Genomic and expression profiling of adrenocortical carcinoma: application to diagnosis, prognosis and treatment
Kimberly J Bussey & Michael J Demeuret
Author for correspondence: Senior Investigator, Clinical Translational Research Division, Translational Genomics Research Institute, Director, Endocrine Tumors Center, Scottsdale Healthcare, 10460 N. 92nd St, Suite 200, Scottsdale, AZ 85258, USA · Tel .: +1 480 323 1250 · Fax: +1 480 323 1259 = mdemeure@tgen.org
Adrenocortical carcinoma (ACC) is an aggressive endocrine tumor with a poor 5-year survival rate of 10-20%. Current therapy is often ineffective and may be associated with intolerable side effects. Although ACC is extremely rare, recent advances in genomic and expression profiling, coupled with knowledge gained from the study of the inherited syndromes that increase ACC risk, are beginning to bring together a picture of a tumor type dependent on p53, the G2/M cell cycle transition and IGF2 stimulation. Nevertheless, ACC remains a heterogeneous disease. Only recently have sufficient tumors been characterized and results published to permit an exploration of this diversity. Advances in treatment will depend on exploiting those pathways already implicated in ACC, along with those yet to be identified, and testing those treatments in better models of the disease than the three cell lines that currently exist and are widely available to the community.
Adrenocortical carcinoma (ACC) is an aggres- sive malignancy of the adrenal cortex. ACC has an incidence of approximately 1-2 per mil- lion and accounts for 0.02-0.2% of all cancer deaths [1-4]. Women are affected approximately 2.5-times more often than men. Over 50% of ACCs are functional, in that there is clinical evidence of excess hormone production, such as cortisol, aldosterone or a sex steroid. Factors making treatment decisions complicated are that histological diagnosis of ACC may be difficult to make, and that there is a paucity of clinical studies to guide therapy owing to the rarity of the disease. Metastatic disease or invasion into a contiguous structure is the only absolute indica- tor of malignant disease in masses of the adre- nal cortex. Local recurrence is not sufficient to establish the diagnosis of cancer because even benign tumors may recur in the surgical bed as a result of capsular disruption during resec- tion. Supportive criteria of malignancy include a diameter greater than 5 cm and a weight in excess of 50 g. The average size of an adreno- cortical cancer when diagnosed is 10-12 cm in diameter [5]. The presence of an irregular border, adjacent lymphadenopathy and heterogeneous pattern seen by computerized tomography also suggest a malignant adrenal lesion. The Weiss score is used to help classify adrenocortical tumors as benign or malignant. Weiss scores of 0-2 are considered to be benign or at least will
exhibit an indolent course, while scores of 4-9 are malignant and biologically aggressive in their clinical behavior [6]. A Weiss score of 3 is consid- ered borderline, and although most patients are treated as presumptively having ACC, a propor- tion of them are treated successfully by complete surgical resection alone, suggesting that a Weiss score of 3 includes a subset of benign tumors.
Using genomic & transcriptomic information in ACC diagnosis
Given the difficulty at times of distinguishing benign from malignant adrenocortical tumors on the basis of histopathologic examination alone, there is great interest in the identification of genomic markers that may aid in the diag- nosis of ACC. Accordingly, the genomic ana- lysis of these tumors supports a difference in the acquired genetic changes characterizing adeno- mas versus ACC. Traditional cytogenetics dem- onstrate that adenomas are almost always near diploid with loss of a sex chromosome [101]. In the two cases that have been reported with structural abnormalities, both had involvement of chromo- some 7 with breakpoint of 7q22 and 7q11. Only four cytogenetic studies of ACC from adults and one study of a pediatric ACC have been reported in the literature [101]. Among the adult tumors, two cases had structurally and numerically com- plex karyotypes, while the other two cases were diploid with a single structural change (TABLE 1).
Review Future Oncology
Keywords
adrenocortical carcinoma cell cycle . expression profiling = genomic profiling = IGF2 . mitotane =p53
future medicine
part of
fsg
| Table 1. Karyotypes of adrenocortical tumors reported in the literature. | |||
|---|---|---|---|
| Patient gender | Patient age (years) | Pathology | Karyotype Ref. |
| Female | Unknown | Adenoma | 46,XX,t(7;17)(q22;p13) [65] |
| Male | 70 | Adenoma | 45,X,-Y,/44,idem,-19 [66] |
| 46 | 45,X,-Y | ||
| 59 | 45,X,-Y | ||
| 48 | 45,X,-Y | ||
| 58 | 45,X,-Y,der(6)t(6;7)(p25;q11) | ||
| Female | 68 | Adenocarcinoma | 46,XX,add(11)(p15) [67] |
| Male | 68 | Carcinoma | 54,XY,+X,+Y,+3,der(4)t(4;22)(p16;q11) [68] x2,+5,+8, -10,+der(12)t(3;12) (p14;p13)x2, add(14)(q?)x2,+der(15)t(15;20)(p11;q11)x2, der(18)t(3;18) (p12;p11), der(18)t(5;18) (p13;p11), +19,der(20)t(9;20)(p11;q11)x2,-22 |
| Female | 25 | Adenocarcinoma | 46,XX,t(4;11)(q35;p13) [69] |
| Female | 82 | Carcinoma | 38,XX,del(1)(q21q42),add(2)(p13), del(3) [70] (p13p24),-5,-6,der(11)t(1;11) (p13;q23), -12,-13,-15,-15,-17,-18,?i(19)(p10),-22,inc |
| Female | 4 | Adenocarcinoma | 56-57,XX,+2,+4,+5,+7,+8,inv(9)(p11q12)c, [71] +10,+add(13)(p11),+14,+15,+19,+20, +20,+mar |
One case was hyperdiploid with net gains of 3p14-pter, 5p13-pter, 9p11-pter, 12p13-qter, 15q, and whole-chromosome gains of X, Y and 19. The same tumor demonstrated a net loss of material from 4p16-pter, 18p11-pter, and whole- chromosome loss of 10 and 22. The other case with a complex karyotype was hypodiploid, with net gains of material from 1p13-pter and 19p, and net loss of material from 1q21-q24, 3p13- p24, 11q23-qter, 19q, and whole-chromosome loss of 5, 6, 12, 13, 15x2, 17, 18 and 22. No common recurrent rearrangement has been seen. The one pediatric ACC karyotyped had a less structurally complex karyotype with the gain of an add(13)(p11) and an undefined marker chromosome as the structural changes. Whole- chromosome gains were observed for chromo- somes 2, 4, 5, 7, 8, 10, 14, 15, 19 and 20 [101].
Conventional comparative genomic hybridiza- tion (CGH) involves hybridizing differentially labeled DNA from tumor and a normal control to metaphase chromosomes. The level of resolu- tion obtainable in such a study depends on the length of the metaphase chromosomes and is accepted as having a limit of 10 Mb. Five studies report the results of applying conventional CGH to ACC (TABLE 2) [3,7-10]. No abnormality has been reported at present in all ACC. Most are present in 50-60% or less of the tumors exam- ined in each series. There is also a great deal of heterogeneity within tumors and between stud- ies. However, taken as a whole, ACC demon- strate gains of 4p16, 5p15, 5q12-13, 5q32-qter, 9q34, 12q13, 12q24 and 19p. Regions of loss
include 1p21-31, 2cen-q21, 9p and 11q24-qter. Similar data emerge from array CGH studies, where instead of metaphase chromosomes as the hybridization target, the samples and controls are hybridized to DNA microarrays. This permits a resolution that is limited only by the genomic distance between two probes on the array. Our group reported that among 25 tumors, there were overall gains within chromosomes 5, 6q, 7, 8q, 12, 16q and 20. We found losses within chromosomes 1, 2q, 3, 6p, 7p, 8p, 9, 10, 11, 13q, 14q, 15q, 16, 17, 19q and 22q [11].
Among pediatric ACC, two CGH studies have revealed a different pattern of abnormali- ties from that observed in adult ACC [12,13]. In contrast to adult adrenal tumors, there were no differences observed in the patterns of gain and loss in pediatric ACC and adenomas. Gain of 9q34, particularly as a highly amplified region, has been an almost universal finding in pediatric ACC. This region harbors the ABL oncogene as well as the transcription factor SF-1, which is known to play a role in normal adrenal cortex development. Gain of chromosomes 19, 12q24, and 11q13 were commonly observed in both studies. Common regions of loss were seen in both studies on 4q and 2q.
CGH studies have revealed that adenomas in adults generally have few, if any, changes, and the number of aberrations accumulates as a func- tion of tumor size. No specific change has been identified, however, that distinguishes adeno- mas from ACC. In this respect, recent advances in the characterization of gene expression may
help with diagnosis. Every gene-expression study that has looked at ACC versus adenomas, both in adults and children, has shown that gene expression is capable of distinguishing between the two entities [14-21]. Two studies have looked for signatures that are as good as or better than the Weiss score for distinguishing benign ade- nomas from ACC. This work was spurred by the findings of de Fraipont and colleagues who demonstrated that a 22-gene signature encom- passing overexpression of IGF2 and related genes, and under-expression of steroidogenic genes, was nearly as good as the Weiss score in distinguishing benign from malignant adreno- cortical tumors [14]. A study by de Reyniès et al. demonstrated that a two-gene signature, PINK1 and DLG7, was capable of accurately classifying tumors as benign or malignant and was compa- rable to Weiss score, being highly predictive of disease-free survival [19]. Soon and colleagues, guided by data from expression profiling, dem- onstrated that immunohistochemisty for IGF2 and Ki-67 was able to accurately classify tumors with a Weiss score of 3 as being either benign or malignant [21].
