Oncology DOI: 10.1159/000486678
Received: October 4, 2017
Accepted after revision: December 22, 2017 Published online: February 14, 2018
Analysis of 10 Adrenocortical Carcinoma Patients in the Cohort of the Precision Medicine Platform MONDTI
Markus Kielerª Leonhard Müllauerb Oskar Koperekb Daniela Bianconia Matthias Unselda Markus Raderera Gerald W. Pragerª
ªDivision of Oncology, Department of Medicine I, Comprehensive Cancer Center, Medical University Vienna, Vienna, Austria; bClinical Institute of Pathology, Medical University Vienna, Vienna, Austria
Keywords
Targeted therapy . Molecular profiling . Adrenocortical carcinoma . Molecular pathology · Genetic aberrations
Abstract
Objective: Adrenocortical carcinoma (ACC) is a rare disease with a dismal prognosis. We aimed to evaluate if a personal- ized medicine approach may be useful for matching patients with ACC to targeted therapies. Methods: This is an analysis of 10 molecularly profiled ACCs that were progressing under standard of care treatment. The profile consisted of a 50- gene next-generation sequencing panel, immunohisto- chemistry (IHC), and fluorescence in situ hybridization for several proteins or chromosomal aberrations. Results: In 6 (60%) tumor samples, no somatic mutation was detected, while in 3 (30%) tumors 1 mutation was detected and in 1 (10%) tumor 2 mutations were detected. These mutations were CTNNB1 (2 samples), TP53 (1 sample), RB1 (1 sample) and APC (1 sample). Expression of phospho-mTOR and of EGFR was commonly detected by IHC (87.5 and 62.5%). In 4 (50%) samples, IHC revealed a weak expression of progester- one receptor. Less frequent alterations were expression of PDGFR-a, c-KIT, and estrogen receptor, each in 1 case. Con-
clusions: Based on the molecular profile, no recommenda- tion for targeted therapy was made by the multi-disciplinary team. Currently, ACC might not be suitable for a precision medicine approach according to our tests.
@ 2018 S. Karger AG, Basel
Introduction
Adrenocortical carcinoma (ACC) is a rare disease with an estimated worldwide annual incidence of 2 affected individuals per million [1]. According to the current stag- ing classification of the European Network for the Study of Adrenal Tumors, ACC can be divided into 4 stages [2]. While stages I and II are potentially curable by complete surgical resection of the tumor, advanced stages III and IV are associated with a poor 5-year overall survival, ranging from 6 to 13% for patients with metastasized dis- ease [2, 3]. Given its high recurrence rate and only modest response to standard systemic chemotherapy with mito- tane plus etoposide, doxorubicin, and cisplatin, treat- ment of ACC still poses a great challenge to the clinician, and new therapies are urgently needed [4]. In the era of comprehensive cancer genomic profiling projects like
KARGER
Prof. Gerald W. Prager Division of Oncology, Department of Medicine I Comprehensive Cancer Center, Medical University Vienna Währinger Gürtel 18-20, AT-1090 Vienna (Austria) E-Mail gerald.prager@meduniwien.ac.at
The Cancer Genome Atlas (TCGA) [5], targeted drugs have now become routine in the clinical practice and offer long-lasting remissions in many malignancies, such as melanoma and non-small cell lung cancer [6, 7]. Recent genomic analyses of ACCs have revealed distinct altera- tions in oncogenic pathways, such as the ß-catenin path- way, and have led to a new understanding of the disease [8, 9]. Despite this progress, promising personalized treatment attempts, for example blocking the insulin-like growth factor-1 receptor, have not shown any improve- ment in survival of patients with ACC [10]. We, therefore, performed a subanalysis of 10 patients with metastatic ACC that have been included in our precision medicine platform MONDTI. Tumors that were refractory to all standard treatment options were molecularly profiled to match them to targeted therapies. The profile consisted of a 50-gene next-generation sequencing (NGS) panel, immunohistochemistry (IHC), and fluorescence in situ hybridization (FISH) for several proteins or chromosom- al aberrations.
