Abstract
Prediction of malignant behaviour of pheochromocytomas/sympathetic paragangliomas (PCCs/PGLs) is very difficult if not impossible on a histopathological basis. In a familial setting, it is well known that succinate dehydrogenase subunit B (SDHB)-associated PCC/PGL very often metastasise. Recently, absence of SDHB expression as measured through immunohistochemistry was shown to be an excellent indicator of the presence of an SDH germline mutation in PCC/PGL. SDHB loss is believed to lead to tumour formation by activation of hypoxia signals. To clarify the potential use of SDHB immunohistochemistry as a marker of malignancy in PCC/PGL and its association with classic hypoxia signalling we examined SDHB, hypoxia inducible factor-1α (Hif-1α) and its targets CA-9 and GLUT-1 expression on protein level using immunohistochemistry on a tissue micro array on a series of familial and sporadic tumours of 115 patients. Survival data was available for 66 patients. SDHB protein expression was lost in the tumour tissue of 12 of 99 patients. Of those 12 patients, 5 had an SDHB germline mutation, in 4 patients no germline mutation was detected and mutational status remained unknown in parts in 3 patients. Loss of SDHB expression was not associated with increased classic hypoxia signalling as detected by Hif-1α, CA-9 or GLUT-1 staining. Loss of SDHB expression was associated with an adverse outcome. The lack of correlation of SDHB loss with classic hypoxia signals argues against the current hypoxia hypothesis in malignant PCC/PGL. We suggest SDHB protein loss as a marker of adverse outcome both in sporadic and in familial PCC/PGL.
Introduction
Pheochromocytomas (PCCs) are rare tumours of neural crest-derived chromaffin cells. Most tumours arise in the adrenal gland, but up to 23% (Karagiannis et al. 2007) are localised in extra-adrenal tissue and are called sympathetic paragangliomas (PGLs). About 30% of these tumours occur in familial tumour syndromes (Neumann et al. 2002, Amar et al. 2005, Tischler 2008, Komminoth 2009) including neurofibromatosis type 1 (NF1), von Hippel–Lindau disease (VHL), multiple endocrine neoplasia type 2 (MEN2) and succinate dehydrogenase (SDH) syndromes.
The underlying molecular mechanisms leading to PCCs/sympathetic PGLs are not fully understood. Several studies suggest that classic hypoxia signalling involving hypoxia inducible factor-1α (Hif-1α) may play a central role, at least in the context of VHL deficiency (Maxwell et al. 1999). VHL inactivation leads to classic hypoxia signalling by inhibiting the degradation of the Hif-1α: under normoxic conditions, Hif-1α is hydroxylated and degraded. Under hypoxia and in some types of cancer it stabilises and accumulates in the nucleus. This leads to an upregulation of the Hif target genes CA-9 and GLUT-1 on the one hand and to an increased expression of VEGF, resulting in an increased microvessel density, on the other hand (Couvelard et al. 2005). This mechanism is supported by expression analysis of PCC: on the RNA level, Eisenhofer et al. (2004) described the activation of hypoxia-driven angiogenic pathways in VHL syndrome PCCs. Hif-1α was not in the list of upregulated genes, possibly because it is regulated on protein rather than on mRNA level (van Uden et al. 2008). Dahia et al. (2005) described a hypoxia-induced expression profile in both VHL- and SDH-induced PCCs/PGLs. By gain- and loss-of-function analyses they additionally suggested that the link between hypoxia signals (through pVHL) and mitochondrial signals (through SDH) is mediated by Hif-1α. Succinate accumulating due to SDH mutations can inhibit the degradation of Hif-1α as do VHL mutations in this in vitro model. This is thought to cause upregulation of Hif targets leading to tumourigenesis (Dahia et al. 2005, Selak et al. 2005, 2006) in SDH-associated PCCs/PGLs. These hopytheses have been challenged by a recent study on hypoxia signalling in hereditary PCCs/PGLs: on mRNA levels there was no difference of Hif-1α expression between the different tumour types but Hif-2α was overexpressed in VHL- and SDH-related tumours, a finding that was also reflected by protein expression analysis (Favier et al. 2009). Hif-2α is reported to be associated with a poor patient outcome in several tumour types (Qing & Simon 2009). Other results also point towards alternative mechanisms than classic hypoxia signalling: VHL mutants leading to the VHL2c phenotype consisting exclusively of PCCs retain their ability to downregulate Hif (Hoffman et al. 2001). In Caenorhabditis elegans, a subset of genes dysregulated in vhl mutants is not normalised in vhl/hif-1 double mutants (Bishop et al. 2004). Failure of developmental apoptosis may be the hypoxia-independent mechanism of PCC pathogenesis (Lee et al. 2005), possibly with a regulatory loop including Hif (Maxwell 2005).