Emerging prognostic indicators
Complete surgical excision is the only treatment that offers a realistic potential for cure and is the best prognostic factor predicting long-term sur- vival. Unfortunately, 40-70% of patients have metastases at the time of diagnosis precluding surgical cure [22,23]. This relatively late stage at diagnosis results in a poor overall 5-year survival rate of 20-35%. Median survival of patients with completely resected tumors is 46 months, com- pared with only 8.5 months if the tumors were
incompletely resected (p <0.005) [24]. In patients whose tumors are not resectable, however, the biologic course is variable, with some patients still having relatively long survival and others succumbing to the disease quickly. Genomics may offer clues to predict biologic behavior.
Only our group’s analysis has examined the association between genomic aberrations detected by array-based CGH and survival. Amplifications of 6q, 7q, 12q and 19p, and losses of 3, 8, 10p, 16q, 17q and 19q, were significantly associated with poor survival [11]. The focus of gene-expression profiling in ACC has been on detecting differences between benign and malig- nant adrenal tumors, as well as establishing bet- ter prognostic indicators. Gene-expression pro- filing studies have found IGF2, FGFR1, FGFR2, FGFR4 and TOP2A to be increased in expres- sion, while CDKN1C, KCNQ1, ADH1, IGFBP6, IGF1R and ABCB1 show decreased expression in ACC relative to normal adrenal tissue. In one of the most recent expression studies, Giordano and colleagues demonstrated that genes differ- entially expressed in ACC relative to normal adrenal tissue were often involved in cell cycle progression, especially mitosis [20]. They also identified 12q and 5q as chromosomal regions with evidence of enrichment for overexpressed genes in ACC, while 11q, 1p and 17p had evi- dence of enrichment for underexpressed genes. Both de Reynies et al. and Giordano et al. saw a clustering of ACC into two groups with a sig- nificant difference in survival [19,20]. Giordano et al. observed an association between cluster and mitotic index, indicating that the clustering based on gene expression approximated group- ing by tumor grade. Given the enrichment for
| Table 2. Summary of previous genomic analyses in adrenocortical carcinoma. | ||||
|---|---|---|---|---|
| Number of ACCs | Genetic analysis | Regions gained | Regions lost | Ref. |
| 12 | Conventional CGH | 5q12-q13, 5q22-qter, 9q32-qter, 12q13-q14, 12q24, 20q, Xq13-q21 | 1p21-p31, 1q23-q41, 2p21-pter, 2q, 3p, 3q, 6q, 9p, 11q14-qter, 18q | [10] |
| 14 | Conventional CGH | 1p34.3-pter, 1q22-q25, 3p24-pter, 3q29, 5, 7, 7p11.2-p14, 8, 9q, 9q34, 11q, 11q12-q13, 12q, 12q13, 12q24.3, 13q34, 14q, 14q11.2-q12, 14q32, 16, 16p, 17q, 17q24-q25, 19, 19p13.3, 19q13.4, 20, 22q, 22q11.2-q12 | 9p | [7] |
| 13 | Conventional CGH | 4, 5, 12, 12q14-q21, 19 | 1p, 1p34-pter, 2q, 2q34-qter, 11q, 11q24-qter, 17p, 17p13-pter, 22 | [8] |
| 8 | Conventional CGH | 4q, 4q31, 5, 12, 12cent-q24, 15q, 15q21-qter, 16q, 19p | 2, 2p23-cen-q21, 3p21-cent, 6q, 8p, 9p, 11p, 11q, 11q22-qter, 17p, 17q, 18q, 22q | [3] |
| 12 | Conventional CGH and oncogene- specific microarray | 1q, 4p15-pter, 5p, 5p15, 5q, 5q13, 5q32-qter, 7p, 7q, 8q, 8q24, 9p, 9q, 12q13-q15, 13q, 16q, 17p, 17q, 20p, 20q | [19] | |
| ACC: Adrenocortical carcinoma; CGH: Comparative genomic hybridization. | ||||
fsg
cell cycle genes, and mitosis in particular, and the fact that mitotic index is one component of tumor grade, one might expect that the approxi- mation of grade by gene-expression data was due to the mitotic index component. However, when a Cox proportional hazards model incorporat- ing tumor grade, log of the mitotic index and the gene-expression data from the first principle component was examined, mitotic index was not found to be significant, while grade and the principle component were. This suggests that the gene-expression data contains informa- tion beyond what is captured by grade. When Giordano and colleagues looked at the enrich- ment of genes overexpressed between the two clusters, they also observed that in addition to enrichment for genes involved with cell cycle and chromosomal instability in the poor prog- nosis group, there was a differential association with chromosomal location. In the cluster with the better prognosis, genes on 16p and 5q were enriched. In the cluster with poor prognosis, genes on 1q, 22q, 6q, 10p and 6p were enriched [20]. Interestingly, only the 6q enrichment cor- responds with a region associated with poor prognosis by array CGH. De Reyniès and col- leagues identified a two-gene signature, PINK1 and BUB1B, as predictive of overall survival. A sample was predicted to have a poor prognosis based on quantitative polymerase chain reaction if the A cross-over threshold of BUB1B minus the A cross-over threshold of PINK1 was less than 6.32 [19].
Moving toward pathway-driven treatments in ACC
No proven chemotherapeutic regimen exists for the treatment of ACC, and the toxicity of the only approved compound, mitotane (1,1-dichloro-2- [o-chlorophenyl]-2-[p-chlorophenyl] ethane, or o,p’-DDD), is often severe and dose-limiting. An isomer of the pesticide dichlorodiphenyltri- chloroethane (DDT), mitotane is an agent that is directly toxic to the cells of the adrenal cor- tex. The biochemical mechanisms of action for mitotane are not well characterized, although it is thought to work after hydroxylation by a mitochondrial cytochrome P450 enzyme and subsequent conversion to an acyl chloride that is cytotoxic [25]. Cytotoxicity ultimately results in mitochondrial disruption and necrosis [26]. The primary metabolites of mitotane are o,p’-DDA and o,p’-DDE. Investigations into the localiza- tion of 3-methylsulfonly-DDE, a metabolite of DDT, and mitotane demonstrated that both compounds accumulate in the zona fasciculata
and the zona reticularis layers of the adre- nal cortex, but not the zona glomerulosa [27]. The activity of mitotane is dependent on the 1,1-dichloro structure; conversion of this moiety to a methyl group significantly reduces activ- ity [28]. Investigations into the protein profile of H295R cells and steroid production after mito- tane exposure suggest that the cytochrome P450 enzyme involved in the initial activation step acts upstream of the steroidogenic cascade [29].
The first report of mitotane use in patients with ACC was presented by Bergenstal and col- leagues in 1959 [30]. The therapeutic value of mitotane is dependent on achieving sufficiently high blood levels of the drug, but the therapeu- tic window is narrow [31]. The side effects are often severe enough to require dose reduction or discontinuation of treatment.
After ACC resection, the role of adjuvant treatment with mitotane is controversial and the rarity of this tumor has made extensive clinical trials to determine optimum therapy difficult. Adjuvant mitotane has been advocated by some, citing the observation that as many as 75-85% of patients who have undergone surgical resection ultimately have relapses of their disease [32,33]. Most studies to date have lacked adequate sta- tistical power to settle the question of whether adjuvant mitotane has efficacy. The most recent and perhaps most often quoted study favoring the use of adjuvant mitotane may be criticized because it is a retrospective, nonrandomized study and lacked quality control of the surgical resection procedures. In this study by Terzolo and colleagues, the use of adjuvant mitotane was associated with longer recurrence-free sur- vival [34]. There was no difference in the benefit seen amongst patients who received mitotane doses of either 3-5 g or 1-3 g daily.
In a study of patients with advanced adreno- cortical cancer led by the Eastern Cooperative Oncology Group, 22% of patients responded to mitotane treatment and experienced prolonged median survival of 50 months compared with 14 months for nonresponders [35]. Van Slooten and associates showed an increase in survival or tumor regression in 57% of patients who had mitotane serum levels greater than 14 µg/ml, with complete remission in one patient; however, no patient with a mitotane serum level less than 10 µg/ml had a significant response to chemo- therapy [36]. Other studies have failed to show response to mitotane [37,38]. Chemotherapeutic strategies using mitotane in combination with other standard agents have been reported with mixed results. The most promising regimen thus
far has been a combination of doxorubicin, eto- poside and cisplatin in conjunction with mito- tane. The authors reported that in 72 patients with metastatic disease, five had a complete response and 30 had a partial response, giving an overall response rate of 48.6%. The median time to progression and overall survival of the entire cohort were 9.1 and 28.5 months, respec- tively, with responders showing approximately double that [39]. As current therapy is clearly suboptimal, it remains vitally important to con- tinue efforts to identify drugs that, alone or in combination with mitotane, can improve patient survival with fewer side effects.
So what insights do the genomic and expres- sion studies offer us regarding potential new avenues for treatment? While the genomic and expression studies to date show some overlap (i.e., the association of increasing chromosomal instability with poor prognosis), there is no one region that is obvious as the necessary and suf- ficient acquired change in ACC comparable to that of the Philadelphia chromosome that leads to the BCR-ABL fusion gene in chronic myeloid leukemia. We should therefore consider the pathways that have already been associated with the disease, as well as knowledge mining the genomic and expression data to elucidate new pathways and therapeutic targets.
