Squamous cell carcinoma-related oncogene is highly expressed in developing, normal, and adenomatous adrenal tissue but not in aggressive adrenocortical carcinomas
Inderpal S. Sarkaria, MD, Alexander Stojadinovic, MD, Simon G. Talbot, MD, Axel Hoos, MD, Maria E. Dudas, Murray F. Brennan, MD, Ronald A. Ghossein, MD, and Bhuvanesh Singh, MD, New York, NY, and Washington, DC
Background. Our previous work has demonstrated squamous cell carcinoma-related oncogene (SCCRO) expression in adult murine adrenocortical tissue. The aim of this study was to assess patterns of SCCRO expression in the embryonic murine adrenal gland, and in normal and neoplastic human adrenocortical tissues in order to determine its role as a marker of differentiation in adrenocortical development and neoplastic progression.
Methods. Murine embryos were procured at developmental stages E8 to E18. A tissue microarray was constructed containing 38 normal, 39 adenomatous, and 87 carcinomatous human adrenocortical specimens. Immunohistochemistry for SCCRO was performed and its expression was graded in suitable tissues.
Results. SCCRO expression was detected in the murine adrenal cortex as early as E15 and persisted into the postnatal period. High-level SCCRO expression was identified in 94% of normal (32/34) and adenomatous (29/31) adrenocortical specimens but in only 63% (45/72) of adrenocortical carcinoma (ACC) specimens (P =. 001). Loss of SCCRO expression in primary ACC (13/34 (34%)) correlated with advanced stage (P =. 06), presence of MI disease at presentation (P =. 03), and worse overall survival (P =. 006).
Conclusions. SCCRO appears to be a marker of adrenocortical differentiation in both murine and human systems. SCCRO expression may be useful in distinguishing adrenocortical adenomas from ACC. Moreover, loss of SCCRO expression in primary ACC is associated with worse outcome and may be a marker of progressive dedifferentiation in these tumors. (Surgery 2004;136:1122-8.)
From the Laboratory of Epithelial Cancer Biology, Departments of Surgery and Pathology, Memorial Sloan- Kettering Cancer Center, New York, NY, and Walter Reed Army Medical Center, Washington, DC
SQUAMOUS CELL CARCINOMA-RELATED ONCOGENE (SCCRO) is a novel gene initially identified by positional cloning of a recurrent amplification at 3q26-27 in squamous cell carcinomas.1-4 Our pre- vious work has shown that SCCRO acts as a tran- scription factor that functions in hedgehog
Presented at the 25th Annual Meeting of the American Association of Endocrine Surgeons, Charlottesville, Virginia, April 4-6, 2004.
Reprint requests: Bhuvanesh Singh, MD, Memorial Sloan- Kettering Cancer Center Head and Neck Surgery, 1275 York Ave, New York, NY 10021.
0039-6060/$ - see front matter
@ 2004 Elsevier Inc. All rights reserved.
doi:10.1016/j.surg.2004.06.041
signaling, one of the critical pathways involved in mammalian development, including that of the adrenal cortex. Furthermore, multi-tissue expres- sion analyses reveal SCCRO expression in the normal adult murine adrenal cortex. These obser- vations suggest that SCCRO may be part of a grow- ing list of “onco-developmental” genes, involved in carcinogenesis in the dysregulated state, but is also important for normal cellular functioning in the regulated state. Based on the associations between SCCRO, hedgehog signaling, and adrenocortical development, we aimed to determine if SCCRO plays a role in adrenal development and adreno- cortical malignant progression. To address these questions, we performed immunohistochemical analyses to assess SCCRO expression in mouse
embryos and tissue microarrays containing normal and neoplastic adrenocortical specimens.
METHODS
Murine embryo procurement and processing. Murine embroys were procured from NIH Swiss mice at development stages E8 to E18 and postnatal day 1, fixed in 4% paraformaldehyde, embedded in paraffin, and serially sectioned (Novagen, Madison, Wis). Representative slides were selected from each developmental stage and corresponding sections submitted to either hematoxylin and eosin to con- firm content or immunohistochemical analysis to assess SCCRO expression.