In the familial setting, it is well known that SDH subunit B (SDHB)-associated PCCs/PGLs very often lead to metastases (Timmers et al. 2007), sometimes many years after resection (Maier-Woelfle et al. 2004). Prediction of malignancy in sporadic tumours is an unsolved problem; the only definite evidence of malignancy is the detection of metastases. The ‘PCC of the adrenal gland scaled score (PASS)’, a morphological scoring system to identify more aggressive tumours (Thompson 2002), has not proved to be useful due to great interobserver variability in a recent study (Wu et al. 2009). An increased risk of malignancy seems to be indicated by Ki-67 proliferation indices >2 or 3%, but this is of limited clinical use (van der Harst et al. 2000, August et al. 2004, Kimura et al. 2005, Strong et al. 2008). Recently, loss of SDHB immunohistochemistry was shown to be an excellent indicator of the presence of an SDH germline mutation in PCCs and PGLs (van Nederveen et al. 2009). Its potential use for predicting biological behaviour is unknown.
To clarify the potential use of SDHB immunohistochemistry as a marker of malignancy and its association with classic hypoxia signalling, we decided to examine SDHB, Hif-1α and its targets CA-9 and GLUT-1 expression on protein level on a series of familial and sporadic PCCs and sympathetic PGLs and to correlate the results with survival data.
Materials and methods
Patients and tumour specimens
All PCCs and sympathetic PGLs analysed at the Institute of Surgical Pathology, University Hospital Zurich in the years from 1975 to 2006 were included.
Clinical data and follow-up information were extracted retrospectively from patient charts. A questionnaire enquiring about tumour relapse or progression was sent to family doctors.
Tumour specimens
We analysed PCC/PGL specimens from 115 patients (62 males, 46 females and sex could not be evaluated in 7 patients). The available paraffin specimens comprised 112 primary tumours and 3 metastases. Of the 112 primary tumours, 26 were of extra-adrenal abdominal localisation (19 males and 8 females), 76 originated from the adrenal medulla (37 males, 32 females and 7 not specified) and 8 originated extra-abdominally. Information about the localisation was not available in two tumours. Of the three metastases one was of an extra-adrenal abdominal primary (patient with known SDHB germline mutation), one of an extra-abdominal primary (patient without SDHB germline mutation) and the one metastasis was of an unknown primary (mutational status of the patient unknown). A tissue micro array (TMA) comprising these 115 tissues was constructed as described previously (Bubendorf et al. 2001).
Syndromic patients
One male patient suffered from NF1 disease clinically, five patients from VHL disease (four with proven VHL mutation, one clinical VHL disease with multiple bilateral clear cell renal cell carcinomas in addition to the PCC (one male, three females and one not specified)) and six patients from MEN2 (proven RET mutation (three males and three females)) (Table 1). An overview of patients and follow-up data is given in Table 2.
Results of immunohistochemical hypoxia stainings in familial tumours
| Patient no. | Syndrome | Sex | Organ | SDHB | Hif-1α | CA-9 |
|---|---|---|---|---|---|---|
| 40 | NF1 | M | Adrenal gland | + | + | − |
| 28 | RET(MEN2) | M | Adrenal gland | + | − | − |
| 39 | RET(MEN2) | M | Adrenal gland | + | − | − |
| 60 | RET(MEN2) | M | Adrenal gland | + | − | − |
| 61 | RET(MEN2) | F | Adrenal gland | + | − | − |
| 71 | RET(MEN2) | F | Adrenal gland | + | − | − |
| 117 | RET(MEN2) | F | Adrenal gland | + | − | − |
| 5 | SDHB | M | Abdominal ea | NA | − | − |
| 11 | SDHB | M | Abdominal ea | − | + | + |
| 33 | SDHB | F | Abdominal ea | − | − | − |
| 77 | SDHB | M | Abdominal ea | − | − | − |
| 78 | SDHB | M | Abdominal ea | − | − | − |
| 21 | VHL | M | Adrenal gland | NA | − | NA |
| 54 | VHL | F | Adrenal gland | + | + | + |
| 115 | VHL | F | Adrenal gland | + | + | − |
| 126 | VHL | UK | Adrenal gland | + | − | − |
| 42 | VHL | F | Adrenal gland | + | − | − |
−, no staining; +, positive staining; NA, not assessable; ea, extra-adrenal; UK, unknown.