Materials and Methods
Patients and Design of the Precision Medicine Platform
Patients with advanced or metastasized solid tumors or lym- phomas with no standard treatment options were eligible for in- clusion in MONDTI, provided archival tissue samples were avail- able or a fresh biopsy was feasible. Patients needed to have an East- ern Cooperative Oncology Group (ECOG) performance status of 0 or 1. The precision medicine project MONDTI has been ap- proved by the local ethical committee, and all patients had to pro- vide informed consent before inclusion in MONDTI. Further- more, this subanalysis has also been approved by the local ethical committee.
Tissue Samples
Formalin-fixed, paraffin-embedded tissue from patients with advanced or metastasized solid tumors or lymphomas that were refractory to all available standard treatment lines were sent to or retrieved from the archive of the Department of Pathology, Medi- cal University Vienna, Vienna, Austria.
Cancer Gene Panel Sequencing
DNA was extracted from paraffin-embedded tissue blocks with a QIAamp Tissue Kit™ (Qiagen, Hilden, Germany). 10 ng DNA per sample was utilized for sequencing. The DNA library was gen- erated by multiplex polymerase chain reaction with the Ion Am- pliSeq Cancer Hotspot Panel v2™M (Life Technologies, Carlsbad, CA, USA).
Immunohistochemistry
IHC was performed using 2-um-thin tissue sections. The fol- lowing antibodies were employed: ALK (clone 1A4; Zytomed, Ber- lin, Germany), CD30 (clone BerH2; Dako, Vienna, Austria), CD20
(clone L26; Dako), epidermal growth factor receptor (EGFR) (clone 3C6; Ventana), estrogen receptor (ER) (clone SP1; Ven- tana), HER2 (clone 4B5; Ventana), HER3 (clone SP71; Abcam), KIT (clone 9.7; Ventana), MET (clone SP44; Ventana), phosphor- ylated-mechanistic target of rapamycin (p-mTOR) (clone 49F9; Cell Signaling Technology, Danvers, MS, USA), PDGFRA (rabbit polyclonal; Thermo Fisher Scientific), PDGFRB (clone 28E1; Cell Signaling Technology), PD-L1 (clone E1L3N; Cell Signaling Tech- nology), progesterone receptor (PR) (clone 1E2; Ventana), PTEN (clone Y184; Abcam), and ROS1 (clone D4D6; Cell Signaling Technology). An immunohistochemical score was determined by multiplying the percentage of positive cells with their respective staining intensity (0 = negative, 1 = weak, 2 = moderate, 3 = strong), as follows: (maximum 300) = (% negative × 0) + (% weak × 1) + (% moderate × 2) + (% strong × 3).
Fluorescence in situ Hybridization
FISH was performed with 4-um-thick formalin-fixed, paraffin- embedded tissue sections. The following FISH probes were em- ployed: ALK (2p23.1; Abbott, Abbott Park, IL, USA), RET (10q11; Kreatech, Berlin, Germany), PTEN (10q23.31)/Centromer 10, and ROS1 (ZytoVision, Bremerhaven, Germany). Two hundred cell nuclei per tumor were evaluated.
Multidisciplinary Boards (MONDTI Boards)
After summarization by an experienced molecular pathologist, the molecular profile of each tumor sample was discussed within the multidisciplinary boards that were held every other week. Members of the board were molecular pathologists, radiologists, clinical oncologists, biostatisticians, and basic scientists. Targeted therapy was chosen based on the individual tumor profile and con- sisted of tyrosine kinase inhibitors, checkpoint inhibitors (anti- PD-L1 monoclonal antibodies), and growth factor receptor anti- bodies with or without endocrine therapy. Treatment decisions by the multidisciplinary team were prioritized according to the level of evidence from high to low according to phase III to phase I tri- als. If more than 1 targetable alteration was detected, combina- tional therapy covering as many molecular targets as possible was chosen, considering the toxicity profile of each drug and their po- tential mutual interactions. Given that patients received all stan- dard treatment options for their specific malignancy before enter- ing the molecular profiling program, almost all matched targeted agents were recommended as off-label use. If the molecular profile met the inclusion criteria of a clinical trial for molecular targets that was ongoing in our cancer center, patients were preferentially asked if they wanted to participate in this trial.
