Improvement of histopathological classification of adrenal gland tumors by genetic differentiation
Torsten Gruschwitz . Jan Breza . Heiko Wunderlich . Kerstin Junker
Received: 21 December 2009 / Accepted: 13 March 2010 / Published online: 3 April 2010 @ Springer-Verlag 2010
Abstract
Purpose There are often problems in differentiating between benign and malignant adrenal gland tumors by imaging and histopathology. Fine-needle biopsy is possible but not used owing to problems in histopathological differ- entiation. On account of considerable differences in the therapy and aftercare of benign and malignant adrenal tumors, correct classification of tumor type is of greatest importance. The purpose of this study was to define specific genetic alterations differentiating between adenomas and carcinomas.
Methods DNA was isolated from tumor areas in paraffin sections and amplified by a modified protocol for DOP- PCR. After labeling of tumor-DNA and normal DNA with biotin-dUTP and digoxigenin-dUTP, respectively, compar- ative genomic hybridization (CGH) was carried out accord- ing to standard protocols. Retrospectively, 26 (16 adenomas and 10 carcinomas) tumors of the adrenal cortex were analyzed.
Results Genetic alterations were found in 5/16 adenomas (31.25%) and in all adrenocortical carcinomas. The mean number of genetic changes per tumor was 8.7 (range 6-12) in carcinomas. The benign cortical tumors present 1.6 changes (range 0-3) per tumor. Only a moderate correlation between number of alterations and size of tumor was seen.
Furthermore, specific chromosomal alterations of carcino- mas were identified.
Conclusions Genetic evaluation facilitates differentiation between adrenal gland tumors. Genetic tests should be used in routine diagnostics of adrenal specimens. Potentially, fine-needle biopsy can be established as standard diagnos- tics of adrenal tumors with unknown genesis.
Keywords Adrenal tumor . Tumor-type differentiation · Genetic evaluation · Fine-needle biopsy
Introduction
Clinically unapparent adrenal masses are incidentally detected after imaging studies conducted for reasons other than the evaluation of the adrenal glands. They have fre- quently been referred to as adrenal incidentalomas. On the basis of 30 autopsy studies, adrenal masses are one of the most prevalent human tumors. The prevalence of adrenal incidentalomas varies in different studies, it approaches 3% in middle age, and rises to as much as 8% in the elderly [1, 2]. Consequently, as our population ages, the management of clinically unapparent masses is becoming an increasing aspect of health care. Moreover, advances in imaging and an increasing availability in imaging technology may reveal a higher incidence, making the management of incidentalomas a challenge for modern medicine.
Differentiation between malignant and benign masses is an essential part of diagnosis. In contrast to the high inci- dence of benign adrenal tumors, adrenocortical carcinoma (ACC) is a rare malignancy with an estimated incidence of up to two cases per million people [3]. Most of the patients present with an advanced stage of the disease and have a median survival time of less than 12 months [4]. Prognosis
T. Gruschwitz () · H. Wunderlich . K. Junker Department of Urology, University Medical Center, Lessingstraße 1, 07743 Jena, Germany
e-mail: torsten.gruschwitz@med.uni-jena.de
J. Breza Department of Urology, University of Bratislava, Bratislava, Slovakia
of these highly malignant tumors is poor owing to late confirmation of correct diagnosis. One reason for that is the restricted capability of imaging systems to recognize carcinomas at early stages. Benign or malignant adrenal tumors presenting classical imaging characteristics of the respective tumor can be detected by CT, MRT or PET with a high sensitivity and specificity [5]. However, tumors like granulomatous inflammation, low-fat adenomas, pheo- chromocytomas and carcinomas at early stages are fre- quently not differentiable. In addition, authors reported that the size of nearly all adrenal gland tumors is underes- timated using CT [6]. Owing to the fact that indication for surgery of an incidentaloma is primarily on the basis of the tumor size, devastating consequences for the patient are possible.