Patient selection and tumor acquisition. Patients with adrenocortical adenoma or adrenocortical carcinoma (ACC) resected at Memorial Sloan- Kettering Cancer Center between January 1982 to November 1999 for whom paraffin-embedded tissue specimens were available were identified from the Endocrine Tumor Database. Corres- ponding paraffin-embedded tissues from primary, metastatic, and/or recurrent neoplasms were col- lected, as available. In addition, paraffin-embed- ded histologically normal adrenocortical tissue was also identified, when possible, from resected adre- nal specimens. Demographic, tumor, and outcome data were collected from fields prospectively main- tained on the Endocrine Tumor Database.
Pathologic review. Hematoxylin and eosin- stained sections from each patient were reviewed to identify and localize the tissue of interest by the reference pathologist (R.A.G.). The criteria for diagnosis of ACC, based on the 11-tiered system by Weiss et al, are described in detail elsewhere.5-7 Briefly, neoplasm size and weight; nuclear grade (determined based on the grading system for renal cell carcinomas described by Fuhrman et al8); number of mitotic figures per high power field; presence of atypical mitoses; clarity of neoplastic cytoplasm (graded as ≤25% or >25%); neoplasm architecture (classified as diffuse if sheets of cells without a characteristic growth pattern composed >33% of the neoplastic mass); presence of necrosis, evidence of venous, sinusoidal, or capsular invasion; and microscopic margins status were used to estab- lish the diagnosis of ACC.
Tissue microarray construction. Tissue core bi- opsies 0.6 mm in diameter were procured from representative areas of neoplastic and normal tissues, and arrayed in triplicate onto a master paraffin block with the use of a manual tissue arrayer (Beecher Instruments, Silver Springs, Md), as previously described.5,9 Five-micron sections
Normal Adrenal
Adrenocortical Carcinoma
Positive
Negative
Positive
were cut, stained with hematoxylin and eosin, and examined to confirm tissue core content. Serial sections were used for immunohistochemi- cal analysis.
Immunohistochemistry. Slides were deparaffi- nized, rehydrated with graded alcohol treatment, and processed with a standard avidin-biotin im- munoperoxidase staining procedure. Quenching was performed with 1% hydrogen peroxide in phosphate-buffered saline (PBS) for 15 minutes. Antigen retrieval was performed by microwave treatment in 0.01 mol/L citrate buffer, pH 6.0, for 15 minutes. Blocking was performed first with avidin-biotin, and then with 10% goat serum in 2% bovine serum albumin-phosphate-buffered saline in a humidified chamber for 30 minutes (Avidin-Biotin Blocking KIT; Vector Laboratories, Inc, Burlingame, Calif). Slides were incubated at 4℃ overnight in a humidity chamber in appropri- ately diluted primary antibody. A novel rabbit polyclonal antihuman SCCRO antibody developed in our laboratory was used at a dilution of 1:5000 (0.2 µg/mL). The use of this antibody for immuno- histochemistry has been optimized and validated previously. After washing, samples were incubated with biotinylated antirabbit immunoglobulins (1:1000 dilution) followed by avidin-biotin perox- idase complexes at a dilution of 1:25 (Vectastain; Vector Laboratories, Inc, Burlingame, Calif). Dia- minobenzidine was used as the chromogen and hematoxylin as the nuclear counterstain. Cell lines known to express the antigen under study were used as positive controls.