Although most cases are sporadic, ACC occurs in the context of inherited syndromes such as the Li-Fraumeni syndrome and Beckwith-Wiedemann syndrome (BWS). Li-Fraumeni syndrome is a dominant syndrome caused by a germline mutation in the p53 sup- pressor gene (TP53) on chromosome 17p13 [6,40]. Patients with Li-Fraumeni syndrome have a higher susceptibility to breast carcinoma, soft tissue sarcomas, brain tumors, osteosarcoma, leukemia and ACC [102]. The spectrum of germline mutations for Li-Fraumeni kindreds with ACC shows a shift outside of the normal hotspots in the DNA-binding domain. 60% of kindreds with ACC have mutations that clus- ter in a region of the p53 protein within non- DNA binding loops, the ß-sheet skeleton and the oligomerization domain [41]. TP53 has been reported to be mutated in approximately 25% of sporadic ACC [6]. In Brazil, a unique germline mutation at R337H has resulted in an increased incidence of ACC of 4-6 per million, but does not result in Li-Fraumeni syndrome in most patients; the result is that one in ten carriers develop ACC. In keeping with the distribution of ACC-associated mutations in Li-Fraumeni, R337H is a pH-sensitive mutation localized to
the oligomerization domain [42]. Polymorphisms in TP53 may also play a role in ACC pathogen- esis. In a study of Polish ACC patients, the pro- line allele of the R72P polymorphism was more prevalent in patients with ACC than in normal controls, suggesting that it may contribute to ACC susceptibility [43]. LOH for 17p has been reported in 85% or more of ACC [6], further supporting the role of p53 and possibly other genes within the region.
Interest in the role of IGF2 in ACC stems from an association of ACC with the overgrowth dis- order BWS. The disease arises from the misex- pression of genes from an imprinted domain at 11p15. Two regions in the imprinted domain have been implicated in the phenotypic variation of the disease, particularly tumor spectrum (i.e., the types of tumors associated with the disease). The telomeric region encompasses paternally expressed IGF2 and maternally expressed H19. Abnormalities that lead to the lack of H19 expres- sion and overexpression of IGF2, such as pater- nal uniparental disomy or aberrant methylation of H19, give rise to BWS with a preponderance of Wilms tumor. The role of loss of imprinting (LOI) at the H19 and IGF2 loci in tumorigen- esis is unclear. Targeted knockouts of H19 that overexpress IGF2 and do not express H19 when maternally inherited, do not demonstrate evi- dence of increased cancer incidence unless crossed into a background that confers an increased risk in itself [44,45]. For example, in the APCmin back- ground, IGF2 overexpression results in twice as many tumors as the APCmin background alone [46]. Targeted overexpression of IGF2 by trans- gene insertion, however, does lead to an increase in tumors in mice. The diversity of the tumor sites, including tissues that express the transgene as well as those that don’t, coupled with a long latency (most tumors are seen after 18 months) suggests that IGF2 overexpression may be tumor promoting but not tumor initiating [47]. The cen- tromeric region of the 11p15 imprinted domain includes the LIT1 transcript of the KvLQT1 gene and CDKN1C, the gene that encodes p57kip. Hypomethylation of LIT1 or mutations in CDKN1C have been implicated in BWS with a tumor spectrum that is skewed toward embryonal tumors such as rabdomyosarcoma, hepatoblas- toma, gondadoblastoma and ACC [48]. p57kip is a known negative regulator of cell cycle progres- sion, making it a prime candidate for a tumor suppressor in this region.
Approximately 90% of ACC demonstrate overexpression of IGF2 and LOH for the IGF2 locus has been found in greater than or equal to
IGF-2
B
IGF-1 receptor
+P
IRS-1
PI3K reg class IA
B
B
+P
B
B
B
14-3-3 5/8
Shc
B
B
PI3K cat class IA
14-3-3 ε
14-3-3 Y
14-3-3 /a
B
Z
ATP + phosphatidylinositol
GRB2
4,5-bisphosphate = ADP +
phosphatidylinositol
B
3,4,5-trisphosphate
B
PDK (PDPK1)
c-Raf-1
TR
B
PtdIns(3,4,5)P3
B
B
B
B
T
B
H-Ras
SOS
+P
B
B
TR
AKT(PKB)
+P
+P
+P
TR
+P
BAD
MDM2
CM
B
p53
*
GSK3 a/B
B
Bcl-XL
TR
TR
TR
B
TR
+P
C/EBPa
EGR1
SP1
c-Myc
95% of ACC [6]. In microarray studies for gene expression, IGF2 and associated binding pro- teins are consistently identified as differentially expressed in ACC as compared with normal adrenal tissue or adenomas, with most studies identifying IFG2 and/or IGF-1R as upregulated. It has been postulated that overexpression of IGF2 forms a positive autocrine loop that pro- motes cell proliferation, and that this is the phe- notype selected for in the ACC. However, this may not be the entire story. As mentioned above, the tumor spectrum of BWS is influenced by what region of the 11p15 imprinted domain is
disrupted, with disruptions in the centromeric region associated with ACC. p57kip is a regulator of CDK1, the cyclin-dependent kinase associ- ated with the G2/M transition, and as previously discussed, gene-expression studies have impli- cated the G2/M transition in ACC. Therefore, overexpression of IGF2 may be a consequence of a selection against p57kip expression, with loss of p57kip expression as an initiating event and overexpression of IGF2 serving to promote continued growth. These findings do not rule out the potential utility of using agents to block the IGF2 pathway in the treatment of ACC.
One must be aware, however, that if IGF2 over- expression is not the initiating event, it is pos- sible that ACC tumors may harbor or develop a subclone that is not dependent on IGF2 for proliferation, but retains the loss of p57kip expres- sion and, therefore, dysregulation of the G2/M transition. Thus, one may postulate that com- bining IGF2 inhibition with agents that perturb G2/M and result in catastrophic mitosis might be more effective than IGF2 inhibition alone.
Indeed, looking at how IGF2 signaling and p53 interact in the cell (FIGURE 1), it becomes apparent that IGF2 signaling induces strong anti- apoptotic responses that involve inactivating the pro-apoptotic protein BAD through phosphory- lation, binding to 14-3-3 proteins, and seques- tering it to the cytosol, as well as through activa- tion of MDM2 and, thus, inactivation of p53. Loss of p53 activity in turn removes a level of repression on the expression of IGF1R. As such, IGF2 signaling serves not only to promote cell proliferation, but it suppresses apoptosis through inactivating p53 as well as limiting the cell’s ability to kick-start the caspase cascade through inactivation of BAD and undergo apoptosis.
By combining all of the results in the literature, a picture of limited and refined genomic aberra- tion appears (TABLE 3, see end of article). Although it remains to be studied, the hypothesis would be that these regions represent the ‘driver’ rather than the ‘passenger’ events, since evidence for their involvement comes from multiple different sources. Indeed, many of the regions of genomic aberration harbor genes that have been implicated in cancer. For example, ACCs commonly have gains of 5p15, which harbors hTERT, the cata- lytic component of telomerase. Such a gain may increase the expression of telomerase and result in greater telomerase activity. A survey of telomere maintenance mechanisms in ACC demonstrated that telomerase activity, as opposed to other telo- mere maintenance mechanisms, was by far the most common mechanism active (79% of tumors examined) [49]. Overexpression of 8-catenin, encoded by CTNND2, has been implicated in regulating cadherin-p120 cell-cell adhesion in prostate cancer progression [50]. IGFBP6, highly overexpressed in ACC [19,20], localizes to 12q13, while IGFBP1 and IGFBP3 localize to 7p12-13. IGF binding proteins alter the bioavailability of IGF-1 and IGF-2. As mentioned above, dysregu- lation of IGF21IGFR1 signaling appears to play a role in ACC pathogenesis. TGFBR3 localizes to a region of loss at 1p21-p31. TGF-ß signaling has been implicated in adrenal development [6]. TGFBR3 may function as a tumor suppressor in
several tumor types, including breast cancer, non- small-cell lung cancer, ovarian cancer, prostate cancer and pancreatic cancer.
Despite the emerging picture of coherence in genomic aberrations, it is important to remem- ber that in adult ACC, no genomic aberra- tion has been found with a prevalence of more than 60%, and as yet, there is no evidence for changes that are both necessary and sufficient for ACC tumorigenesis. This suggests that there is significant heterogeneity in these tumors, and it is likely that no one treatment will be suc- cessful in all cases. The enrichment of G2/M pathways in the genes that are overexpressed in ACC with poor prognosis suggest that looking at treatments that target this transition may be beneficial. Inhibitors of Polo-like kinase 1, the Aurora kinases, and survivin are all under devel- opment or in early clinical trials. These types of agents could prove interesting for study in ACC. Another pathway with evidence for perturbation in ACC is the MAP-kinase pathways. Pathway analysis for the genomic regions associated with poor prognosis (using GeneGo) - the most sta- tistically significant map - shows a dysregulation of the MAP-kinase signaling cascade through the deletion of genes that negatively regulate the activation of Erk1/2. This leads to a system where the balance is tipped towards Erk1/2 activation. Little is known regarding the status of the MAP kinase cascade in ACC, and it is unclear where in the cascade an intervention would be most useful.