For differentiation between benign and malignant tumors, pathologists commonly use classification systems on the basis of histopathological characteristics (Weiss et al., Hough et al., van Slooten et al. [7-10]). The mean- ingfulness of these classification systems is limited. Fur- thermore, immunohistochemistry too provides restricted information on the distinction between benign and malig- nant tumors [11].
Various criteria, immunological and cytoskeletal mark- ers, DNA ploidy and cell phase markers have been pro- posed for the differentiation of adrenocortical and medullar masses, but have yielded inconsistent results so far [12]. Stojadnovic et al. [13] examined the use of molecular markers to predict prognosis, unfortunately no molecular profile with prognostic value was identified. At present, no biomarkers exist for routine diagnostics of adrenal masses.
Fine-needle biopsy represents an additional diagnostic method. According to the NIH Conference 2002, it is indicated in patients suffering from a malignancy with no other signs of metastasis and a heterogeneous adrenal mass with high attenuation value (>20 HU) in CT [14]. Owing to the radiological difficulties in the diagnostics of the smallest adrenal tumors, some authors recommend fine-needle biopsy in all adrenal tumors of less than 2 cm in diameter [15].
The aim of this study was to characterize adrenal tumors regarding their genetic alterations in order to differentiate between benign and malignant tumors. With that, conclu- sions could be drawn as to the pathogenesis of these tumors. On the basis of these data, genetic tests can be used
as an additional tool in routine diagnostics of adrenal tumors.
Materials and methods
Tissue samples were available from 26 sporadic adrenal cortex tumors. The tissue samples were obtained at the time of adrenalectomy by surgical excision. All tumors were his- tologically proven by two independent pathologists. There were 16 benign tumors of the cortex, and 10 tumors were classified as adrenocortical carcinomas. The patients’ data are summarized in Table 1.
Unfortunately, no follow-up data were collected. Owing to this limitation the results of histological analysis could not be confirmed by clinical follow-up.
CGH
In order to obtain sufficient amounts of DNA for CGH analysis, tumor-DNA was amplified according to a modi- fied protocol for DOP-PCR [16]. This protocol employs Sequenase during the first 8 cycles of non-specific PCR, followed by 30 additional cycles under specific conditions using Taq polymerase (Stoffel fragment). Labeling of tumor-DNA and normal-DNA was achieved by 20 PCR cycles using Biotin-16dUTP and Digoxigenin-11dUTP, respectively. One µg each of tumor-DNA and normal-DNA was hybridized with 50 µg Cot-1-DNA on normal metapha- ses at 37°℃ for 48 h. Detection of fluorescence signals was carried out with Avidin-FITC (tumor-DNA) and anti- digoxigenin-rhodamine (normal DNA). DAPI-Antifade was used for chromosome counterstaining. Fifteen meta- phases were analyzed in each case using an Axioplan microscope (Zeiss, Germany) and a computer system from Metasystems, Germany. Chromosomal alterations can be detected as shifts of the profile to the red borderline (loss of chromosomal region in the tumor-DNA) or to the green borderline (gain of chromosomal region in the tumor- DNA).
Statistical analysis
To compare the number of alterations between the tumor groups we used the Wilcoxon-Mann-Whitney analysis.
| Histology | Number of tumors | Sex w/m | Age of patient (mean/range) | Size of tumor (mean/range) |
|---|---|---|---|---|
| Adenoma | 16 | 7/9 | 53.2/26-65 | 34.8/16-65 |
| Carcinoma | 10 | 7/3 | 47.7/19-64 | 94.5/42-150 |
The Kendall-Tau-b test was used to examine the relation- ship between number of alterations and tumor size.
Results
Correlation between number of genetic alterations and histological classification
Genetic alterations were found in 5 out of 16 adenomas (31.25%) and in all adrenocortical carcinomas. The mean number of genetic changes per tumor was 8.7 (range 6-12) in carcinomas. The benign cortical tumors present 1.6 changes (range 0-3) per tumor. The Wilcoxon-Mann-Whitney analy- sis showed a highly significant difference between benign and malignant tumors concerning the number of observed CGH changes (P <0.001).