Scoring of SCCRO expression. The pattern of SCCRO expression in mouse embryos was assessed by the reference pathologist (R.A.G.) In tissue microarray specimens with 2 or more cores suitable for analysis, the intensity of cytoplasmic SCCRO
Embryonic Day 15
Embryonic Day 17
₹
F
X40
X
1
X100
AM
AM
AC
AC
| Specimen type | Number analyzed/total on array (%) | SCCRO positive/number analyzed (%) | P value* |
|---|---|---|---|
| Normal vs neoplasm | .001+ | ||
| Normal | 34/38 (89%) | 32/34 (94%) | |
| Adenoma | 31/33 (94%) | 29/31 (94%) | |
| ACC | 73/87 (84%) | 45/73 (62%) | |
| ACC type | NS | ||
| Primary | 34/38 (89%) | 21/34 (62%) | |
| Metastasis | 28/36 (78%) | 20/28 (71%) | |
| Recurrence | 11/13 (85%) | 4/11 (36%) |
SCCRO, Squamous cell carcinoma-related oncogene; ACC, adrenocortical carcinoma.
*P values shown are for correlations between the neoplastic characteristic and SCCRO expression phenotype.
+P value represents comparison between both normal vs ACC and adenoma vs ACC.
expression was scored with a grade of 0 (absent), 1 (low), 2 (moderate), or 3 (high) by two investiga- tors (R.A.G. and I.S.S.) blinded to the clinical data. Only cores with moderate to high levels of expres- sion (IHC grades 2 or 3) were considered to have a positive SCCRO phenotype (Fig 1).
Statistical analysis. Correlations between SCCRO expression and clinicopathologic predictors were determined with the chi-square or Fisher exact test. Survival curves were generated by the Kaplan-Meier method and compared with the use of the log-rank
test. Cox regression analysis was used to test multi- variate correlations. All analyses were performed with SPSS statistical software (SPSS, Inc, Chicago, Ill).
RESULTS
SCCRO expression in murine embryogenesis. SCCRO expression was seen early in murine de- velopment, starting in extramedullary hematopoi- etic rests. Neuroectodermal and epithelial
| Patient/tumor characteristic | Number (%)* | SCCRO positive/number analyzed (%) | P value* |
|---|---|---|---|
| Sex | NS+ | ||
| Male | 14 (41%) | 8/14 (57%) | |
| Female | 20 (59%) | 13/20 (65%) | |
| Stage | .06 | ||
| I+II | 12 (35%) | 12/12 (100%) | |
| III+IV | 22 (65%) | 11/22 (50%) | |
| Distant metastases at presentation | .03 | ||
| Yes | 15 (43%) | 6/15 (40%) | |
| No | 19 (57%) | 15 (79%) | |
| Size of neoplasm t | NS | ||
| ≤5cm | 3 (9%) | 2/3 (67%) | |
| 6-10 cm | 5 (15%) | 3/5 (60%) | |
| 11-20 cm | 21 (62%) | 13 (38%) | |
| >20 cm | 3 (9%) | 3/3 (100%) | |
| Weight of neoplasm t | NS | ||
| ≤100 g | 2 (6%) | 1/2 (50%) | |
| 101-250 g | 2 (6%) | 1/2 (50%) | |
| 251-1000 g | 15 (43%) | 8/15 (53%) | |
| >1000 g | 6 (18%) | 5/6 (83%) | |
| Venous invasion t | NS | ||
| Yes | 18 (53%) | 9/18 (50%) | |
| No | 15 (44%) | 12/15 (80%) | |
| Sinusoidal invasion | NS | ||
| Yes | 31 (91%) | 19/31 (61%) | |
| No | 3 (9%) | 2/3 (67%) | |
| Capsular invasion t | NS | ||
| Yes | 16 (47%) | 8/16 (50%) | |
| No | 17 (50%) | 12/17 (71%) | |
| Adjacent organ invasion | NS | ||
| Yes | 3 (9%) | 1/3 (33%) | |
| No | 31 (31%) | 20/31 (65%) | |
| Necrosis present in neoplasm | NS | ||
| Yes | 29 (85%) | 16/29 (55%) | |
| No | 5 (15%) | 5/5 (100%) | |
| Mitotic rate/50 high power fields | NS | ||
| ≤5 | 2 (6%) | 0/2 (0%) | |
| 6-20 | 8 (24%) | 6/8 (75%) | |
| 21-50 | 9 (26%) | 5/9 (56%) | |
| >50 | 15 (44%) | 10/15 (67%) | |
| Atypical mitotic figures | NS | ||
| Yes | 21 (62%) | 11/21 (52%) | |
| No | 13 (38%) | 10/13 (77%) | |
| Nuclear grade | NS | ||
| 1 | 0 (0%) | 0/0 (0%) | |
| 2 | 0 (0%) | 0/0 (0%) | |
| 3 | 11 (32%) | 6/11 (55%) | |
| 4 | 23 (68%) | 15/23 (65%) | |
| Diffuse architecture | NS | ||
| Yes | 27 (68%) | 17/27 (63%) | |
| No | 7 (21%) | 4/7 (57%) | |
| Cytoplasm | NS | ||
| 0%-25% clear | 24 (71%) | 17/24 (71%) | |
| 26%-100% clear | 10 (29%) | 4/10 (40%) | |
| Positive microscopic margint | NS | ||
| Yes | 7 (21%) | 6/7 (86%) | |
| No | 20 (59%) | 13/20 (65%) |
SCCRO, Squamous cell carcinoma-related oncogene.