Clinical characteristics of 115 pheochromocytomas (PCC) patients
| Total | Men | Women | Unknown | |
|---|---|---|---|---|
| Patients | 115 | 62 (53.9%) | 46 (40.0%) | 7 (6.1%) |
| Multifocal | 19 (16.5%) | 11 (9.6%) | 7 (6.1%) | 1 (0.9%) |
| Follow-up available | 66 (57.4%) | 35 (30.4%) | 31 (27.0%) | 0 (0%) |
| Tumour-related death | 15 (13.0%) | 5 (4.3%) | 3 (2.6%) | 7 (6.1%) |
| No tumour-related death/alive | 59 (51.3%) | 30 (26.1%) | 29 (25.2%) | 0 (0%) |
| Follow-up | ||||
| Range (month) | 2–291 | 5–291 | 2–188 | – |
| Mean (month) | 78.39 | 77.4 | 79.52 | – |
| Median (month) | 55.5 | 52 | 71 | – |
| Syndromic patients | ||||
| NF1 | 1 (0.9%) | 1 (0.9%) | 0 (0%) | 0 (0%) |
| VHL | 5 (4.3%) | 1 (0.9%) | 3 (2.6%) | 1 (0.9%) |
| MEN2 | 6 (5.2%) | 3 (2.6%) | 3 (2.6%) | 0 (0%) |
| SDHB | 5 (4.3%) | 4 (3.5%) | 1 (0.9%) | 0 (0%) |
| Localisation (primary tumour) | ||||
| Unknown | 3 (2.6%) | 2 (1.7%) | 1 (0.9%) | 0 (0%) |
| Adrenal | 76 (66.1%) | 37 (32.2%) | 32 (27.8%) | 7 (6.1%) |
| Extra-adrenal | 33 (28.7%) | 22 (19.1%) | 11 (9.6%) | 0 (0%) |
Immunohistochemistry
The analysis was performed on 4 μm sections from the TMA, which was stained with antibodies against CA-9, Hif-1α, CD34, GLUT-1 and SDHB. The immunohistochemical staining for the antigens was performed on automated staining systems (CA-9 on Bond Refine, Vision BioSystems Ltd, Newcastle upon Tyne, UK; Hif-1α on Bond Refine, Vision BioSystems Ltd; CD34 on Ventana BenchMark, Ventana Medical Systems, Tucson, AZ, USA; GLUT-1 on Ventana BenchMark). For SDHB staining, the slides were pretreated by microwave heating in Tris/EDTA buffer, pH 9.0 at 100 °C for 40 min or citrate buffer, pH 6.0 for 15 min. After rinsing in tap water followed by incubation in 3% H2O2 in PBS for 15 min the SDHB antibody was incubated overnight at 4 °C. The presence of tumour tissue was verified by synaptophysin and H&E stainings in all punch cylinders.
The following antibodies were used: CA-9 polyclonal antibody ab15086 (Abcam, Cambridge, UK), dilution 1:200; Hif-1α monoclonal antibody ab16066 (Abcam), dilution 1:500; GLUT-1 polyclonal antibody AB1341 (Chemicon International, Temecula, CA, USA), dilution 1:1000 and SDHB rabbit polyclonal HPA002868 (Sigma–Aldrich Corp.), dilution 1:250; CD34 clone QBEND/10 (MCAP 547, Serotec, MorphoSys, Oxford, UK), dilution 1:800.
Visualisation was accomplished using the avidin–biotin complex method leading to a brown staining signal. As controls, for SDHB and CD34 endothelial cells served as internal positive control, for CA-9 and GLUT-1 normal liver tissue was used and for Hif-1α glioblastoma tissue was used.
For CA-9 cytoplasmic and/or membranous staining was scored positive. For Hif-1α nuclear and cytoplasmic stainings were evaluated separately. For GLUT-1 cytoplasmic staining was scored positive. For Hif-1α and GLUT-1, depending on the intensity of staining, a semiquantitative scoring system was used, comprising strongly positive, weakly positive and negative immunoreactivities. For CA-9 stainings were scored as either negative or positive. In all stainings tumours with <5% positive tumour cells were scored as negative.
SDHB was scored as positive if the cytoplasm showed a strong dot-like positivity. We categorised the tumour as negative if the cytoplasm was negative in the presence of internal positive control in endothelial cells. Tumours with homogeneous faint cytoplasmic staining were scored negative (van Nederveen et al. 2009).