Current challenges
As previously noted, because ACC is a rare tumor, the accrual of sufficient numbers of patients for study in clinical trials is difficult. Furthermore, pharmaceutical firms may be reluctant to pursue development of agents where the size of the potential market for the agent is relatively small. Progress in the development of novel therapeutic agents would be greatly enhanced were there better experimental models for adrenocortical cancer. There are only three widely available ACC cell lines, SW13, H295 and H295R - a derivative of H295 selected for adherent growth. These cell lines do not adequately represent the heterogeneity of ACC. Three additional lines have been reported in the literature, but they are not widely available to the scientific community, and therefore the lines are not well characterized. ACC is rare in mice, although there are mouse models in which ACC has been observed. Most ACC has been observed in either gonadectomized animals or in
the context of specific targeted mutations such as PTEN, p53 and CDNK1C, but the disease has not been characterized in these models [51].
One model that may prove useful for ACC and other rare tumor types is the NCI-60 cell line panel. Used by the Developmental Therapeutics Program (DTP) of the National Cancer Institute (MD, USA) for screening and secondary testing of potential anticancer agents, the NCI-60 panel is diverse. It includes leu- kemias, melanomas and carcinomas of breast, ovary, kidney, colon, prostate, lung and CNS origin. The NCI-60 has been more fully charac- terized at the molecular level than any other set of cell lines. Since 1990, when the NCI-60 assay went into full operation, more than 100,000 chemical compounds (plus a large number of natural product extracts) have been tested in a 48-h growth inhibition assay [52,53]. In addition to the resulting pharmacological profiles of the NCI-60, the cells have been profiled for mRNA expression using cDNA microarrays [54,55] and oligonucleotide chips [56] for protein expression using 2D protein gel electrophoresis [57,58] and ‘reverse-phase’ lysate arrays [59,60], and for chro- mosomal aberrations [61,62]. These datasets can be combined to reveal relationships between drug sensitivity and acquired genetic changes at the molecular level.
The high-throughput methods that have been used to characterize ACC are compatible with those used to characterize the NCI-60. One can envision mining the NCI-60 data for correlations between a given pattern of gene expression or genomic alteration and drug response. Such methodologies have been suc- cessful in showing the association of L-aspara- ginase activity and the expression of asparagine synthetase in ovarian cancers, [55,61,63], as well as deriving gene-expression signatures predic- tive of response to therapy [64]. For example, given the identification of PINK1 and BUB1B as predictors of poor overall survival in ACC when the A crossover threshold of BUB1 minus the A crossover threshold of PINK1 is less than 6.32 [19], the next step might be to mine the NCI-60 data for that relationship and corre- late it to drug response. The hypothesis would be that certain drugs might be more or less effective depending on whether BUB1B and PINK1 show less difference in their expression levels. One would then measure the expression of PINK1 and BUB1B in SW13 and H295R, and use that to classify the cell lines as more or less like the poor prognosis group. Based on that classification, one could then predict how
SW13 and H295R would respond to drugs identified from the NCI-60 data. The next step would be to would be to test those pre- dictions by evaluating the sensitivity of SW13 and H295R in vitro, and possibly in a xenograft model if the in vitro results are promising. Such an experimental flow could rapidly translate a finding associated with poor prognosis into a testable therapeutic intervention.
Conclusion
The genomic and expression profiling of ACC to date has supported the role of the IGF2/IGF1R pathway in the disease and highlighted the importance of the p53 pathway. Additionally, the G2/M phase of the cell cycle has been high- lighted as enriched in differentially expressed genes between ACC and adenomas. This same set of processes has also been implicated in driv- ing the division of ACC into two groups with differences in survival. Continued knowledge mining will likely reveal additional pathways and processes that can be exploited for treatment.
Future perspective
We are now at a stage where there are sufficient data available on a reasonable number of ACC samples in the literature that, even though no single study has the ideal sample numbers for some of the advanced data mining techniques and further subclassification of tumors by molecular profile, a meta-analysis of the data is now feasible. In the same vein, diagnostic and prognostic indicators can be tested against data sets other than the ones from which they are derived, speeding up the translation to clinical practice. Such analyses will help direct in vitro and in vivo therapeutic models that can then be applied clinically with some accuracy. Additionally, although clonal in origin, ACC is heterogeneous at the cellular level. Thus, it will be imperative to investigate the clonal hetero- geneity within the individual adrenocortical tumors initially at the DNA level, and in all likelihood as technology allows, at the RNA and protein levels. One can envision obtain- ing data that not only suggests what combi- nations of pathways to target therapeutically to attack all clones, but also demonstrates the changes that the cancer incurs as it progresses from localized to metastatic disease, as well as how treatment affects those changes. The role of repeated biopsies and genomic surveillance to guide therapy becomes more realistic as the armamentarium of targeted agents increases. Finally, rarity of ACC in comparison with the
proportion of incidentally discovered adrenal tumors owing to the more frequent use of imag- ing studies suggests that ACC is likely occur- ring in the context of a specific background that confers vulnerability. Discovering what that context is will not only provide a further means of determining how to treat such adre- nal incidentalomas, but also contribute further to our understanding of how to treat ACC with targeted therapies.
Financial & competing interests disclosure
The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the sub- ject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
No writing assistance was utilized in the production of this manuscript.
Executive summary
Using genomic & transcriptomic information in adrenocortical carcinoma diagnosis
· Adrenocortical carcinoma (ACC) can be distinguished from adenomas by both genomic profiles and gene expression.
· Adult and pediatric ACC show different acquired changes, with adult ACC demonstrating significant heterogeneity in both ploidy and structural rearrangements within and between tumors. Pediatric ACC is almost universally characterized by amplification of 9q34 in addition to other changes.
· The mRNA expression levels of PINK1 and DLG7 are an effective two-gene signature for classifying tumors as benign or malignant.
Immunohistochemistry for IGF2 and Ki-67 is capable of more accurately classifying adrenal masses, with a Weiss score of 3 as benign or malignant.
Emerging prognostic indicators
· Amplifications of 6q, 7q, 12q and 19p and losses of 3, 8, 10p, 16q, 17q and 19q are significantly associated with poor survival.
· Expression of genes on 16p and 5q are enriched in tumors with better prognosis. In contrast, tumors with poor prognosis have enrichment for the expression of genes on 1q, 22q, 6q, 10p and 6p.
· A two-gene signature of PINK1 and BUB1B is as good as the Weiss score at predicting overall survival.
Moving toward pathway-driven treatments in adrenocortical carcinoma
· Mitotane, the current approved therapy for ACC, is effective in only 22% of cases. New treatments are needed.
The association of ACC with Li-Fraumeni and Beckwith-Wiedemann syndromes has implicated p53 and IGF2 as important in the disease.
· Expression profiling and mutational analysis supports the roles of p53 and IGF2 in ACC pathogenesis.
Additional pathways, such as the processes involved in the G2/M transition of the cell cycle, may also play a role in ACC, and therefore represent good therapeutic targets.
There are sufficient data to start comparing the genomic aberrations in ACC with gene expression. Telomerase, TGF-ß and 8-catenin all fall into regions of genomic aberration and have been implicated in tumorigenesis, or in the case of hTERT, ACC in particular.
Current challenges
· The three widely available cell lines are insufficient to represent the diversity of ACC. Additional cell lines are needed.
Tumors arising in mouse models of ACC need to be characterized and compared with the human disease.
The NCI-60 panel of cell lines may be a resource for rare tumor therapeutic development.
Bibliography
Papers of special note have been highlighted as:
· of interest
== of considerable interest
1. Greenberg PH, Marks C: Adrenal cortical carcinoma: a presentation of 22 cases and a review of the literature. Am. Surg. 44, 81-85 (1978).
2. Karakousis CP, Rao U, Moore R: Adrenal adenocarcinomas: histologic grading and survival. J. Surg. Oncol. 29, 105-111 (1985).
3. Kjellman M, Kallioniemi OP, Karhu R et al .: Genetic aberrations in adrenocortical tumors detected using comparative genomic hybridization correlate with tumor size and malignancy. Cancer Res. 56, 4219-4223 (1996).
4. Venkatesh S, Hickey RC, Sellin RV, Fernandez JF, Samaan NA: Adrenal cortical carcinoma. Cancer 64, 765-769 (1989).
5. Thompson NW, Cheung PS: Diagnosis and treatment of functioning and nonfunctioning adrenocortical neoplasms including incidentalomas. Surg. Clin. North Am. 67, 423-436 (1987).
6. Libe R, Fratticci A, Bertherat J: Adrenocortical cancer: pathophysiology and clinical management. Endocr. Relat. Cancer 14, 13-28 (2007).
Excellent review of the pathobiology of adrenocortical carcinoma (ACC).
7. Dohna M, Reincke M, Mincheva A et al .: Adrenocortical carcinoma is characterized by a high frequency of chromosomal gains and
high-level amplifications. Genes Chromosomes Cancer 28, 145-152 (2000).
8. Sidhu S, Marsh DJ, Theodosopoulos G et al .: Comparative genomic hybridization analysis of adrenocortical tumors. J. Clin. Endocrinol. Metab. 87, 3467-3474 (2002).
9. Zhao J, Roth J, Bode-Lesniewska B et al .: Combined comparative genomic hybridization and genomic microarray for detection of gene amplifications in pulmonary artery intimal sarcomas and adrenocortical tumors. Genes Chromosomes Cancer 34, 48-57 (2002).