Correlation between number of genetic alterations and tumor size
The mean size of malignant tumors was 94.5 mm (range 42-150) as compared to the benign tumor mean size of 34.8 mm (range 16-65). When separating benign and malignant tumors, Kendall-Tau-b test showed middle and low correlation coefficients for the number of genetic changes in relation to tumor size (benign 0.493, low signifi- cance; malignant 0.101, no significance).
Specific chromosomal alterations of adrenal gland tumors
Genetic losses were detected much more often than gains. The distribution was nearly 3:1 in carcinomas and 2:1 in benign tumors. In adrenal carcinomas, losses were detected frequently on chromosomes 1p (50%), 1q (40%), 3/3p (70%), 4/4q (30%), 5/5q (30%), 6p (40%), 6q (60%), 7q (30%), 9p (30%), 9q (30%), 10q (30%), 11/11p (30%), 13q (40%), 14q (30%), 15q (30%) and 18q (70%). Gains were frequently seen on chromosomes 4q (20%), 5p (20%), 5q (20%), 7/7q (30%), 9p (20%) and 16 (20%). In contrast, specific alterations with high frequencies did not occur in benign tumors (Table 2).
Comparison of specific chromosomal aberrations between benign and malignant tumors of the adrenal cortex
Seventy-two percent of aberrations found in benign tumors also occurred in malignant tumors. In general these altera- tions were frequently detected in malignant tumors; for example loss of 3p was seen in 70% of the carcinomas. On the other side, losses of genetic material on chromosomal regions 1p (50%), 1q (40%), 6p (40%), 6q (60%), 9q (30%), 10q (30%), 13q (40%), 15q (30%), 18q (70%) and
gain on chromosome arm 5/5p (20%) were seen exclusively and at a high percentage in carcinomas.
Correlation between hormone activity and genetic alterations
Only 2 of 16 adenomas were hormonally active (aldoste- ronism); in both tumors no genetic alterations were detected. Of ten carcinomas, three presented hormonal activity, two Cushing syndromes and one Conn’s syn- drome. The number of genetic alterations detected in these tumors was 7, 8 and 8, respectively. Only losses of genetic material were seen. The other tumors were all true inciden- talomas.
Discussion
It is well known that ACC is a rare and aggressive malig- nancy; the only chance of cure is diagnosis at an early stage of the disease. Depending on series reviewed, approxi- mately 40% of patients present with an advanced stage of the disease (stage IV) not amenable for surgical resection with a median survival time of less than 12 months [17]. The reason for late diagnosis and delayed beginning of ther- apy is the lack of clinical or biochemical markers reliably defining malignancy. Serum tumor marker measurements and endocrinologic tests are not useful in differentiation between benign and malignant adrenal tumors. Manage- ment strategies exclusively based on size criteria are neither sensitive nor specific. Although imaging techniques have improved over the past years, radiologists still have problems in discovering ACC at an early stage.
The purpose of this study was to analyze whether benign and malignant tumors of the adrenal gland show different genetic patterns.
A high correlation was provable between tumor type and occurrence of genetic alterations. While approximately only a third of adenomas (5/16) showed genetic alterations, in all carcinomas chromosomal changes occurred. These results are comparable with other CGH-supported examina- tions of sporadic adrenal cortex tumors. Zhao et al. found genetic alterations in 15/23 adenomas and in 12/12 carcino- mas; Sidhu et al. [18, 19] evaluated genetic alterations in 11/18 adenomas and in 13/13 carcinomas.
Furthermore, the number of chromosomes altered per tumor (adenomas 1.6; carcinomas 8.7 chromosomes per tumor, respectively) and also the range of number of altera- tions between both tumor types (0-3 vs. 6-12 alterations per tumor, respectively) is highly significantly different.