*P values shown are for correlations between the given patient/tumor characteristic and SCCRO expression phenotype.
+Characteristics with numbers not totaling 34 (100%) are explained by unavailable or missing data points for the given category.
1.2
P = . 006
1.0
Cumulative Survival
#
.8
.6
.4
SCCRO negative, n = 13
SCCRO positive, n = 21
.2
0.0
0
10
20
Overall Survival (years)
| Covariates | RR | 95% CI | P value |
|---|---|---|---|
| SCCRO phenotype and staget | |||
| SCCRO negative relative to SCCRO positive | 1.72 | 1.02-3.00 | .04 |
| Stage I/II relative to Stage III/IV | 0.24 | 0.05-0.53 | < . 001 |
ACC, Adrenocortical carcinoma; SCCRO, squamous cell carcinoma-related oncogene.
*Analysis includes SCCRO phenotype with stage.
+Model: Chi-square = 18.89; degrees of freedom = 2; P < . 001.
expression was observed early in embryonic de- velopment and persisted throughout all in utero stages. For the purposes of this project, we focused on expression in relation to adrenal development. Prominent SCCRO expression in the adrenal cortex was seen in the E15 and E17 stages and persisted in the postnatal adrenal tissue (Fig 2). The adrenal medulla, despite its neuroectodermal origin, did not show any SCCRO expression.
SCCRO expression in adult human normal and neoplastic adrenal tissues. Eighty-seven patients with ACC, 33 patients with adrenocortical adeno- mas, and 38 patients with histologically normal adrenocortical tissues with available paraffin-em- bedded resection specimens were included in the study. A comprehensive description of these pa- tients has been previously published.5,9 Overall, 83% (72/87) of ACC, 94% (31/33) of adrenocor- tical adenomas, and 89% (34/38) of normal adrenocortical specimens on the tissue microarray were suitable for analysis. Rates of core loss (6%- 17%) were within the range of previously pub-
| Covariates | RR | 95% CI | P value |
|---|---|---|---|
| SCCRO phenotype and presence of metastasest | |||
| SCCRO negative relative to SCCRO positive | 1.6 | 0.94-2.83 | .08 |
| Absence vs presence of metastases at diagnosis | 0.39 | 0.20-0.66 | < . 001 |
ACC, Adrenocortical carcinoma; SCCRO, squamous cell carcinoma-related oncogene.
*Analysis includes SCCRO phenotype with presence of metastatic disease at presentation.
+Model: Chi-square = 24.10; degrees of freedom = 2; P < . 001.
lished reports.10,11 Most core losses occurred during processing of paraffin blocks for slide preparation.