Results
The data of 6 female and 4 male patients with treat- ment-refractory metastatic ACC and a median age at di- agnosis of 50 years (interquartile range between 44 and 65 years) were available for this subgroup analysis of 297 tu- mors that have been molecularly profiled since Novem- ber 2013. Of the 10 tissue samples, 7 were from metastat- ic and 3 from primary lesions. Eight patients presented
Kieler/Müllauer/Koperek/Bianconi/ Unseld/Raderer/Prager
| Patient No. | Age at diagnosis, years | Stage at diagnosis | Tissue tested | Next-generation sequencing gene panel | Immunohistochemistryª | Fluorescence in situ hybridization |
|---|---|---|---|---|---|---|
| 1 | 36 | III | metastatic | no mutation detected | not done (insufficient tissue material) | not done |
| 2 | 69 | II | metastatic | CTNNB1 (missense), TP53 (missense) | p-mTOR +2 (170) | no alteration |
| 3 | 43 | II | metastatic | no mutation detected | EGFR +1 (140), PR +1 (IRS 2), p-mTOR +1 (90) | no alteration |
| 4 | 66 | II | metastatic | RB1 (missense) | c-KIT +1, EGFR +1 (130), PR +1 (IRS 2), p-mTOR +1 (70) | PTEN heterozygous deletion in 64% of cells |
| 5 | 25 | III | primary | CTNNB1 (nonframeshift deletion) | EGFR +1 (60), PR +2 (IRS 3), PDGFR-a +1, p-mTOR +1 (90) | no alteration |
| 6 | 53 | II | metastatic | no mutation detected | p-mTOR +1 (80) | no alteration |
| 7 | 47 | IV | primary | no mutation detected | EGFR +3 (220), PR +1, ER +1, p-mTOR +1 (5) | no alteration |
| 8 | 55 | III | metastatic | no mutation detected | no alteration detected | no alteration |
| 9 | 46 | IV | metastatic | no mutation detected | not done (insufficient tissue material) | not done |
| 10 | 65 | II | primary | APC (nonsense) | EGFR +1 (30), p-mTOR +1 (60) | no alteration |
CTNNB1, catenin-ß1; TP53, tumor protein P53; p-mTOR, phosphorylated-mechanistic target of rapamycin; RB1, RB transcrip- tional corepressor 1; APC, adenomatous polyposis coli; EGFR, epidermal growth factor receptor; PR, progesterone receptor; c-KIT, c- KIT proto-oncogene receptor tyrosine kinase; PDGFR, platelet derived growth factor receptor; ER, estrogen receptor; IRS, immunore- active score; PTEN, phosphatase and tensin homolog. ª Values in parentheses indicate the immunohistochemical score that was calcu- lated as mentioned in the Materials and Methods section.
with recurrent disease and had undergone surgery of their primary tumors in the past, while 2 patients had stage IV disease at the time of diagnosis. All patients re- ceived standard chemotherapy with mitotane plus epiru- bicin, cisplatin, and etoposide.
All tumor samples were successfully sequenced with the NGS panel. In 6 (60%) cases, no somatic mutation was de- tected, while in 3 (30%) cases 1 mutation was detected and in 1 (10%) case 2 mutations were identified. There were 2 samples with a mutation in the catenin-ß1 (CTNNB1) gene. One of these 2 samples had a concomitant mutation in the tumor protein P53 (TP53) gene. Two samples were either mutated in the RB transcriptional corepressor 1 (RB1) or adenomatous polyposis coli (APC) gene.