10. Zhao J, Speel EJ, Muletta-Feurer S et al .: Analysis of genomic alterations in sporadic adrenocortical lesions. Gain of chromosome 17 is an early event in adrenocortical tumorigenesis. Am. J. Pathol. 155, 1039-1045 (1999).
fsg
Review Bussey & Demeure
11. Stephan EA, Chung TH, Grant CS et al .: Adrenocortical carcinoma survival rates correlated to genomic copy number variants. Mol. Cancer Ther. 7, 425-431 (2008).
12. Figueiredo BC, Stratakis CA, Sandrini R et al .: Comparative genomic hybridization analysis of adrenocortical tumors of childhood. J. Clin. Endocrinol. Metab. 84, 1116-1121 (1999).
13. James LA, Kelsey AM, Birch JM, Varley JM: Highly consistent genetic alterations in childhood adrenocortical tumours detected by comparative genomic hybridization. Br. J. Cancer 81, 300-304 (1999).
14. de Fraipont F, El Atifi M, Cherradi N et al .: Gene expression profiling of human adrenocortical tumors using complementary deoxyribonucleic acid microarrays identifies several candidate genes as markers of malignancy. J. Clin. Endocrinol. Metab. 90, 1819-1829 (2005).
15. Giordano TJ, Thomas DG, Kuick R et al .: Distinct transcriptional profiles of adrenocortical tumors uncovered by DNA microarray analysis. Am. J. Pathol. 162, 521-531 (2003).
16. Slater EP, Diehl SM, Langer P et al .: Analysis by cDNA microarrays of gene expression patterns of human adrenocortical tumors. Eur. J. Endocrinol. 154, 587-598 (2006).
17. Velazquez-Fernandez D, Laurell C, Geli J et al .: Expression profiling of adrenocortical neoplasms suggests a molecular signature of malignancy. Surgery 138, 1087-1094 (2005).
18. West AN, Neale GA, Pounds S et al .: Gene expression profiling of childhood adrenocortical tumors. Cancer Res. 67, 600-608 (2007).
19. de Reynies A, Assie G, Rickman DS et al .: Gene expression profiling reveals a new classification of adrenocortical tumors and identifies molecular predictors of malignancy and survival. J. Clin. Oncol. 27, 1108-1115 (2009).
· Identifies gene signatures for diagnosis and prognosis of ACC.
20. Giordano TJ, Kuick R, Else T et al .: Molecular classification and prognostication of adrenocortical tumors by transcriptome profiling. Clin. Cancer Res. 15, 668-676 (2009).
· A good example of taking a pathway- enrichment approach to characterizing ACC from expression data.
21. Soon P, Gill A, Benn D et al .: Microarray gene expression and immunohistochemistry analyses of adrenocortical tumours identify IGF2 and Ki-67 as useful in differentiating carcinomas from adenomas. Endocr. Relat. Cancer (2009) (Epub ahead of print).
22. Cohn K, Gottesman L, Brennan M: Adrenocortical carcinoma. Surgery 100, 1170-1177 (1986).
23. Lee JE, Berger DH, el-Naggar AK et al .: Surgical management, DNA content, and patient survival in adrenal cortical carcinoma. Surgery 118, 1090-1098 (1995).
24. Didolkar MS, Bescher RA, Elias EG, Moore RH: Natural history of adrenal cortical carcinoma: a clinicopathologic study of 42 patients. Cancer 47, 2153-2161 (1981).
25. Martz F, Straw JA: The in vitro metabolism of 1-(o-chlorophenyl)-1-(p-chlorophenyl)-2,2- dichloroethane (o,p’-DDD) by dog adrenal mitochondria and metabolite covalent binding to mitochondrial macromolecules: a possible mechanism for the adrenocorticolytic effect. Drug Metab. Dispos. 5, 482-486 (1977).
26. Martz F, Straw JA: Metabolism and covalent binding of 1-(o-chlorophenyl)-1-(p- chlorophenyl)-2,2-dichloroethane (o,p’- DDD). Correlation between adrenocorticolytic activity and metabolic activation by adrenocortical mitochondria. Drug Metab. Dispos. 8, 127-130 (1980).
27. Lindhe O, Skogseid B, Brandt I: Cytochrome P450-catalyzed binding of 3-methylsulfonyl- DDE and o,p’-DDD in human adrenal zona fasciculata/reticularis. J. Clin. Endocrinol. Metab. 87, 1319-1326 (2002).
28. Schteingart DE, Sinsheimer JE, Counsell RE et al .: Comparison of the adrenalytic activity of mitotane and a methylated homolog on normal adrenal cortex and adrenal cortical carcinoma. Cancer Chemother. Pharmacol. 31, 459-466 (1993).
29. Stigliano A, Cerquetti L, Borro M et al .: Modulation of proteomic profile in H295R adrenocortical cell line induced by mitotane. Endocr. Relat. Cancer 15, 1-10 (2008).
30. Bergenstal DM, Lipsett M, Moy RN, Hertz R: Regression of adrenal cancer and suppression of adrenal function in man by o,p’-DDD. Trans. Am. Physicians 72, 341-350 (1959).
31. Haak HR, Hermans J, van de Velde CJ et al .: Optimal treatment of adrenocortical carcinoma with mitotane: results in a consecutive series of 96 patients. Br. J. Cancer 69, 947-951 (1994).
32. Pommier RF, Brennan MF: An eleven-year experience with adrenocortical carcinoma. Surgery 112, 963-970; discussion 970-971 (1992).
33. Stojadinovic A, Ghossein RA, Hoos A et al .: Adrenocortical carcinoma: clinical, morphologic, and molecular characterization. J. Clin. Oncol. 20, 941-950 (2002).
34. Terzolo M, Angeli A, Fassnacht M et al .: Adjuvant mitotane treatment for adrenocortical carcinoma. N. Engl. J. Med. 356, 2372-2380 (2007).
. First study showing efficacy of mitotane in the adjuvant setting.
35. Decker RA, Elson P, Hogan TF et al .: Eastern Cooperative Oncology Group study 1879: mitotane and adriamycin in patients with advanced adrenocortical carcinoma. Surgery 110, 1006-1013 (1991).
36. van Slooten H, Moolenaar AJ, van Seters AP, Smeenk D: The treatment of adrenocortical carcinoma with o,p’-DDD: prognostic implications of serum level monitoring. Eur. J. Cancer Clin. Oncol. 20, 47-53 (1984).
37. Barzon L, Fallo F, Sonino N, Daniele O, Boscaro M: Comment - is there a role for low doses of mitotane (o,p’-DDD) as adjuvant therapy in adrenocortical carcinoma? J. Clin. Endocrinol. Metab. 84, 1488-1489 (1999).
38. Vassilopoulou-Sellin R, Guinee VF, Klein MJ et al .: Impact of adjuvant mitotane on the clinical course of patients with adrenocortical cancer. Cancer 71, 3119-3123 (1993).
39. Berruti A, Terzolo M, Sperone P et al .: Etoposide, doxorubicin and cisplatin plus mitotane in the treatment of advanced adrenocortical carcinoma: a large prospective Phase II trial. Endocr. Relat. Cancer 12, 657-666 (2005).
. The combination in this study is now known as the Italian protocol and comprises one arm of the FIRM-ACT trial to identify the most efficacious combination for ACC from current treatments.
40. Srivastava S, Zou ZQ, Pirollo K, Blattner W, Chang EH: Germ-line transmission of a mutated p53 gene in a cancer-prone family with Li-Fraumeni syndrome. Nature 348, 747-749 (1990).
41. Olivier M, Goldgar DE, Sodha N et al .: Li-Fraumeni and related syndromes: correlation between tumor type, family structure, and TP53 genotype. Cancer Res. 63, 6643-6650 (2003).
42. Zambetti GP: The p53 mutation ‘gradient effect’ and its clinical implications. J. Cell Physiol. 213, 370-373 (2007).
· Good review of p53 mutations, particularly hypomorphs, and their role in cancer.
43. Ignaszak-Szczepaniak M, Horst-Sikorska W, Sawicka J, Kaczmarek M, Slomski R: The TP53 codon 72 polymorphism and predisposition to adrenocortical cancer in Polish patients. Oncol. Rep. 16, 65-71 (2006).
Genomic & expression profiling of adrenocortical carcinoma
44. Blake JA, Bult CJ, Eppig JT, Kadin JA, Richardson JE: The Mouse Genome Database genotypes: phenotypes. Nucleic Acids Res. 37, D712-D719 (2009).
45. Bult CJ, Eppig JT, Kadin JA, Richardson JE, Blake JA: The Mouse Genome Database (MGD): mouse biology and model systems. Nucleic Acids Res. 36, D724-D728 (2008).
46. Sakatani T, Kaneda A, Iacobuzio-Donahue CA et al .: Loss of imprinting of Igf2 alters intestinal maturation and tumorigenesis in mice. Science 307, 1976-1978 (2005).
47. Moorehead RA, Fata JE, Johnson MB, Khokha R: Inhibition of mammary epithelial apoptosis and sustained phosphorylation of Akt/PKB in MMTV-IGF-II transgenic mice. Cell Death Differ. 8, 16-29 (2001).
48. Weksberg R, Nishikawa J, Caluseriu O et al .: Tumor development in the Beckwith- Wiedemann syndrome is associated with a variety of constitutional molecular 11p15 alterations including imprinting defects of KCNQ1OT1. Hum. Mol. Genet. 10, 2989-3000 (2001).
· Demonstrates the association of tumor spectrum with the chromosomal location of the abnormality giving rise to Beckwith-Wiedemann syndrome.
49. Else T, Giordano TJ, Hammer GD: Evaluation of telomere length maintenance mechanisms in adrenocortical carcinoma. J. Clin. Endocrinol. Metab. 93, 1442-1449 (2008).