An only restricted correlation was seen between number of alterations and size of tumor. Although smaller tumors presented fewer alterations than the larger ones, an increase
| Histology | Size (mm) | Gain | Loss | Number of altered chromosomes |
|---|---|---|---|---|
| Adenoma | 23 | – | – | 0 |
| Adenoma | 28 | 9p | 21q | 2 |
| Adenoma | 48 | 9q | – | 1 |
| Adenoma | 16 | – | – | 0 |
| Adenoma | 20 | – | – | 0 |
| Adenoma | 23 | – | – | 0 |
| Adenoma | 27 | – | – | 0 |
| Adenoma | 30 | – | – | 0 |
| Adenoma | 32 | – | – | 0 |
| Adenoma | 35 | – | – | 0 |
| Adenoma | 35 | – | – | 0 |
| Adenoma | 36 | – | – | 0 |
| Adenoma | 36 | – | – | 0 |
| Adenoma | 39 | – | 3p14pter | 1 |
| Adenoma | 63 | – | 5q31qter, 11p, 14q21qter | 3 |
| Adenoma | 65 | 9p | – | 1 |
| Carcinoma | n.k. | – | 1p, 2q, 3p, 4, 6q, 13q, 18q | 7 |
| Carcinoma | 42 | – | 1, 4, 5, 9, 10q11.2q23q, 13q, 16, 18q | 8 |
| Carcinoma | 50 | 7,16 | 1, 3, 6, 10, 11, 18q | 8 |
| Carcinoma | 70 | – | 1p, 3, 5q23qter, 6q, 7p, 11/11p, 13q, 14q23qter, 15q23qter, 17, 18q | 11 |
| Carcinoma | 80 | – | 1, 2p, 3, 6, 8, 9, 11, 18q | 8 |
| Carcinoma | 100 | 4q, 5, 7, 14q, 16 | 6p, 9q, 22qter | 8 |
| Carcinoma | 100 | – | 3p, 6q, 7q, 9p, 13q, 14q, 17p, 18q | 8 |
| Carcinoma | 120 | 1p33p35, 6q, 7q, 10p, 13q31qter, 18p | 2q, 3p, 4q, 5q, 10q, 15q, Xp | 12 |
| Carcinoma | 140 | 3q, 49, 59, 9p | 7,8 | 6 |
| Carcinoma | 150 | 5pterq23, 9p, 12p | 1q31qter, 3p12.pter, 7q, 10p, 12q21qter, 14q23qter, 15q21qter, 16q, 18q | 11 |
in tumor size was not always accompanied by an increase of the number of alterations. When separating benign and malignant tumors, Kendall-Tau-b correlation analysis showed middle and low correlation coefficients (adenoma 0,493; carcinoma 0,101). Therefore, it is more likely that the number of alteration correlates much more with tumor type than with tumor size. Because therapy decision is mainly based on tumor size, this is an important result regarding indication for operation of adrenal gland tumors.
The recent NIH consensus conference did not make a definitive recommendation about management of non-func- tioning incidentalomas of 4-6 cm in size, therefore observ- ing surgery is possible [14]. The assumption that malignant tumors are huge in size perhaps misleads physicians to a waiting strategy. It would be hazardous to deprive patients with small carcinomas and good prognosis of adequate treatment by surgery. Regarding adenomas and carcinomas examined in our study between 4 and 6 cm in size, tumors
were classifiable on the basis of the number of alterations. Two carcinomas (4.2 and 5.0 cm) presented eight chromosomes with genetic alterations each; the four adenomas in size rang- ing 3.9-6.5 cm presented up to three affected chromosomes (1, 1, 1 and 3 alterations per tumor, respectively). These results imply that a genetic examination can precisely differentiate between tumor types independently of tumor size.
The existence of specific chromosomal aberrations for each tumor type would endorse the informative value of genetic tests. Especially losses of genetic material were detected in high percentages of the carcinomas, for example loss on chromosome 18q was found in 70%. These losses of genetic material suggest that tumor suppressor genes, which are located in these areas, play an important role in the tumorigenesis of adrenal cortex carcinomas.