Prevalence of SCCRO expression in normal and neoplastic tissue is summarized in Table I. The median age of patients with ACC, 48 ± 18 years (range, 2-77), did not vary by expression status of SCCRO. Consistent with our finding in murine embryos, SCCRO expression was present in 32 of 34 normal adrenocortical glands (94%). The same prevalence of expression (29/31; 94%) was seen in adrenocortical adenomas. Staining was seen pri- marily in the cytoplasm; occasional nuclear stain- ing was also identified. When staining was present, it was identified in all of the adrenocortical cells in the tissue sections. There were no distinguishing histologic features in the 2 normal or adenoma- tous adrenocortical tissue samples that did not express SCCRO.
Overall, the prevalence of SCCRO expression was lower in ACC (63%) than in adrenocortical adenomas or normal adrenal cortical tissue (P =. 001). Although recurrent ACC had a lower prevalence of positive SCCRO phenotype (4/11; 36%) than primary (21/34; 62%) or metastatic (20/28; 71%) lesions, the difference was not statistically significant.
For the purposes of outcome assessment, we limited our analysis to the 38 patients presenting with primary ACC, 34 (89%) of which were suitable for analysis. Clinicopathologic characteristics and SCCRO expression patterns for all patients under- going surgery for primary ACC are summarized in Table II. Loss of SCCRO expression in primary ACC correlated with the presence of distant metastases at presentation (P= . 03) and advanced pathologic stage (P =. 06). There was no correlation between expression of SCCRO and other clinical and path- ologic factors, including sex, neoplasm size or
weight, signs of neoplastic invasion (sinusoidal, venous, capsular, or into adjacent organs), mitotic indices (mitotic rate or presence of atypical mito- ses), cell and neoplastic morphology (nuclear grade, presence of diffuse architecture or clear cytoplasm, presence of necrosis), or presence of a positive microscopic margin.
The overall survival of patients with ACC losing SCCRO expression was poorer than those with a positive phenotype (Fig 3; P= . 006). No patients with non-SCCRO-expressing ACC survived more than 6 years post-treatment. Given the small sample size, we limited multivariate models to 2 variables and performed analyses serially to control for confounding factors. The association between loss of SCCRO expression and outcome remained significant after controlling for the confounding effects of tumor stage by multivariate analysis (Table III; RR = 1.72, 95% CI = 1.02-3.00; P =. 04) and approached significance when controlling for presence of metastatic disease at presentation (Table IV; RR = 1.6, 95% CI = 0.94-2.83; P= . 08).
DISCUSSION
Adrenal glands develop from 2 distinct lineages. The adrenal medulla develops from neuroectodermal tissue, while the adrenal cortex is derived from coelomic epithelium of mesoder- mal origin.12,13 The molecular basis for adrenal development remains obscure, but it is clear that an interaction between the medullary and cortical primordia plays a key role. Several studies have linked Shh signaling to adrenal development. 14-16 Mice lacking fibroblast growth factor receptor 2- IIIb, an upstream member of the hedgehog signaling pathway, display abnormal adrenal de- velopment. We have previously shown that SCCRO is a transcription factor that is also involved in hedgehog signaling. Given the striking pattern of SCCRO expression seen during murine adreno- cortical organogenesis and the established role of the hedgehog pathway in adrenocortical develop- ment, we speculated that SCCRO may also play a role in this process. Accordingly, the aim of this study was to assess the pattern of SCCRO expres- sion in murine and human adrenocortical tissue and to determine whether its expression is dysre- gulated in neoplastic progression to ACC.
We showed that SCCRO is expressed in the majority (94%) of both normal and adenomatous adrenocortical tissues compared to only 63% of ACCs (P= . 001). This suggests that SCCRO may be a marker of terminal differentiation in adrenocor- tical tissue, with loss of SCCRO expression associ-
ated with a dedifferentiated phenotype. In congruence with this hypothesis, loss of SCCRO expression in primary ACC was associated with a highly aggressive clinical phenotype, showing correlations with advanced stage (P =. 06), pres- ence of metastatic disease at presentation (P= . 03), and poorer overall survival (P =. 006), with no patients with loss of SCCRO phenotype surviving more than 6 years post-treatment. The association between the status of SCCRO expression and outcome remained significant after controlling for the confounding effects of stage by multivariate analysis (P =. 04), but only approached signifi- cance after controlling for presence of metastatic disease at presentation (P =. 08). Although these data suggest loss of SCCRO expression may be an independent predictor of outcome in these pa- tients, larger studies with greater specimen num- bers are needed to validate and confirm this putative association.
