.

Planar/Suspension Antibody-Based Arrays

So far, planar or suspension antibody-based arrays are the most commonly used platforms for measuring protein kinase signaling in clinical samples.13 A major advantage of these platforms is their ability to generate high-throughput, and in most cases, multiplexed data allowing for the concomitant analysis of hundreds of analytes across a large number of samples. Although antibody specificity can be challenging from time to time, these methodologies have the advantage of accurately measuring low abundance proteins with detection limit in the range of ng-pg/mL. As recently described by Pierobon and colleagues,14 when compared to other antibody-based assays suitable for the analysis of clinical samples, the Reverse Phase Protein Microarray (RPPA) has distinguished itself for the ability of reproducibly and sensitively quantifying the activation level of hundreds of kinases starting from a limited amount of biological materialfrom hundreds of patients at once.14,15 As such, this technology captures the linear dynamic range of most targetable kinases and downstream substrates.16
Because standard curve and internal controls for quality assurance can be mounted on each array, this platform is been used as a Clinical Laboratory Improvement Amendments (CLIA)/College of American Pathologists (CAP)-based technology for measuring protein signaling activation of individual tumors and for identifying patients who can benefit from specific targeted treatment (Figure 1).8

Upfront tissue processing techniques such as Laser Capture Microdissection (LCM) have been effectively coupled to the RPPA workflow and found to be an essential and necessary component in generating accurate protein activation-based clinical data, especially when used to guide treatment selection.17,18 The use of core needle biopsies, even under Formaldehyde Fixed-Paraffin Embedded (tissue) (FFPE) fixation, is an appropriate process for the optimal preservation of the phosphoproteome, and has been successfully analyzed using an LCM/RPPA workflow.16,19,20
We recently used the RPPA technology to measure the activation of HER2 in patients with inflammatory breast cancer (IBC) and non-small cell lung cancers (NSCLC) harboring HER2 mutations by evaluating the level of phosphorylation at the tyrosine 1248 residue, a well-described site of modification known to control and modulate transmission of downstream signaling (Figure 2).21  In particular, we evaluated whether measuring the activation level of HER2 is clinically relevant and has added value to conventional genomic characterization in terms of outcome prediction. For these case studies, the protein activation level of HER2 in the tested samples was compared to the distribution of activated HER2 in a large cohort of breast cancer samples with known HER2 expression and amplification (data not shown). In brief, activation and dimerization of HER2 with other receptor tyrosine kinases (RTKs), including other members of the HER family, regulate a number of important cellular processes by modulating different signaling cascades. In particular, HER2 can stimulate cell proliferation by activating the MAP kinase pathway. Activation of the AKT/mTOR kinase pathway, on the other hand, regulates cell survival. Because most of the members of the MAP and AKT/mTOR pathways are kinases, their activity depends on the phosphorylation status of each of the pathway components.

HER2 Mutation and Activating Base Substitutions

Ali and colleagues recently reported the first IBC case with HER2 mutation and activating base substitutions.22 The patient had extensive local and distant recurrent disease classified by standard pathologic testing as a triple-negative tumor that had progressed despite several cytotoxic treatment regimens. After the molecular test results and the implementation of HER2 targeted therapy, she showed remarkable symptomatic improvement and clinical response. RPPA analysis of a tumor biopsy collected from the same lesion showed an HER2 activation level similar to those seen in breast cancer tumors with overexpression/amplification of HER2, a subgroup of patients who routinely benefit from anti-HER2 targeted treatments (Figure 3). These data indicated that a small subgroup of IBC patients with unamplified (by fluorescent in situ hybridization [FISH] and immunohistochemistry [IHC]), but mutated HER2 and high levels of HER2 protein activation, could benefit from anti-HER2 treatment.
HER2 activation was then evaluated in 2 patients with NSCLC harboring HER2 mutations. Both patients were treated with the small kinase inhibitor targeting HER2, neratinib, in combination with a downstream mTOR inhibitor, temsirolimus, as part of a phase I study.23 Retrospective analysis by RPPA showed heterogeneity in terms of HER2 phosphorylation/ activation across the 2 clinical specimens. In the first sample, the activation level of HER2 was comparable to that in patients with HER2 FISH/IHC-negative breast cancer (Figure 3). As expected, because the drug target was not activated, the patient did not benefit from treatment with the HER2 inhibitor. Indeed, the lack of activation of the receptor indicates that, although a HER2 mutation was present, this genetic event did not result in high levels of activation/phosphorylation of the receptor in this particular lesion. The development of feedback mechanisms in the receptor tyrosine kinases expression and/or turnover as well as the interaction with specific ligands are a few possible explanations for these findings. On the contrary, the activation level of HER2 in the second NSCLC patient analyzed was similar to activation levels seen in patients with breast cancer with overexpressed/amplified HER2 (Figure 3). Increased activation of HER2 in this patient was associated with clinical response that lasted close to a year, consistent with the benefit seen with other effective targeted therapies against true oncogenic drivers.