IHC and FISH were not performed in 2 cases due to insufficient tumor material. p-mTOR was expressed in 7 (87.5%) of the 8 remaining tumor samples. The expression of p-mTOR was moderate (IHC score between 100 and
200) in 1 sample and weak in 6 samples (IHC score <100). EGFR expression was expressed in 5 (62.5%) tumor sam- ples. The expression was weak in 4 samples (IHC score <200) and strong in 1 sample (IHC score >200). In 4 (50%) samples, IHC revealed a positive expression for PR, which was weak in 3 cases and moderate in 1 case. Less frequent alterations were expression of platelet-derived growth fac- tor receptor alpha (PDGFR-a) (12.5%), c-KIT proto-on- cogene receptor tyrosine kinase (c-KIT) (12.5%), and ER (12.5%), each in 1 sample. The expression of these pro- teins was found to be weak in all samples. FISH identified a heterozygous deletion of PTEN in 64% of the tumor cells in 1 (12.5%) tumor sample. We refer to Table 1 for a sum- mary of the results.
All molecular profiles were discussed in the multidis- ciplinary board of the precision medicine platform MONDTI. Based on these results, no targeted therapy was recommended.
Discussion
In this report of a precision medicine approach for ad- vanced ACCs, including NGS, IHC, and FISH, we sought to verify our results in the light of previously published precision medicine studies for ACC [11, 12]. We identi- fied 4 samples with at least 1 mutation in our 50-gene panel. There were 2 samples with a mutation in the CTNNB1 gene. One of these 2 samples had a concomitant mutation in the TP53 gene. Two samples were either mu- tated in the RB1 or APC gene. In large genomic charac- terization projects, whole-exome sequencing confirmed CTNNB1 and TP53 as 2 of the most recurrently mutated genes among sporadic ACC, and the prevalence for both genes ranges from 16 to 21% in larger ACC series [8, 9]. RB1 and APC mutations are less common but also well described in the genomic landscape of ACC with a rate of 7 and 2%, respectively [9]. CTNNB1 and APC are both related to the ß-catenin pathway, while TP53 and RB1 are part of the p53/RB tumor suppressor pathway. None of these genes are currently targetable.
p-mTOR, as assessed by IHC, was expressed in 87.5% of tumor samples that were available for advanced profil- ing. Presently, there is no evidence that clearly supports a key role of this pathway in the pathogenesis of ACCs. This is supported by the finding that adrenocortical adenomas have higher mRNA expression of p-mTOR than ACCs [13]. Currently, there is no clear rationale behind the use of mTOR inhibitors as a targeted therapy in ACCs.
EGFR was expressed in 5 (62.5%) tumor samples. Its expression is a common finding in ACC samples at com- parable rates as reported in our cohort [14]. Single EGFR inhibition did not inhibit tumor growth in a preclinical tumor model and was not considered as a therapeutic op- tion in MONDTI for this disease [15].
In 4 (50%) samples, IHC revealed a positive expression for PR, which was weak in 3 cases and moderate in 1 case. One sample was weakly positive for ER. PR and ER over- expression is another typical finding in ACC [16]. The results of a study in a preclinical model suggest that si- lencing of PR and ER in ACC may have anti-hormone- secretive and antitumor activity, perhaps mediated through inhibition of the Wnt/B-catenin pathway [17]. We did not consider PR or ER inhibition monotherapy as a therapeutic option because of the low evidence for clin- ical efficacy. Further research regarding the role of the PR and the ER in the therapy of ACC is needed.
Less frequent alterations were expression of PDGFR-a (12.5%), c-KIT (12.5%), and only weak expression of PTEN (12.5%). The expression of these proteins was
found to be weak in all samples. PDGFR-a and c-KIT overexpression have not been reported in ACC to date.