Microvessel density was calculated by counting all vessel sections in tumour tissue of each TMA cylinder with the help of a grid and a counting device in relation to the area consisting of tumour tissue. The number of vessel sections per square millimetre was calculated. Examples of immuohistochemical stainings are given in Fig. 1.
Immunohistochemistry of PCC/PGL. (A) Negative for CA-9; (B) positive for CA-9; (C) nuclear negativity for HIF-1α; (D) nuclear positivity for HIF-1α; (E) negative for SDHB (positive signal in endothelial cells); (F) positive for SDHB (note granular cytoplasmic staining); (G) first quartile microvessel density (MVD); (H) second quartile MVD; (I) third quartile MVD; (J) fourth quartile MVD.
Citation: Endocrine-Related Cancer 17, 4; 10.1677/ERC-09-0316
All immunohistochemical stainings were evaluated in a blinded manner by at least two pathologists seperately (A B, A P and F v N).
SDHB mutation analysis
In patients with negative SDHB immunohistochemistry germline mutation analysis of the SDH genes was performed on peripheral blood or non-neoplastic paraffin tissue after obtaining informed consent. Where no non-neoplastic tissue was available, mutation analysis was performed in tumour tissue. All mutation analyses were performed at least twice in two laboratories (Bern or Zurich and Rotterdam).
DNA was extracted from peripheral blood using the Puregene kit (GentraSystems, Minneapolis, MN, USA) according to the manufacturer's instructions. DNA was extracted from paraffin tissue as described previously (Maier-Woelfle et al. 2004). Mutation analysis was performed by denaturing gradient gel electrophoresis-based mutation analysis as described previously (Maier-Woelfle et al. 2004). PCRs were repeated for all samples with abnormal banding patterns followed by cycle sequencing. Mutation analysis was repeated in a second laboratory by cycle sequencing in all patients where no SDHB-mutation was detected.
SDHB deletion analysis
In all SDHB immunonegative cases where no SDHB mutation was detected fluorescence in situ hybridisation (FISH) analysis was performed to detect copy number losses. Deletion analysis was performed by double-target FISH as described before (Schmitt et al. 2009) using a chromosome 1-specific centromere probe in combination with a 1p36.1–1p35-specific probe containing the SDHB gene. Images were recorded with Analysis software (Olympus Biosystems, Hamburg, Germany).
Statistical analysis
The statistical analysis was performed with SPSS version 16.0.1 (SPSS Software, Chicago, IL, USA). We used two-sided Pearson's χ2-test and Spearman correlation coefficient to analyse dependence of the data. Kaplan–Meier curves were used for demonstration of survival. P values <0.05 were considered to indicate statistical significance.
Ethics
The study was approved by the local ethical committee (Kantonale Ethikkommission, StV 37-2006).
Results
Follow-up
Follow-up data were available for 66 patients (57.4%). The survival data ranged from 2 to 291 months (mean: 78.39 months; median: 55.5 months). In 15 patients (13.0%), death was caused by tumour (5 males, 3 females and 7 not specified) (Table 2).
CA-9 immunohistochemistry
Tumours of 111 patients (96.5%) could be evaluated for CA-9 protein expression by immunohistochemistry. In total, 15 (13.5%) of these were scored positive (11 males and 4 females), 96 (86.5%) were negative (48 males, 41 females and 7 without details about sex).
Hif-1α immunohistochemistry
Hif-1α could be evaluated in 114 patients (99.1%). Nuclear staining was negative in 110 of 114 (96.5%) patients (60 males, 43 females and sex of 7 patients unknown), only 4 of 114 (3.5%) patients showed nuclear staining (2 males and 2 females).
Regarding cytoplasmic staining, 6 of 114 (5.3%) showed a strong positive staining (4 males, 1 female and sex of 1 patient unknown) and 14 of 114 (12.3%) patients were weakly positive (6 males, 4 females and sex of 4 patients unknown) whereas the majority of patients (94 of 114, 82.5%) were negative (52 males, 40 females and sex of 2 patients unknown).
GLUT-1 immunohistochemistry
GLUT-1 could be evaluated in 112 patients (97.5%), 6 of them showed a strong positivity (3 males, 2 females and sex of 1 patient unknown), 2 showed a strong positivty (1 male and 1 female) and 104 of 112 patients (56 males, 42 females and sex of 6 patients unknown) were negative.