50. Lu Q, Dobbs LJ, Gregory CW et al .: Increased expression of 8-catenin/neural plakophilin-related armadillo protein is associated with the downregulation and redistribution of E-cadherin and p120ctn in human prostate cancer. Hum. Pathol. 36, 1037-1048 (2005).
51. Krupke DM, Begley DA, Sundberg JP, Bult CJ, Eppig JT: The Mouse Tumor Biology Database. Nat. Rev. Cancer 8, 459-465 (2008).
” Good resource for finding descriptions of tumors and tumor phenotypes for transgenic mice.
52. Boyd MR, Paull KD: Some practical considerations and applications of the National Cancer Institute in vitro anticancer drug discovery screen. Drug Dev. Res. 34, 91-109 (1995).
53. Shoemaker RH, Monks A, Alley MC et al .: Development of human tumor cell line panels for use in disease-oriented drug screening. Prog. Clin. Biol. Res. 276, 265-286 (1988).
54. Ross DT, Scherf U, Eisen MB et al .: Systematic variation in gene expression patterns in human cancer cell lines. Nat. Genet. 24, 227-235 (2000).
55. . Scherf U, Ross DT, Waltham M et al .: A gene expression database for the molecular pharmacology of cancer. Nat. Genet. 24, 236-244 (2000).
56. Staunton JE, Slonim DK, Coller HA et al .: Chemosensitivity prediction by transcriptional profiling. Proc. Natl Acad. Sci. USA 98, 10787-10792 (2001).
57. Li G, Waltham M, Anderson NL, Unsworth E, Treston A,Weinstein JN: Rapid mass spectrometric identification of proteins from two-dimensional polyacrylamide gels after in gel proteolytic digestion. Electrophoresis 18, 391-402 (1997).
58. Myers TG, Anderson NL, Waltham M et al .: A protein expression database for the molecular pharmacology of cancer. Electrophoresis 18, 647-653 (1997).
59. Nishizuka S, Charboneau L, Young L et al .: Proteomic profiling of the NCI-60 cancer cell lines using new high-density reverse- phase lysate microarrays. Proc. Natl Acad. Sci. USA 100, 14229-14234 (2003).
60. Nishizuka S, Chen ST, Gwadry FG et al .: Diagnostic markers that distinguish colon and ovarian adenocarcinomas: identification by genomic, proteomic, and tissue array profiling. Cancer Res. 63, 5243-5250 (2003).
61. Bussey KJ, Chin K, Lababidi S et al .: Integrating data on DNA copy number with gene expression levels and drug sensitivities in the NCI-60 cell line panel. Mol. Cancer Ther. 5, 853-867 (2006).
62. Roschke AV, Tonon G, Gehlhaus KS et al .: Karyotypic complexity of the NCI-60 drug-screening panel. Cancer Res. 63, 8634-8647 (2003).
63. Lorenzi PL, Reinhold WC, Rudelius M et al .: Asparagine synthetase as a causal, predictive biomarker for L-asparaginase activity in ovarian cancer cells. Mol. Cancer Ther. 5, 2613-2623 (2006).
64. Potti A, Dressman HK, Bild A et al .: Genomic signatures to guide the use of chemotherapeutics. Nat. Med. 12, 1294- 1300 (2006).
· Demonstrates how one can derive clinically applicable gene signatures for drug response from the NCI-60 data.
65. Bettio D, Rizzi N, Giardino D et al .: Translocation (7;17)(q22;p13) as a sole karyotypic change in an adrenal adenoma. Cancer Genet. Cytogenet. 103, 180-181 (1998).
66. Gordon RD, Stowasser M, Martin N et al .: Karyotypic abnormalities in benign adrenocortical tumors producing aldosterone. Cancer Genet. Cytogenet. 68, 78-81 (1993).
67. Herrmann ME, Rydstedt LL, Talpos GB et al .: Chromosomal aberrations in two adrenocortical tumors, one with a rearrangement at 11p15. Cancer Genet. Cytogenet. 75, 111-116 (1994).
68. Limon J, Dal Cin P, Kakati S, Huben RP, Rao U, Sandberg AA: Cytogenetic findings in a primary adrenocortical carcinoma. Cancer Genet. Cytogenet. 26, 271-277 (1987).
69. Limon J, Dal Cin P, Gaeta J, Sandberg AA: Translocation t(4;11)(q35;p13) in an adrenocortical carcinoma. Cancer Genet. Cytogenet. 28, 343-348 (1987).
70. Marks JL, Wyandt HE, Beazley RM et al .: Cytogenetic studies of an adrenal cortical carcinoma. Cancer Genet. Cytogenet. 61, 96-98 (1992).
71. Mertens F, Kullendorff C-M, Moell C, Alumets J, Mandahl N: Complex karyotype in a childhood adrenocortical carcinoma. Cancer Genet. Cytogenet. 105, 190-192 (1998).
Websites
101. Mitelman Database of Chromosome Aberrations in Cancer. 2009. Mitelman F, Johansson B, Mertens F(Eds.) cgap.nci.nih.gov/Chromosomes/Mitelman
102. Online Mendelian Inheritance in Man, OMIM (TM). Johns Hopkins University, Baltimore, MD, USA. MIM number: #151623, last edited: December 11, 2008. www.ncbi.nlm.nih.gov/omim
Affiliations
” Kimberly J Bussey Associate Investigator, Clinical Translational Research Division, Translational Genomics Research Institute, 445 N. 5th Street, Phoenix, AZ 85004, USA Tel .: +1 602 343 8817 Fax: +1 602 343 8740 kbussey@tgen.org
n Michael J Demeure Senior Investigator, Clinical Translational Research Division, Translational Genomics Research Institute, Director, Endocrine Tumors Center, Scottsdale Healthcare, 10460 N. 92nd St, Suite 200, Scottsdale, AZ 85258, USA
Tel .: +1 480 323 1250 Fax: +1 480 323 1259 mdemeure@tgen.org
fsg
| Smallest region of overlap | Array CGH refinement | Genes | |
|---|---|---|---|
| Segment starts | Segment stops | ||
| Gains | |||
| 1q22-q25 | |||
| 4p15-pter 5p15 | |||
| 0 | 16832642 | SLC6A18, IRX4, LOC728178, LOC100130744, EEF1AL11, LOC391741, FLJ44896, TERT, LOC645763, IRX1, ZNF622, LOC100133292, PDCD6, MARCH11, MED10, PLEKHG4B, LOC731559, LOC25845, LOC100130063, LOC100132773, LOC100132536, RPL29P13, ZDHHC11B, LOC100132605, CEP72, UNQ1870, LOC100132778, SEMA5A, NDUFS6, EXOC3, SDHA, LOC442132, SLC9A3, IRX2, TPPP, LOC100129064, ROPN1L, LOC100129204, ADAMTS16, LOC391739, BRD9, LOC100128803, AHRR, FAM105B, LOC285577, ANKRD33B, LOC116349, LOC100128508, C5orf38, FAM134B, KIAA0947, MRPL36, SLC12A7, TAS2R1, DAP, SRD5A1, LOC642954, ADCY2, C5orf49, FLJ25076, LOC100131397, SNORD123, LOC442131, NSUN2, LOC100130748, FBXL7, DNAH5, CLPTM1L, FAM105A, NKD2, TRIO, MYO10, CMBL, ZDHHC11, FAM173B, LOC255167, MTRR, POLS, SDHAP3, LOC100128645, SLC6A3, LOC728613, ANKH, CTNND2, MARCH6, TRIP13, LOC651419, SLC6A19, LOC100132531, FLJ33360, FASTKD3, LOC340094, LOC100128382, LPCAT1, CCDC127, LOC389257, LOC402198, CCT5, LOC729506 | |