There is a need for better therapies for nonresectable or metastatic ACC, as this malignant disease has a poor response to cytotoxic treatment, and the median overall survival of patients treated with first-line treatment with mitotane plus etoposide, doxorubicin, and cisplatin is only 14.8 months [4]. Here, we present a subanalysis of 10 ACCs that have been included in our precision med- icine platform MONDTI to molecularly characterize ad- vanced or metastatic solid tumors in order to identify actionable genetic alterations. In MONDTI, of the 297 included tumor samples, 288 had 1 or more genetic al- terations detected and 160 (55.6%) were matched to tar- geted therapies. The aim of this subanalysis was to eval- uate if our personalized medicine approach holds its promise in this rare tumor entity. In contrast to the matching rate of 55.6% in the original MONDTI collec- tive, no targeted therapy could be recommended. Based on the results of the data from our cohort, we cannot further recommend to include patients with ACC main- ly because of 2 reasons. First, somatic mutations detect- ed by the NGS panel (CTNNB1, TP53, RB1, and APC) are currently not targetable. Second, no targeted therapy was recommended in the whole ACC subset of MOND- TI because the expression was found to be only weak or because there is a lack of evidence for an effective tar- geted therapy approach for the detected molecular al- terations.
We are aware of 2 other precision medicine studies for ACC. In 1 study, the molecular characterization of 29 ACC samples was performed by Foundation Medicine using NGS to detect alterations in 236 cancer-related genes and 47 introns of 19 genes commonly rearranged in cancer [11]. According to the results published in that study, 17 (59%) of 29 ACC samples had a targetable genomic alteration revealing mutations in the NF1, CDKN2A/B, ATM, EGFR, PTCH1, and PIK3CA gene or amplification of the CDK4 and PDGFR-ß gene. In the other study, hot spot gene sequencing of 46 different on- cogenes in 40 patients as well as comparative genomic hybridization of 130 different genes in 28 patients was performed [12]. The authors concluded that no simple molecular target emerged; however, amplification of the CDK4 oncogene (5 of 28; 17.9%) and deletion of the CDKN2A (4 of 28; 14.3%) as well as of the CDKN2B (3 of 28; 10.7%) tumor suppressor genes were detected, and drugs targeting these alterations may be the most relevant in ACC. Data about the clinical outcome were not report- ed in neither of the 2 studies, and whether targeting of
these alterations might provide an effective therapy still needs to be validated.
Considering our findings and in light of the previous- ly published precision medicine studies, we conclude that precision medicine has not yet demonstrated a meaning- ful clinical benefit in ACC. We expect that precision med- icine will witness the integration of transcriptomic data in addition to whole-genome or whole-exome sequenc- ing in the near future. RNA sequencing adds another di- mension to genomic profiling by assessing expressed variants in relation to nonexpressed genes, gene fusions, and activated oncogenic pathways. Initial attempts of an integrated analysis of whole-genome, exome, and tran- scriptome data have already been evaluated in the clinical setting and have even led to the discovery of novel thera- peutic targets [18-21]. This will allow a more precise identification of the underlying driving events and path-
ways in the individual case and thus might also have an impact on the personalized treatment of rare cancers such as ACC.
Disclosure Statement
G.W.P. received speaker’s fees from Bayer, Roche, Merck-Se- rono, Amgen, Servier, Celgene, Shire, MSD, Lilly, and Sanofi- Aventis. M.K. received travel support from Merck, Bayer, Bristol- Myers Squibb, and Roche. All other authors have nothing to de- clare.
Funding Sources
This work was supported by a grant from Initiative Krebs- forschung, grant No. UE 71104027.
References
1 Germano A, Rapa I, Duregon E, et al: Tissue expression and pharmacological in vitro anal- yses of mTOR and SSTR pathways in adreno- cortical carcinoma. Endocr Pathol 2017;28: 95-102.
2 Fassnacht M, Johanssen S, Quinkler M, et al: Limited prognostic value of the 2004 Interna- tional Union against Cancer staging classifi- cation for adrenocortical carcinoma. Cancer 2009;115:243-250.
3 Else T, Williams AR, Sabolch A, et al: Adju- vant therapies and patient and tumor charac- teristics associated with survival of adult pa- tients with adrenocortical carcinoma. J Clin Endocrinol Metab 2014;99:455-461.