SDHB immunohistochemistry
SDHB immunohistochemistry could be evaluated in 99 patients (86.1%). In 16 patients SDHB immunohistochemistry could not be evaluated due to lack of a positive internal control in the endothelium of the intratumoural vessels. In all, 12 of 99 (12%) patients were SDHB negative, including all the 5 patients, in whom an SDHB germline mutation was detected (10 males and 2 females); 87 of 99 (87%) were weakly or strongly positive (41 males, 39 females and 7 unknown). Follow-up was available for 60 of these 99 patients including 9 of the 12 SDHB immunonegative patients.
Eleven of the 12 (92%) SDHB-negative tumours were localised outside the adrenal gland (i.e. extra-adrenal sympathetic PGLs).
Three of the 12 (25%) SDHB-negative tumours showed a strong positivity in the CA-9 staining, and 9 (75%) were negative (not significant in cross tabulation).
None of the SDHB-negative patients showed a strong nuclear positivity for Hif-1α, all 12 were negative for Hif-1α in the nucleus. Two (17%) SDHB-negative patients showed a strong cytoplasmic positivity for Hif-1α, 1 (8%) was weakly positive and 9 (75%) were negative for Hif-1α.
Mutation status in SDHB-negative tumours
In two brothers an SDHB germline mutation was already known. Mutation analysis of the SDHB, SDHC and SDHD genes was performed in all the remaining 10 patients. SDHB germline mutation was excluded in one patient by analysing non-neoplastic tissue. In three additional patients SDHB mutations were excluded in tumour tissues as no non-neoplastic tissue was available. In the remaining three patients SDHB mutations could be excluded in parts of the SDHB gene in tumour tissue, the remaining exons were not amplifiable (Table 3). All patients without an SDHB germline mutation were tested for SDHC and SDHD mutations. SDHC mutations were absent in all three informative patients and SDHD mutations were absent in all five informative patients.
Clinical characteristics of SDHB immunonegative tumours
| PGL-no. | Sex | Syndrome | SDHB mutation | Tumour | Localisation of primary tumour | Localisation of metastasis | Multifocal | Follow-up (age at time of death) |
|---|---|---|---|---|---|---|---|---|
| aPGL001 | M | No | All ex neg (tu) | Primary tumour | Abdominal ea/paraaortal | No | No | AWD |
| aPGL007 | M | SDHB | Del 632G in Exon 5 | Metastasis | Abdominal ea/pararenal | Spine | No | DOD (37) |
| aPGL030 | M | No | All ex neg. (tu+nnt) | Primary tumour | Abdominal ea/pararenal | Liver, bones, lymph nodes | No | DOD (62) |
| aPGL031 | M | SDHB | C. 307insC/p.103 Trp fs | Primary tumour | Abdominal ea/paraaortal | No | NA | |
| aPGL047 | M | No | Ex 4/6 neg. remaining NA (tu) | Primary tumour | Adrenal gland | No | No | NA |
| aPGL058 | M | No | Ex 1/2/3/4/5/6/7 neg, 8 NA (tu) | Primary tumour | Abdominal ea/paraaortal | Aorta, small bowel, bones | Yes | AWD |
| aPGL091 | F | SDHB | Del 632 G in Exon 5 | Primary tumour | Abdominal ea/pararenal | No | No | AWD |
| aPGL204 | M | SDHB | H132P | Primary tumour | Abdominal ea/paraaortal | Lymph nodes | No | DOD (43) |
| aPGL205 | M | SDHB | H132P | Metastasis | Abdominal ea/paraaortal | Bones | Yes | DOD (60) |
| aPGL243 | F | No | All ex neg. (tu) | Metastasis | Mediastinal | Brain | Yes | DOD (56) |
| aPGL320 | M | No | Ex 1/2/4/5/6 neg (tu) | Primary tumour | Abdominal ea/paraaortal | No | No | AWD |
| aPGL347 | M | No | Ex 6 neg. (tu) | Primary tumour | Abdominal ea/paraaortal | No | Yes | NA |
SDHB, succinate dehydrogenase subunit B; Del, deletion; ins, insertion; ea, extra-adrenal; NA, not assessable; tu, tumour; nnt, non-neoplastic tissue; AWD, alive with disease; DOD, death of disease.
Microvessel density
In 109 evaluable cases the number of vessel sections in the area of the cylinder (0.28 mm2), corrected for non-neoplastic portions if present, ranged from 3.75 to 1144 (mean 229 and median 122). Tumours were subdivided into quartiles according to the number of vessel sections per mm2. The first quartile ranged from 13 to 535 vessel sections per mm2, the second from 548 to 1083 vessel sections per mm2, the third from 1268 to 1530 vessel sections per mm2 and the fourth from 1641 to 4046 vessel sections per mm2.