SCCRO is part of a growing list of onco- developmental genes thought to be involved not only in normal cellular regulation but also in tumorigenesis when their activity is not tightly regu- lated.1-4,17 Our findings suggest SCCRO is biologically active in mammalian adrenal development, with loss of gene expression serving as either a marker or instigator of dedifferentiation in ACC. Although more work needs to be done to determine any putative mechanistic relationship with SCCRO, our findings raise several intriguing possibilities for both adrenal development and ACC pathogenesis. From a clinical perspective in this small and selected-albeit well-characterized-cohort, analy- sis of SCCRO expression may be a useful molecular marker in differentiating adrenocortical adenomas from ACC. In addition, loss of SCCRO expression in primary ACC is associated with an aggressive tumor phenotype and may be an indicator of pro- gressive dedifferentiation.
REFERENCES
1. Singh B, Gogineni SK, Sacks PG, Shaha AR, Shah JP, Stoffel A, et al. Molecular cytogenetic characterization of head and neck squamous cell carcinoma and refinement of 3q amplification. Cancer Res 2001;61:4506-13.
2. Singh B, Stoffel A, Gogineni S, Poluri A, Pfister DG, Shaha AR, et al. Amplification of the 3q26.3 locus is associated with progression to invasive cancer and is a negative prognostic factor in head and neck squamous cell carcino- mas. Am J Pathol 2002;161:365-71.
3. Singh B, Gogineni S, Goberdhan A, Sacks P, Shaha A, Shah J, et al. Spectral karyotyping analysis of head and neck squamous cell carcinoma. Laryngoscope 2001;111:1545-50.
4. Singh B, Reddy PG, Goberdhan A, Walsh C, Dao S, Ngai I, et al. p53 regulates cell survival by inhibiting PIK3CA in squamous cell carcinomas. Genes Dev 2002;16:984-93.
5. Stojadinovic A, Ghossein RA, Hoos A, Nissan A, Marshall D, Dudas M, et al. Adrenocortical carcinoma: clinical, mor- phologic, and molecular characterization. J Clin Oncol 2002;20:941-50.
6. Harrison LE, Gaudin PB, Brennan MF. Pathologic features of prognostic significance for adrenocortical carcinoma after curative resection. Arch Surg 1999;134:181-5.
7. Weiss LM, Medeiros LJ, Vickery AL, Jr. Pathologic features of prognostic significance in adrenocortical carcinoma. Am J Surg Pathol 1989;13:202-6.
8. Fuhrman SA, Lasky LC, Limas C. Prognostic significance of morphologic parameters in renal cell carcinoma. Am J Surg Pathol 1982;6:655-63.
9. Stojadinovic A, Brennan MF, Hoos A, Omeroglu A, Leung DH, Dudas ME, et al. Adrenocortical adenoma and carci- noma: histopathological and molecular comparative analy- sis. Mod Pathol 2003;16:742-51.
10. Kononen J, Bubendorf L, Kallioniemi A, Barlund M, Schraml P, Leighton S, et al. Tissue microarrays for high- throughput molecular profiling of tumor specimens. Nat Med 1998;4:844-7.
11. Hoos A, Urist MJ, Stojadinovic A, Mastorides S, Dudas ME, Leung DH, et al. Validation of tissue microarrays for immunohistochemical profiling of cancer specimens using the example of human fibroblastic tumors. Am J Pathol 2001;158:1245-51.
12. Nyska A, Maronpot R. Adrenal gland. In: Maronprot R, Boorman G, Gaul B, editors. Pathology of the mouse. Reference and atlas. 1st ed. Vienna, IL: Cache River Press; 1999: 509-36.