4 Baudin E, Haak H, Berruti A, et al: Combina- tion chemotherapy in advanced adrenocorti- cal carcinoma. N Engl J Med 2012;366:2189- 2197.
5 Cancer Genome Atlas Research Network; Weinstein JN, Collisson EA, Mills GB, Shaw KR, Ozenberger BA, et al: The Cancer Ge- nome Atlas Pan-Cancer analysis project. Nat Genet 2013;45:1113-1120.
6 Larkin J, Ascierto PA, Dréno B, et al: Com- bined vemurafenib and cobimetinib in BRAF- mutated melanoma. N Engl J Med 2014;371: 1867-1876.
7 Maemondo M, Inoue A, Kobayashi K, et al: Gefitinib or chemotherapy for non-small-cell lung cancer with mutated EGFR. N Engl J Med 2010;362:2380-2388.
8 Zheng S, Cherniack AD, Dewal N, et al: Com- prehensive pan-genomic characterization of adrenocortical carcinoma. Cancer Cell 2016; 29:723-736.
9 Assié G, Letouzé E, Fassnacht M, et al: Inte- grated genomic characterization of adrenocor- tical carcinoma. Nat Genet 2014;46:607-612.
10 Haluska P, Worden F, Olmos D, et al: Safety, tolerability, and pharmacokinetics of the anti- IGF-1R monoclonal antibody figitumumab in patients with refractory adrenocortical car- cinoma. Cancer Chemother Pharmacol 2010; 65:765-773.
11 Ross JS, Wang K, Rand JV, et al: Next-gener- ation sequencing of adrenocortical carcinoma reveals new routes to targeted therapies. J Clin Pathol 2014;67:968-973.
12 De Martino MC, Al Ghuzlan A, Aubert S, et al: Molecular screening for a personalized treatment approach in advanced adrenocorti- cal cancer. J Clin Endocrinol Metab 2013;98: 4080-4088.
13 De Martino MC, Feelders RA, De Herder WW, et al: Characterization of the mTOR pathway in human normal adrenal and adre- nocortical tumors. Endocr Relat Cancer 2014; 21:601-613.
14 Adam P, Hahner S, Hartmann M, et al: Epi- dermal growth factor receptor in adrenocorti- cal tumors: analysis of gene sequence, protein expression and correlation with clinical out- come. Mod Pathol 2010;23:596-604.
15 Xu L, Qi Y, Xu Y, et al: Co-inhibition of EGFR and IGF1R synergistically impacts therapeu- tically on adrenocortical carcinoma. Oncotar- get 2016;7:36235-36246.
16 De Cremoux P, Rosenberg D, Goussard J, et al: Expression of progesterone and estradiol receptors in normal adrenal cortex, adreno- cortical tumors, and primary pigmented nod- ular adrenocortical disease. Endocr Relat Cancer 2008;15:465-474.
17 Fiorentini C, Fragni M, Perego P, et al: Anti- secretive and antitumor activity of abi- raterone acetate in human adrenocortical cancer: a preclinical study. J Clin Endocrinol Metab 2016;101:4594-4602.
18 Roychowdhury S, Iyer MK, Robinson DR, et al: Personalized oncology through integrative high-throughput sequencing: a pilot study. Sci Transl Med 2011;3:111ra121.
19 Mody RJ, Wu Y-M, Lonigro RJ, et al: Integra- tive clinical sequencing in the management of refractory or relapsed cancer in youth. JAMA 2015;314:913.
20 Craig DW, O’Shaughnessy JA, Kiefer JA, et al: Genome and transcriptome sequencing in prospective metastatic triple-negative breast cancer uncovers therapeutic vulnerabilities. Mol Cancer Ther 2013;12:104-116.
21 Borad MJ, Champion MD, Egan JB, et al: In- tegrated genomic characterization reveals novel, therapeutically relevant drug targets in FGFR and EGFR pathways in sporadic intra- hepatic cholangiocarcinoma. PLOS Genet 2014;10:e1004135.