SDHB deletion analysis
In seven SDHB immunonegative cases without a detectable SDH mutation SDHB deletion analysis was performed by FISH. One SDHB immunonegative case with an SDHB germline mutation and a known deletion served as a positive control. In five of these seven SDHB immunonegative cases no SDHB deletion was detected. Two cases were not evaluable in repeated analyses.
Statistical analysis
SDHB immunoreactivity did neither correlate with cytoplasmic CA-9 staining (P=0.150; rs=0.000) nor with nuclear or cytoplasmic Hif-1α staining (P=0.555; rs=0.109 and P=0.399; rs=−0.010 respectively) nor with GLUT-1 (P=0.780; rs=0.083).
Significant correlation was observed between both nuclear and cytoplasmic Hif-1α immunoreactivity and CA-9 staining (P=0.000; rs=0.348 and P<0.000; rs=0.123 respectively).
Survival analysis was performed for all immunohistochemical markers. The survival of patients with SDHB immunonegative tumours was significantly adverse compared with SDHB-positive tumours (P<0.0001; Fig. 2). No survival difference between a weak and strong SDHB positivity was found. An extra-adrenal localisation of the primary tumour was also associated with a shortened tumour-specific survival (P=0.005). The survival depending on hypoxia markers CA-9, Hif-1α or GLUT-1 offered no significant trend.
Kaplan–Meier curve depicting percentage of tumour-specific survival for PCC/PGL with (upper curve) and without (lower curve) SDHB immunoreactivity.
Citation: Endocrine-Related Cancer 17, 4; 10.1677/ERC-09-0316
Increased microvessel density correlated significantly with loss of SDHB (P=0.017; rs=−0.289) but not with Hif-1α in the nucleus (P=0.192; rs=0.171), Hif-1α in the cytoplasm (P=0.496; rs=0.048), GLUT-1 (P=0.990; rs=−0.040) and CA-9 (P=0.186; rs=0.070). An overview over the correlation results is given in Table 4.
Correlations and correlation coefficient of immunohistochemistry and number of vessels
| Hif-1α nucleus | Hif-1α cytoplasma | SDHB | GLUT1 | Number of vessels | Localisation | |
|---|---|---|---|---|---|---|
| CA9 | 0.000 | 0.000 | 0.150 | 0.834 | 0.186 | 0.406 |
| Spearman correlation | 0.348 | 0.123 | 0.000 | −0.006 | −0.070 | |
| Hif-1α nucleus | 0.643 | 0.555 | 0.853 | 0.192 | 0.291 | |
| Spearman correlation | −0.088 | 0.109 | −0.053 | 0.171 | ||
| Hif-1α cytoplasma | 0.399 | 0.301 | 0.496 | 0.174 | ||
| Spearman correlation | −0.010 | 0.148 | 0.048 | |||
| SDHB | 0.780 | 0.017 | 0.000 | |||
| Spearman correlation | 0.083 | −0.289 | ||||
| GLUT1 | 0.990 | 0.665 | ||||
| Spearman correlation | −0.40 |
P values bold numbers indicate statistical significance.