| 5q12-q13 5q32-qter | 61912472 145524108 | 64949791 174887118 | RNF180, FAM159B, SFRS12IP1, CENPK, MRPL49P1, LRRC70, ADAMTS6, HTR1A, PPWD1, TRIM23, ISCA1L, RGS7BP, SDCCAG10 GPX3, PCYOX1L, LOC100130088, PDE6A, C5orf47, RPL7AP33, LOC345471, ZNF300, LARP1, CSF1R, RBM27, MST150, SPINK5, DOCK2, RPS14, SPARC, MAT2B, C5orf41, ERGIC1, SH3TC2, HAND1, SLIT3, FBLL1, NUDCD2, DRD1, FGF18, LOC285591, hCG_1980447, NDST1, LOC728287, SNRPEP1, GALNT10, TIGD6, MSX2, LOC100131743, GABRB2, LOC651815, LOC100131740, TIMD4, LOC100129458, HMGXB3, RPS20P4, LOC133874, ADRB2, GPR151, G3BP1, CCDC69, STK32A, ODZ2, DUSP1, ATP10B, WWC1, RNF145, SCGB3A2, ANXA6, SLC36A3, BNIP1, SLC2A3P1, LOC134466, PANK3, GAPDHL16, LOC100128833, LOC100128619, TLX3, RANBP17, PTTG1, DPYSL3, IRGM, CYFIP2, SOX30, CCNG1, TCERG1, GLULL1, LOC100131520, LOC645398, USP12P1, FLJ40453, ICHTHYIN, GRIA1, FOXI1, RPL10P8, TCOF1, LOC724105, SAP30L, FBXW11, LOC727947, LOC100129748, IL12B, CPEB4, RPL10P9, NEURL1B, UNQ9374, SPINK1, LOC100129887, HAVCR2, LOC574080, GLRA1, LOC153469, JAKMIP2, HAVCR1, LOC391847, ATOX1, SFXN1, TTC1, HTR4, SLC36A2, LOC728095, ATP6V0E1, ARSI, GABRA6, PWWP2A, LARS, LOC729421, LSM11, CCNJL, NPM1, LOC100127922, RPL7P, ADRA1B, LOC728264, FAM114A2, LOC100131033, LOC257358, C5orf46, C5orf50, SYNPO, LOC100128482, SNORA74B, LOC100130394, CCDC99, EFCAB9, STC2, LOC391844, FLJ16171, GM2A, LOC728145, LOC100129026, SLC36A1, LOC100128543, MRP63P6, EBF1, GABRP, RPLP0P9, LOC442142, MRPL22, UBLCP1, LOC727846, CDX1, MED7, MFAP3, ITK, GABRG2, LOC100128116, GRPEL2, PDGFRB, POU4F3, LOC100131897, LOC401218, SLU7, CNOT8, DCTN4, RPL7P20, CLINT1, RBM22, LOC100130177, FLJ41603, ABLIM3, ERPL2, TNIP1, CAMK2A, SPINK6, FAT2, FABP6, C5orf4, KIF4B, NKX2-5, FBXO38, GLRXL, GEMIN5, LOC100132936, SGCD, MYOZ3, BOD1, CSNK1A1, IL17B, THG1L, LOC402233, KRT18P41, hCG_1641617, FAM71B, LCP2, LOC644762, LOC100128769, C1QTNF2, GABRA1, NMUR2, AFAP1L1, SH3PXD2B, LOC729170, PPP1R2P3, KCNIP1, UBTD2, RPLP2P2, CD74, RARS, ADAM19, SPINK9, SLC26A2, LOC100128059, PPARGC1B, SLC6A7, LOC100128898, PPP2R2B, STK10, SPINK5L2, SPINK5L3, C5orf54, SPINK7, KCNMB1, RPS15P6, RPL26L1, HMMR, C5orf40, HMP19, LOC100130260 |
| 6q15-6q16.1 | 93777323 | 97353792 | LOC100132830, KIAA0776, MANEA, FUT9, KRT18P50, FHL5, EPHA7, CYCSP17, TSG1, COPS5P, GPR63, RPS7P8 |
| CGH: Comparative genomic hybridization. | |||
| Smallest region of overlap | Array CGH refinement | Genes | |
|---|---|---|---|
| Segment starts | Segment stops | ||
| Gains (cont.) | |||
| 7p11.2-p14 | 44697321 | 55524868 | LOC392027, ADCY1, SUNC1, IGFBP1, LANCL2, MRPS23P1, EGFR, GDI2P, MGC16075, LOC222052, HUS1, TBRG4, LOC100129050, LOC100129276, C7orf65, PURB, LOC100131871, LOC647102, SLC25A5P3, CALM1P2, COBL, LOC730338, ZMIZ2, VSTM2A, SNORA5B, DKFZp564N2472, LOC100132119, GRB10, LOC100128364, NACAD, LOC100131447, LOC100128019, LOC653175, LOC643168, CDC14C, DDC, IKZF1, LOC100133256, LOC100130988, RNU1P7, LOC100133258, ABCA13, MRPL42P4, LOC100129159, SNORA9, PKD1L1, SNORA5C, LOC730234, MYO1G, FLJ21075, LOC100128600, IGFBP3, LOC100132224, SEPT13, LOC442304, SNORA5A, FLJ45974, UPP1, TNS3, ECOP, CDC14BL, LOC100130122, LOC100130121, LOC100129343, LOC392030, CCM2, LOC100132447, LOC100128734, H2AFV, LOC100129904, LOC642663, SEC61G, OGDH, tcag7.940, PPIA, LOC100129427, LOC100133177, LOC100130914, C7orf57, LOC100133275, LOC100129619, FIGNL1, LOC730235, RAMP3, LOC100132575, C7orf40, ZPBP, LOC100130913, VWC2 |
| 7q36.1 | 140163610 | 147901426 | tcag7.926, TAS2R39, FAM115B, TRPV5, RPL26P24, LOC100130415, TAS2R38, KIAA1147, OR2A42, TRBV6-3, TRBV6-2, TRBV6-1, TRBV5-8, TRBV6-7, TRBV6-6, TRBV6-5, TRBV6-4, TRBV5-7, TRBV5-6, TRYX3, LOC136242, CLCN1, ZYX, TRB@, tcag7.1231, OR2A14, TRY6, FLJ43692, BRAF, OR9A2, PS3, OR2A13P, TAS2R62P, OR9P1P, OR9N1P, OR6V1, LOC100132804, TRBV5-3, LOC100129105, WEE2, LOC93432, AGK, LOC780811, CASP2, TRPV6, OR9A3P, ARHGEF5, GSTK1, LOC441294, OR9A4, PRSS1, FAM115C, LOC643308, C7orf34, CTAGE6, MGAM, CTAGE4, LOC642627, OR10AC1P, TRBV10-2, TRBV10-1, TRBV9, TRBV8-2, TRBV11-3, TRBV11-2, TRBV11-1, TRBV10-3, OR2Q1P, TAS2R41, TRBV8-1, TRBV7-9, FLJ40852, tcag7.1217, TRBVA, TRBV30, CLEC5A, OR2A2, PRSS2, TRBV29-1, TRBV28, MRPS33, FAM115A, TRBV22-1, TAS2R4, TAS2R3, SSBP1, TRY2P, LOC100131883, TPK1, OR2A7, TRBV7-4, TRBV7-3, TRBV7-2, TRBV7-1, OR2A9P, TRBV7-7, TRBV20-1, TRBV7-5, RPL26P22, TRBV6-9, TRBV6-8, TRY5, LOC100124692, TRBV23-1, LOC100130169, TRBV21-1, TRY7, TRBV27, TRBV26, TRBV25-1, TRBV24- 1, TRBV19, TRBV18, TRBV5-1, TRBV4-3, TRBV4-2, TRBV4-1, TRBV5-5, TRBV5-4, LOC652678, TRBV5-2, OR2R1P, LOC728377, TRBV3-2, TRBV3-1, TRBV7-8, TRBV7-6, MOXD2, OR2A12, TAS2R40, LOC780812, OR2F2, OR2A41P, EPHB6, TRBV13, TRBV12- 5, TRBV12-4, TRBV12-3, TRBV17, TRBV16, TRBV15, TRBV14, PIP, TRBV12-2, TRBV12-1, TRBJ2-5, TRBJ2-4, TRBJ2-3, TRBJ2-2P, TRBV2, TRBV1, TRBJ2-7, TRBJ2-6, TRBJ2-2, TRBJ2-1, LOC650172, OR6B1, KEL, OR2F1, LOC100131199, EPHA1, OR6W1P, LOC100129514, LOC100133146, TRBJ1-2, TRBJ1-1, TRBD2, TRBD1, TRBJ1-6, TRBJ1-5, TRBJ1-4, TRBJ1-3, CNTNAP2, OR9A1P, TRBC2, TRBC1, OR2A25, TAS2R5, FAM131B, OR2A1, OR2A20P, OR2A5, TMEM139, TRY3, TAS2R60, OR2A15P, TRBVB, OR2A3P, OR2AO1P, NOBOX |
| 9q34 | |||
| 12q13 | 50124921 | 52868703 | PFDN5, SPRYD3, ATF7, KRT72, C12orf10, SP1, KRT5, AMHR2, GRASP, HOTAIR, PCBP2, KRT80, EIF4B, LOC100129362, MAP3K12, SCN8A, HOXC13, KRT1, HOXC5, HOXC10, KRT85, KRT73, HOXC11, LOC100128683, KRT86, ZNF740, KRT81, KRT122P, KRT121P, LOC100128678, CSAD, KRT124P, C12orf44, KRT123P, KRT18, KRT76, KRT6A, KRT126P, HIGD1D, LOC643898, LOC341412, ACVRL1, KRT71, TARBP2, SLC4A8, LOC100127967, KRT2, KRT75, HOXC6, KRT74, LOC728503, IGFBP6, TENC1, NPFF, ESPL1, LOC728522, FIGNL2, LOC100129802, LOC100127971, PRR13, LOC644222, MFSD5, KRT3, KRT6B, LOC100129509, KRT7, AAAS, LOC100128815, SP7, LOC100129218, LOC100127976, KRT6C, NR4A1, HOXC8, KRT82, CALCOCO1, OR7E47P, KRT79, KRT84, FLJ33996, LOC400036, SMUG1, LOC100128395, ACVR1B, KRT4, SOAT2, ATP5G2, KRT8, KRT77, RARG, KRT78, ITGB7, HOXC12, ANKRD33, HOXC4, HOXC9, KRT83, LOC728698 |