As shown by our data, directly measuring protein signaling activity in human samples has the unique advantage of providing an ex vivo readout of the in vivo cellular signaling network even in patients presenting with similar genomic characteristics, and provides information on actual drug-target activation in the presence or absence of genomic alterations. This approach could greatly assist in identifying responding patients missed by genomic-only means as well as help credential and prioritize which genomic alterations are truly functional drivers of malignant progression. Moreover, pathway activation as well as the activation of proteins like HER2 (regardless of their expression) are much more frequent events than genomic alterations, including the rare frequency of HER2 mutation.16 The addition of functional proteomic analysis to precision medicine is poised to alter the landscape of clinical trials and routine molecular profiling that rely on genomic analysis alone, because superior and increased clinical benefit using the addition of proteomics and phosphoproteomics to genomic analysis is beginning to be demonstrated. As shown in the published results from a recent clinical trial where treatment selection was recommended based on integrated proteomic, phosphoproteomic, and genomic data, this ‘multi-omic’ approach can improve progression-free survival in patients with cancer, providing optimism for the impact of this strategy going forward.24
Affiliations: Mariaelena Pierobon, MD, and Emanuel F. Petricoin III, PhD, are from Center for Applied Proteomics and Molecular Medicine, George Mason University, Manassas, VA. Leena Gandhi, MD, is from Department of Medical Oncology, Thoracic Oncology Section, Dana-Farber Cancer Institute, Harvard Medical School, Boston, MA. Massimo Cristofanilli, MD, is from Department of Medicine, Division of Hematology and Oncology, Robert H Lurie Comprehensive Cancer Center, Northwestern University, Chicago, IL. Debu Tripathy, MD, is from Department of Breast Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX.
Disclosures: Dr Pierobon has stock options in Theranostics Health Inc. Dr Pierobon is a paid consultant to Perthera Inc. Dr Pierobon is a co-inventor on technologies licensed from Theranostics and receives royalty fee distributions under US law. Dr Gandhi is on the scientific advisory boards (SAB) for Pfizer, Abbvie, Astra-Zeneca, Genentech/Roche, and Merck. Dr Gandhi receives research funding from Bristol-Myers Squibb. Dr Cristofanilli is a consultant for Agendia. Dr Tripathy has no relevant financial relationships to disclose. Dr Petricoin is a co-founder and equity interest holder in Theranostics Health Inc, and receives compensation from them as well as serving on its SAB. Dr Petricoin is a co-founder and equity interest holder in Perthera Inc, and serves as its chief science officer, and on its SAB. Dr Petricoin is a co-inventor of RPPA technology and receives royalty fee distributions under US law.
Corresponding Author: Emanuel Petricoin, PhD, Center for Applied Proteomics and Molecular Medicine, George Mason University, Institute for Biomedical Innovation, 10920 George Mason Circle, Room 2006, Manassas, VA 20110. Phone: 703-993-8606; Fax: 703-993-8606. E-mail:


  1. Burrell RA, McGranahan N, Bartek J, Swanton C. The causes and consequences of genetic heterogeneity in cancer evolution. Nature. 2013;501(7467):338-345. doi: 10.1038/nature12625. 
  2. Berger MF, Lawrence MS, Demichelis F, et al. The genomic complexity of primary human prostate cancer. Nature. 2011;470(7333):214-220. doi: 10.1038/nature09744. 
  3. Cancer Genome Atlas Research Network. Comprehensive molecular characterization of urothelial bladder carcinoma. Nature. 2014;507(7492):315-322. doi: 10.1038/nature12965. 
  4. Kandoth C, McLellan MD, Vandin F, et al. Mutational landscape and significance across 12 major cancer types. Nature. 2013; 502(7471):333-339. doi: 10.1038/nature12634. 
  5. Gomez-Martín C, Lopez-Rios F, Aparicio J, et al. A critical review of HER2-positive gastric cancer evaluation and treatment: from trastuzumab, and beyond. Cancer Lett. 2014;351(1):30-40. doi: 10.1016/j.canlet.2014.05.019. 
  6. Siena S, Sartore-Bianchi A, Lonardi S, et al. Trastuzumab and lapatinib in HER2-amplified metastatic colorectal cancer patients (mCRC): The HERACLES trial. J Clin Oncol. 2015;33:(suppl) Abstract 3508. 
  7. Bose R, Kavuri SM, Searleman AC, et al. Activating HER2 mutations in HER2 gene amplification negative breast cancer. Cancer Discov. 2013;3(2):224-237. doi: 10.1158/2159-8290.CD-12-0349. 
  8. Arnedos M, Vicier C, Loi S, et al. Precision medicine for metastatic breast cancer-limitations and solutions. Nat Rev Clin Oncol. 2015;12(12):693-704. doi: 10.1038/nrclinonc.2015.123. 
  9. Hyman DM, Puzanov I, Subbiah V, et al. Vemurafenib in multiple nonmelanoma cancers with BRAF V600 mutations. N Engl J Med. 2015;373(8):726-736. doi: 10.1056/NEJMoa1502309.
  10. Jhaveri TZ, Woo J, Shang X, Park BH, Gabrielson F. AMP-activated kinase (AMPK) regulates activity of HER2 and EGFR in breast cancer. Oncotarget. 2015;6(17):14754-14765. 
  11. Gregory CW, Whang YE, McCall W, et al. Heregulin-induced activation of HER2 and HER3 increases androgen receptor transactivation and CWR-R1 human recurrent prostate cancer cell growth. Clin Cancer Res. 2005;11(5):1704-1712. 
  12. Kolch W, Pitt A. Functional proteomics to dissect tyrosine kinase signalling pathways in cancer. Nat Rev Cancer. 2010;10(9):618-629. doi: 10.1038/nrc2900. 
  13. Pierobon M, Wulfkuhle J, Liotta L, Petricoin F. Application of molecular technologies for phosphoproteomic analysis of clinical samples. Oncogene. 2015;34(7):805-814. doi: 10.1038/ onc.2014.16. 
  14. Pierobon M, Belluco C, Liotta LA, Petricoin EF 3rd. Reverse phase protein microarrays for clinical applications. Methods Mol Biol. 2011;785:3-12. doi: 10.1007/978-1-61779-286-1_1. 
  15. VanMeter A, Signore M, Pierobon M, Espina V, Liotta LA, Petricoin EF 3rd. Reverse-phase protein microarrays: application to biomarker discovery and translational medicine. Expert Rev Mol Diagn. 2007;7(5):625-633.
  16. Wulfkuhle JD, Berg D, Wolff C, et al. Molecular analysis of HER2 signaling in human breast cancer by functional protein pathway activation mapping. Clin Cancer Res. 2012;18(23):6426-6435. doi: 10.1158/1078-0432.CCR-12-0452. 
  17. Baldelli E, Haura EB, Crinò L, et al. Impact of upfront cellular enrichment by laser capture microdissection on protein and phosphoprotein drug target signaling activation measurements in human lung cancer: implications for personalized medicine. Proteomics Clin Appl. 2015;9(9-10):928-937. doi: 10.1002/ prca.201400056. 
  18. Mueller C, deCarvalho AC, Mikkelsen T, et al. Glioblastoma cell enrichment is critical for analysis of phosphorylated drug targets and proteomic-genomic correlations. Cancer Res. 2014;74(3):818-828. doi: 10.1158/0008-5472.CAN-13-2172. 
  19. Pierobon M, Silvestri A, Spira A, et al. Pilot phase I/ II personalized therapy trial for metastatic colorectal cancer: evaluating the feasibility of protein pathway activation mapping for stratifying patients to therapy with imatinib and panitumumab. J Proteome Res. 2014;13(6):2846-2855. doi: 10.1021/pr401267m.
  20. Espina V, Mueller C, Edmiston K, Sciro M, Petricoin EF, Liotta LA. Tissue is alive: new technologies are needed to address the problems of protein biomarker pre-analytical variability. Proteomics Clin Appl. 2009;3(8):874-882.
  21. Montgomery RB, Makary E, Schiffman K, Goodell V, Disis ML. Endogenous anti-HER2 antibodies block HER2 phosphorylation and signaling through extracellular signal-regulated kinase. Cancer Res. 2005;65(2):650-656.
  22. Ali SM, Alpaugh RK, Downing SR, et al. Response of an ERBB2-mutated inflammatory breast carcinoma to human epidermal growth factor receptor 2-targeted therapy. J Clin Oncol. 2014;32(25):e88-e91. doi: 10.1200/JCO.2013.49.0599.
  23. Gandhi L, Bahleda R, Tolaney SM, et al. Phase I study of neratinib in combination with temsirolimus in patients with human epidermal growth factor receptor 2-dependent and other solid tumors. J Clin Oncol. 2014;32(2):68-75. doi: 10.1200/ JCO.2012.47.2787. 
  24. Jameson GS, Petricoin EF, Sachdev J, et al. A pilot study utilizing multi-omic molecular profiling to find potential targets and select individualized treatments for patients with previously treated metastatic breast cancer. Breast Cancer Res Treat. 2014;147(3):579-588. doi: 10.1007/s10549-014-3117-1.