Gains of chromosomal regions were less frequently seen. In 30% of carcinomas gain on 7/7q occurred. Furthermore, we can confirm results reported by Kjellman et al., Zhao et al.,
Healthy adrenal gland tissue
Adenoma
Carcinoma
Losses on
3/3p, 5/5q, 11/11p, 14q Gain on 9p
Losses on 1p, 1q, 6p, 6q, 9q, 10q, 11q, 13q, 15q, 17p, 18q Gain on 5/5p
Dohna et al. and Sidhu et al. regarding gains on chromo- somes four and five, for which an oncogenic function may be assumed [18-21].
No correlation was found between hormonal activity and number or type of genetic alterations for adenomas as well as for carcinomas.
Specific chromosomal aberrations were seen rarely in adenomas. Though, 72% of the evaluated alterations of adenomas were also seen in carcinomas. These results sup- port the thesis of a stepwise progression from adenoma to carcinoma, which is reported for other tumor diseases. However, there are only few clinical reports which indicate such a sequence [22]. Progression models were suggested in previous CGH-supported studies by Kjellman et al. and Sidhu et al. [19, 23].
A limitation is the absence of data relating to the clinical follow-up of the patients. Because genetic analysis found clear genetic differences between the tumor types, espe- cially regarding the number of alterations independent to tumor size, we are convinced that genetic analysis will sup- port and tighten statements of tumor type by pathologists.
Image-guided biopsy seems to be a safe and sensitive procedure. But, because of problems in histopathological examination biopsy is still not recommended in standard diagnostics. In our opinion the results of this genetic study, which evaluated a marked and statistically proven differ- ence with respect to the number of alterations between the both tumor types and detected specific genetic alterations, encourage the use of fine-needle biopsy in selected patients. Genetic tests could improve the validity of histopatholo- gical diagnosis by the pathologist and facilitate the decision of therapy for the clinician/physician.
The occurrence of a high number of genetic alterations in our study suggests that tumorigenesis of adrenal carci- noma is very complex and tumor progression as well as malignant transformation is accompanied by an accumulation
of multiple genetic aberrations. According to our results we suggest a tumor progression model: gain on chromosome 9p and loss of genetic material on chromosome/-arms 3/3p, 5/5q, 11/11p and 14q are detectable in adenomas and, par- tially at high percentages, in carcinomas. These alterations could already have appeared at an early stage of tumor pro- gression. Losses on chromosomal regions 1p, 1q, 6p, 6q, 9q, 10q, 11q, 13q, 15q, 17p, 18q and gain on chromosome 5/5p were seen exclusively and at a high percentage in car- cinomas. These alterations have to be judged as factors of malignant transformation (Fig. 1).
Conclusions
The genetic examination of sporadic adrenal gland tumors could determine a high degree of genetic changes with clear differences between benign and malignant tumors. A dis- tinction between these both tumor types seems possible according to the number of genetic changes. Tumor size is insufficiently suitable as a diagnostic marker. Conse- quently, also small tumors must be examined intensively.
We can assume that diagnostically utilizable molecular genetic markers are useful in addition to clinical and histo- pathological characteristics.
Genetic tests should be used in routine diagnostics of specimen of the adrenals. Potentially fine-needle biopsy will be established as standard diagnostics of adrenal tumors with unknown genesis.
Conflict of interest statement None.