Discussion
Prediction of malignant behaviour of PCCs is very difficult if not impossible on a histopathological basis. A proposed morphological scoring scheme did not prove useful in a recent analysis due to large inter-observer variablility (Thompson 2002, Kimura et al. 2005, Wu et al. 2009). The present survival analysis on 66 unselected PCC/sympathetic PGL patients with follow-up revealed SDHB immunohistochemistry as a possible prognostic marker. The survival of patients with SDHB immunonegative tumours was significantly worse than of patients with SDHB-positive tumours. Five of nine patients with SDHB immunonegative tumours and available follow-up died because of the tumours; of those five patients three had an SDHB germline mutation. A tumour-related death did not occur in the six strongly SDHB-positive tumours and only twice in the 46 weakly SDHB-positive tumours with available follow-up. A possible explanation of these findings is that the SDHB-negative tumours arise in patients with the SDHB-associated PCC/PGL syndrome. Indeed, we could demonstrate an SDHB germline mutation in five of the nine immunonegative patients we were able to examine. SDHB mutations were excluded in all coding regions of four patients. We were only able to exclude SDHB mutations in the remaining three immunonegative patients in a minority of SDHB exons. In tumours with SDHB germline mutation inactivation of the wild-type allele occurs by loss of heterozygosity (LOH) (Gimenez-Roqueplo et al. 2002, Maier-Woelfle et al. 2004); we were able to exclude such LOH by FISH analysis in five of the seven immunonegative cases without germline mutation. SDHB germline deletions are described in up to 30% of SDHB kindreds (McWhinney et al. 2004, Tischler 2008). In these tumours LOH would also be expected. Therefore, we postulate that at least these five patients with SDHB immunonegative tumours without mutation or deletion suffer from sporadic disease. van Nederveen et al. (2009) described SDHB loss by immunohistochemistry in 11–16% of sporadic tumours, a similar frequency as described in this study. Complete loss of granular SDHB immunopositivity has been shown as a predictor of SDH germline mutation with a positive predictive value of at least 92% and a negative predictive value of 100% (van Nederveen et al. 2009). In our series, both tumours with a known SDH germline mutation stained negative for SDHB immunohistochemistry, confirming the high negative predictive value of a positive SDHB staining. From the clinical point of view our data also stress the role of localisation of the tumour. In our series there was a strong correlation of extra-adrenal localisation with poor prognosis and SDHB germline mutation, which is in line with previous findings (Neumann et al. 2002, Maier-Woelfle et al. 2004, Amar et al. 2005, Lenders et al. 2005, Tischler 2008).
One aim of our study was to examine the possible link between SDHB immunostaining, which is indicative of enzymatic activity of the SDH complex (van Nederveen et al. 2009) with classic hypoxia signalling, as suggested by several authors (Dahia et al. 2005, Pollard et al. 2006). Classic hypoxia signalling through Hif-1α and its target genes CA-9 and GLUT-1 plays a central role in the development of clear cell renal cell carcinoma and glioblastoma (Kaelin 2004, Kaur et al. 2005). Thus, we examined the expression of Hif-1α, CA-9 and GLUT-1 by immunohistochemistry. Classic hypoxia signalling as characterised by an accumulation of these proteins in our study could be detected only in a small minority of tumours and did not have any influence on survival. There was no association of SDHB loss with CA-9, GLUT-1 and Hif-1α protein level. In contrast, SDHB protein loss correlated with a high microvessel density. Moreover, SDHB loss correlated with an adverse outcome, while a high microvessel density did not significantly correlate with a poor survival (Fig. 3). The two markers of classical hypoxia, Hif-1α and CA-9, showed a strong correlation with each other, arguing against technical problems in identifying classic hypoxia signals. These findings argue against an important role of classic hypoxia signalling in the development of PCCs/PGLs. However, pseudohypoxia, which is mediated by Hif-2α, might play a role as suggested by Favier et al. (2009) who found a high Hif-2α expression in both VHL- and SDH-associated PCCs. However, while the same authors described an upregulation of glucose transporters solely in VHL-associated PCC we cannot confirm these findings: in our study all VHL-associated cases were immunonegative for GLUT-1.
Kaplan–Meier curve depicting percentage of tumour-specific survival for PCC/PGL in dependence of microvessel density (displayed in quartiles). There is no significant correlation between microvessel density and survival.
Citation: Endocrine-Related Cancer 17, 4; 10.1677/ERC-09-0316
Alternatively, other mechanisms than pseudohypoxia signals might be involved in the genesis of SDHB-associated and most sporadic PCCs even in the setting of the VHL syndrome, such as apoptosis and inhibition of microtubule stabilisation as suggested by other authors (Hergovich et al. 2003, Lee et al. 2005). Further studies will be needed to elucidate the mechanisms of tumourigenesis in PCCs/PGLs.
In summary, we suggest SDHB protein loss as a marker of adverse outcome both in sporadic and in familial PCCs/PGLs. Inclusion of this marker in the assessment of PCCs/PGLs might be mandatory for two reasons: first, for direction of molecular genetic testing towards the SDH genes in the case of absent staining, and second as a prognostic marker. We suggest to include patients with sporadic SDHB-negative tumours in more stringent follow-up protocols.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
References
Amar L, Bertherat J, Baudin E, Ajzenberg C, Bressac-de Paillerets B, Chabre O, Chamontin B, Delemer B, Giraud S & Murat A et al. 2005 Genetic testing in pheochromocytoma or functional paraganglioma. Journal of Clinical Oncology 23 8812–8818 doi:10.1200/JCO.2005.03.1484.