| 12q24 | |||
| 19p13.3 | |||
| CGH: Comparative genomic hybridization. | |||
| Smallest region of overlap | Array CGH refinement | Genes | |
|---|---|---|---|
| Segment starts | Segment stops | ||
| Gains (cont.) | |||
| 20q12 | 37107705 | 39462651 | LOC100128988, LOC100130936, EMILIN3, MAFB, ZHX3, HSPEP1, PLCG1, PRO0628, LPIN3, LOC100127898, TOP1, PRO0628 |
| 20q13.2 | 51955422 | 54373434 | C20orf108, RPL12P4, CYP24A1, DOK5, PFDN4, MC3R, CBLN4, BCAS1 |
| 20q13.32- 20q13.33 | 56663800 | 60164420 | LOC100131806, CDH4, LOC100128291, GNASAS, TUBB1, PHACTR3, CDH26, LOC729296, SYCP2, EDN3, C20orf177, C20orf197, TH1L, ATP5E, SLMO2, TAF4, CTSZ, MRPS16P, PPP1R3D, MTCO2L, ZNF831, PSMA7, STX16, RP11-429E11.3, LOC284757, LOC100131417, GNAS, LOC100131710, NPEPL1, SS18L1, LSM14B, LOC645605 |
| Losses | |||
| 1p21-31 | 91953420 | 93311732 | RPAP2, BTBD8, EVI5, LOC100128094, FAM69A, RPL5, LOC100127934, LOC100133115, GFI1, KIAA1107, TGFBR3, LOC646821, SNORD21, GLMN, BRDT, SNORA66, LOC646817, C1orf146, ABHD7 |
| 2p21-23 | |||
| 2q33.1- 2q33.2 | 202317641 | 204129616 | FLJ39061, WDR12, NBEAL1, BMPR2, ALS2, SNORD11, hCG_2044152, LOC442063, SUMO1, SNORD70, LOC442064, LOC339809, RPL39P14, LOC645805, SNORD11B, FAM117B, NOP58, ABI2, RAPH1, ICA1L, FZD7, CYP20A1, LOC100132064, ALS2CR8, RPL7P14, LOC100133283, MRPL50P2, PFTK2, KRT8P15, RPL12P16 |
| 2q35 | 218772082 | 219445588 | VIL1, ARPC2, RQCD1, PLCD4, STK36, AAMP, USP37, RPL19P5, PNKD, C2orf62, GPBAR1, SLC11A1, TMBIM1, ZNF142, TTLL4, PRKAG3, HMGB1L9, BCS1L, CYP27A1, WNT6, CTDSP1, RNF25 |
| 3p21.31 | 46549279 | 50342593 | CDH29, NCKIPSD, CELSR3, ALS2CL, CCDC51, DAG1, LOC729756, LOC100133032, TRAIP, SEMA3B, TESSP2, KLHDC8B, C3orf71, PLXNB1, NAT6, LRRC2, QRICH1, LOC100131840, MST1, SLC38A3, C3orf62, KLHL18, WDR6, CCDC12, RASSF1, ZNF589, SPINK8, TUSC2, PRKAR2A, BSN, TMIE, IMPDH2, LAMB2, CSPG5, RBM5, RBM6, NME6, C3orf45, LOC100129060, UBA7, CAMP, RNF123, PTPN23, C3orf54, TESSP5, DALRD3, CCDC36, FBXW12, CDC25A, PFKFB4, RHOA, PTH1R, TREX1, IP6K2, MST1R, GMPPB, ATRIP, RPL17P16, MON1A, GPX1, MRPS18AP1, ARIH2, MRP63P3, SLC26A6, IP6K1, HYAL2, USP4, LUZPP1, CAMKV, UQCRC1, GNAT1, SEMA3F, LOC389120, SCAP, APEH, IFRD2, C3orf75, TSP50, TDGF1, KIF9, NBEAL2, MAP4, LOC646498, HYAL3, LOC729280, MYL3, SHISA5, PHF5EP, CCDC72, TMEM89, QARS, COL7A1, HYAL1, USP19, AMIGO3, LOC100129354, STGC3, C3orf60, LOC100131951, UCN2, GNAI2, AMT, DHX30, CCDC71, TCTA, SMARCC1, NICN1, P4HTM, SLC25A20, LOC100132677, SETD2, LOC100132146, LOC100130278 |
| 3p21.1 | 52539782 | 53736895 | PRKCD, TKT, MUSTN1, PBRM1, LOC100132069, SNORD19B, SNORD19, ITIH3, NT5DC2, GNL3, LOC100130124, RPS25P4, LOC553148, DCP1A, SNORD69, SFMBT1, RFT1, CACNA1D, GLT8D1, LOC401068, SPCS1, TMEM110, LOC440957, ITIH4, NEK4, ITIH1 |
| 3p14.3 | 56211013 | 58459226 | RPS8P5, PDE12, SLMAP, CCDC66, PDHB, LOC100128209, PXK, PPIAP16, LOC100128353, ARF4, RPP14, KCTD6, FLNB, DNASE1L3, SPATA12, C3orf63, DNAH12, APPL1, ARHGEF3, ERC2, ASB14, HESX1, ABHD6, IL17RD, LOC100132534, FAM116A |
| 9p13.3 | 33114815 | 34371776 | SNORD121B, LOC100129970, C9orf24, NFX1, TRBV20OR9-2, CHMP5, B4GALT1, TRBVAOR9-2, AQP3, ANKRD18B, KIF24, UBE2R2, AQP7, LOC100132563, SUGT1P, ANXA2P2, WDR40A, PRSS3, NUDT2, SNORD121A, UBAP1, UBAP2, PTENP1, SERPINHP1, IMPDH1P1, TRBV24OR9-2, SPINK4, TRBV22OR9-2, RPL35AP2, TRBV26OR9-2, bA255A11.4, RPS8P9, TRBV21OR9-2, TRBV29OR9-2, NOL6, KIAA1161, TRBV25OR9-2, TRBV23OR9-2, BAG1 |
| 11q24-qter | |||
| CGH: Comparative genomic hybridization. | |||
| Smallest region of overlap | Array CGH refinement | Genes | |
|---|---|---|---|
| Segment starts | Segment stops | ||
| Losses (cont.) | |||
| Tel-17p13.2 | 0 | 5552067 | CTNS, PAFAH1B1, FAM57A, ATCD1, LOC284009, C17orf87, RPAIN, MST075, LOC100130876, BHLHA9, GLOD4, ANKFY1, GGT6, MYO1C, VMO1, PLD2, P2RX1, DPH1, LOC100131873, ABR, TRPV3, SNORD91A, VPS53, OR1A2, SPATA22, LOC100130241, TMEM93, GPR172B, RILP, OR1E3, CHRNE, OR3A2, KIAA0664, ZZEF1, ELP2P, CRK, OR1D5, LOC100128142, MYBBP1A, GARNL4, PITPNA, LOC100130311, SNORD91B, YWHAE, RPL21P125, SGSM2, OR1G1, TIMM22, TUSC5, NUP88, LOC100129673, DERL2, MED11, ZFP3, LOC100128013, MIS12, RABEP1, LOC642746, CAMKK1, MNT, GLTPD2, LOC653166, OR1D2, UBE2G1, FAM101B, SERPINF1, ZNF232, ZMYND15, OR1P1, LOC100129974, OR3A3, LOC727845, OR1A1, LOC440396, OR1E2, ASPA, LOC727796, INCA1, C17orf97, WDR81, hCG_1776980, ATP2A3, GSG2, USP6, PFN1, TSR1, GP1BA, LOC100130950, C1QBP, HIC1, OR1E1, C17orf85, GEMIN4, SERPINF2, SRR, PSMB6, METT10D, PELP1, SHPK, TRPV1, RNF167, C17orf91, CYB5D2, CAMTA2, SPNS2, NXN, SPAG7, OR1D3P, INPP5K, ALOX15, MRPL14P1, LOC124974, TAX1BP3, RPA1, OR3A4, P2RX5, ENO3, PRPF8, RPH3AL, ZNF594, RPS4P17, KIF1C, SMTNL2, ARRB2, SLC25A11, SPNS3, TLCD2, OR1AC1P, MINK1, DOC2BL, LOC100132822, RTN4RL1, CXCL16, RYKP, OR1R1P, ITGAE, OVCA2, SLC43A2, SCARF1, OR3A1, SMYD4, SMG6, NLRP1, DHX33, RNMTL1, TM4SF5 |
| 17p13.1 | 7085898 | 10165595 | TP53, TMEM102, ALOX15B, EIF4A1, PFAS, RANGRF, C17orf44, CNTROB, SNORA48, MFSD6L, RPL29P2, NEURL4, KCNAB3, SNORA67, LOC100128288, VAMP2, LOC284023, PIK3R6, PER1, ARHGEF15, GPS2, KRBA2, C17orf74, ACAP1, MYH13, PIK3R5, ZBTB4, DULLARD, ALOX12B, GABARAP, SAT2, LOC728867, RCVRN, CYB5D1, HES7, WDR16, LOC100129677, C17orf61, POLR2A, LOC390760, KCTD11, C17orf59, SNORD10, NLGN2, RPS26P53, LOC100130387, TMEM88, STX8, RPL26, SENP3, SPEM1, CHD3, YBX2, LOC100129978, SLC25A35, TMEM107, GLP2R, TRAPPC1, CLDN7, AMAC1L3, SLC2A4, LOC644070, C17orf81, TNFSF13, TNFSF12, EIF5A, TNK1, DHRS7C, NDEL1, SOX15, C17orf68, MYH10, EFNB3, RNF222, GAS7, ODF4, MPDU1, GUCY2D, ALOXE3, RPS27AP1, ATP1B2, NTN1, SHBG, CCDC42, SCARNA21, PLSCR3, LSMD1, CHRNB1, FXR2, DNAH2, SPDYE4, LOC100128281, AURKB, WRAP53, FGF11, CD68, TNFSF12-TNFSF13, USP43, TMEM95, JMJD3 |
| CGH: Comparative genomic hybridization. | |||