References
1. Kloos RT, Gross MD, Francis IR, Korobkin M, Shapiro B (1995) Incidentally discovered adrenal masses. Endocrine rev 16:460-484
2. Barzon L, Sonino N, Fallo F, Palù G, Boscaro M (2003) 315 Prev- alence and natural history of adrenal incidentalomas. Eur J Endo- crinol 149(4):273-285
3. Third National Cancer Survey: Incidence Data, National Cancer Institute Monograph (1975) United States Department of Health, Education and Welfare, Public Health Service, National Institute of Health, National Cancer Institute
4. Ng L, Libertino JM (2003) Adrenocortical carcinoma: diagnosis, evaluation and treatment. J Urol 169(1):5-11
5. Blake MA, Holakere NS, Boland GW (2008) Imaging tech- niques for adrenal lesion characterization. Radiol Clin North Am 46(1):65-78
6. Cerfolio RJ, Vaughan ED Jr, Brennan TG Jr, Hirvela ER (1993) Accuracy of computed tomography in predicting adrenal tumor size. Surg Gynecol Obstet 176(4):307-309
7. Weiss LM (1984) Comparative histologic study of 43 metastasiz- ing and nonmetastasizing adrenocortical tumors. Am J Surg Pathol 8(3):163-169
8. Weiss LM, Medeiros LJ, Vickery AL Jr (1989) Pathologic features of prognostic significance in adrenocortical carcinoma. Am J Surg Pathol 13(3):202-206
9. Hough AJ, Hollifield JW, Page DL, Hartmann WH (1979) Prog- nostic factors in adrenal cortical tumors. A mathematical analysis of clinical and morphologic data. Am J Clin Pathol 72(3):390-399
10. Van Slooten H, Schaberg A, Smeenk D, Moolenaar AJ (1985) Morphologic characteristics of benign and malignant adrenocorti- cal tumors. Cancer 55(4):766-773
11. Saeger W (2000) Histopathological classification of adrenal tumours. Eur J Clin Invest 30(Suppl. 3):58-62
12. Mansmann G, Lau J, Balk E, Rothberg M, Miyachi Y, Bornstein SR (2004) The clinically inapparent adrenal mass: update in diag- nosis and management. Endocr Rev 25(2):309-340
13. Stojadinovic A, Ghossein RA, Hoos A et al (2002) Adrenocortical carcinoma: clinical, morphologic, and molecular characterization. J Clin Oncol 20(4):941-950
14. NIH (2002) NIH state-of-the-science statement on management of the clinically inapparent adrenal mass (“incidentaloma”). NIH Consens State Sci Statements 19(2):1-25
15. Lumachi F, Borsato S, Tregnaghi A et al (2003) CT-scan, MRI and image-guided FNA cytology of incidental adrenal masses. Eur J Surg Oncol 29(8):689-692
16. Telenius H, Carter NP, Bebb CE, Nordenskjold M, Ponder BA, Tunnacliffe A (1992) Degenerate oligonucleotide-primed PCR: general amplification of target DNA by a single degenerate primer. Genomics 13(3):718-725
17. Dackiw AP, Lee JE, Gagel RF, Evans DB (2001) Adrenal cortical carcinoma. World J Surg 25(7):914-926
18. Zhao J, Speel EJ, Muletta-Feurer S et al (1999) Analysis of geno- mic alterations in sporadic adrenocortical lesions. Gain of chromo- some 17 is an early event in adrenocortical tumorigenesis. Am J Pathol 155(4):1039-1045
19. Sidhu S, Marsh DJ, Theodosopoulos G et al (2002) Comparative genomic hybridization analysis of adrenocortical tumors. J Clin Endocrinol Metab 87(7):3467-3474
20. Kjellman M, Kallioniemi OP, Karhu R et al (1996) Genetic aber- rations in adrenocortical tumors detected using comparative geno- mic hybridization correlate with tumor size and malignancy. Cancer Res 56(18):4219-4223
21. Dohna M, Reincke M, Mincheva A, Allolio B, Solinas-Toldo S, Lichter P (2000) Adrenocortical carcinoma is characterized by a high frequency of chromosomal gains and high-level amplifica- tions. Genes Chromosomes Cancer 28(2):145-152
22. Bernard MH, Sidhu S, Berger N et al (2003) A case report in favor of a multistep adrenocortical tumorigenesis. J Clin Endocrinol Metab 88(3):998-1001
23. Kjellman M, Larsson C, Backdahl M (2001) Genetic background of adrenocortical tumor development. World J Surg 25(7):948- 956