13. Heikkila M, Peltoketo H, Leppaluoto J, Ilves M, Vuolteena- ho O, Vainio S. Wnt-4 deficiency alters mouse adrenal cortex function, reducing aldosterone production. Endo- crinology 2002;143:4358-65.
14. Ma Y, Chai L, Cortez SC, Stopa EG, Steinhoff MM, Ford D, et al. SALL1 expression in the human pituitary-adrenal/ gonadal axis. J Endocrinol 2002;173:437-48.
15. Kohlhase J, Wischermann A, Reichenbach H, Froster U, Engel W. Mutations in the SALL1 putative transcription factor gene cause Townes-Brocks syndrome. Nat Genet 1998;18:81-3.
16. Revest JM, Spencer-Dene B, Kerr K, De Moerlooze L, Rosewell I, Dickson C. Fibroblast growth factor receptor 2- IIIb acts upstream of Shh and Fgf4 and is required for limb bud maintenance but not for the induction of Fgf8, Fgf10, Msx1, or Bmp4. Dev Biol 2001;231:47-62.
17. Estilo CL, P OC, Ngai I, Patel SG, Reddy PG, Dao S, et al. The role of novel oncogenes squamous cell carcinoma- related oncogene and phosphatidylinositol 3-kinase p110al- pha in squamous cell carcinoma of the oral tongue. Clin Cancer Res 2003;9:2300-6.
DISCUSSION
Dr Herbert Chen (Madison, Wis). I have two ques- tions for you. First, in your conclusions you state that loss of SCCRO may be a marker for loss of differentiation. I wonder if you have done any studies to compare SCCRO expression to actual markers of differentiation.
Secondly, you stated in your introduction that SCCRO signals to the hedgehog signaling pathway, and I was wondering if you had the opportunity to examine for expression of hedgehog factors such as GLI1 or patched.
Dr Quan-Yang Duh (San Francisco, Calif). Do you have specimens that have come from the same patient, so you can compare the normal tissue, the primary tumor, and the recurrence. If so, are they different?
Dr Alan P. B. Dackiw (Baltimore, Md). Both the murine and human adrenal have a fetal zone which regresses. In your immunohistochemical stains for SCCRO in the mouse, was there any correlation in the degree of localization of the staining with the age and sex of the mouse? Did you see any difference in SCCRO expression between the fetal and permanent zones of the developing adrenal cortex.
Dr Sarkaria. Dr Chen asked about comparison of SCCRO expression to actual markers of differentiation. We have not correlated loss of SCCRO expression with markers of differentiation in ACC mainly because there are no reliable markers of differentiation described for ACC. Most indices considered markers of differentia- tion, such as nuclear atypia or mitotic rate, are not used to distinguish ACC from adrenocortical adenomas. The concept of a well versus poorly differentiated ACC is one that is emerging with newer molecular markers, such as SCCRO, that better characterize the biological state of these lesions.
To answer his second question about looking for expression of hedgehog factors such as GLI1 or patched, Shh pathway components, including GLI1 and patched, are upregulated in embryonic stem cells stably trans- fected with SCCRO. Furthermore, SCCRO and GLI1 levels correlate in NSCLC and in other tumors with known Shh-dependent oncogenic mechanisms, such as medulloblastomas. In addition, we have shown that SCCRO directly upregulates GLI1 expression by binding to its promoter region and driving transcription.
Dr Duh asked if specimens of normal tissue, primary tumor, and recurrent tumor from the same patient were available. We did not have such specimens. These archived pathology specimens, especially the normal and ACC tissues, are from different patients. This study design with independent cohorts of adrenocortical tissues from normal and neoplastic adrenal glands was utilized to minimize bias due to specimens from the same patient with a similar genetic makeup.
Dr Dackiw asked if the mice’s age and sex affected SCCRO expression and if the expression differed in the fetal and permanent zones of the developing adrenal cortex. SCCRO expression was seen throughout the developing adrenal cortex, in both the fetal and perma- nent zone, regardless of the age or sex of the mice.