August C, August K, Schroeder S, Bahn H, Hinze R, Baba HA, Kersting C & Buerger H 2004 CGH and CD 44/MIB-1 immunohistochemistry are helpful to distinguish metastasized from nonmetastasized sporadic pheochromocytomas. Modern Pathology 17 1119–1128 doi:10.1038/modpathol.3800160.
Bishop T, Lau KW, Epstein AC, Kim SK, Jiang M, O'Rourke D, Pugh CW, Gleadle JM, Taylor MS & Hodgkin J et al. 2004 Genetic analysis of pathways regulated by the von Hippel–Lindau tumor suppressor in Caenorhabditis elegans. PLoS Biology 2 e289 doi:10.1371/journal.pbio.0020289.
Bubendorf L, Nocito A, Moch H & Sauter G 2001 Tissue microarray (TMA) technology: miniaturized pathology archives for high-throughput in situ studies. Journal of Pathology 195 72–79 doi:10.1002/path.893.
Couvelard A, O'Toole D, Turley H, Leek R, Sauvanet A, Degott C, Ruszniewski P, Belghiti J, Harris AL & Gatter K et al. 2005 Microvascular density and hypoxia-inducible factor pathway in pancreatic endocrine tumours: negative correlation of microvascular density and VEGF expression with tumour progression. British Journal of Cancer 92 94–101 doi:10.1038/sj.bjc.6602245.
Dahia PL, Ross KN, Wright ME, Hayashida CY, Santagata S, Barontini M, Kung AL, Sanso G, Powers JF & Tischler AS et al. 2005 A HIF1alpha regulatory loop links hypoxia and mitochondrial signals in pheochromocytomas. PLoS Genetics 1 72–80 doi:10.1371/journal.pgen.0010008.
Eisenhofer G, Huynh TT, Pacak K, Brouwers FM, Walther MM, Linehan WM, Munson PJ, Mannelli M, Goldstein DS & Elkahloun AG 2004 Distinct gene expression profiles in norepinephrine- and epinephrine-producing hereditary and sporadic pheochromocytomas: activation of hypoxia-driven angiogenic pathways in von Hippel–Lindau syndrome. Endocrine-Related Cancer 11 897–911 doi:10.1677/erc.1.00838.
Favier J, Briere JJ, Burnichon N, Riviere J, Vescovo L, Benit P, Giscos-Douriez I, De Reynies A, Bertherat J & Badoual C et al. 2009 The Warburg effect is genetically determined in inherited pheochromocytomas. PLoS ONE 4 e7094 doi:10.1371/journal.pone.0007094.
Gimenez-Roqueplo AP, Favier J, Rustin P, Rieubland C, Kerlan V, Plouin PF, Rotig A & Jeunemaitre X 2002 Functional consequences of a SDHB gene mutation in an apparently sporadic pheochromocytoma. Journal of Clinical Endocrinology and Metabolism 87 4771–4774 doi:10.1210/jc.2002-020525.
van der Harst E, Bruining HA, Jaap Bonjer H, van der Ham F, Dinjens WN, Lamberts SW, de Herder WW, Koper JW, Stijnen T & Proye C et al. 2000 Proliferative index in phaeochromocytomas: does it predict the occurrence of metastases? Journal of Pathology 191 175–180 doi:10.1002/(SICI)1096-9896(200006)191:2<175::AID-PATH615>3.0.CO;2-Z.
Hergovich A, Lisztwan J, Barry R, Ballschmieter P & Krek W 2003 Regulation of microtubule stability by the von Hippel–Lindau tumour suppressor protein pVHL. Nature Cell Biology 5 64–70 doi:10.1038/ncb899.
Hoffman MA, Ohh M, Yang H, Klco JM, Ivan M & Kaelin WG Jr 2001 von Hippel–Lindau protein mutants linked to type 2C VHL disease preserve the ability to downregulate HIF. Human Molecular Genetics 10 1019–1027 doi:10.1093/hmg/10.10.1019.
Kaelin WG Jr 2004 The von Hippel–Lindau tumor suppressor gene and kidney cancer. Clinical Cancer Research 10 6290S–6295S doi:10.1158/1078-0432.CCR-sup-040025.
Karagiannis A, Mikhailidis DP, Athyros VG & Harsoulis F 2007 Pheochromocytoma: an update on genetics and management. Endocrine-Related Cancer 14 935–956 doi:10.1677/ERC-07-0142.
Kaur B, Khwaja FW, Severson EA, Matheny SL, Brat DJ & Van Meir EG 2005 Hypoxia and the hypoxia-inducible-factor pathway in glioma growth and angiogenesis. Neuro-Oncology 7 134–153 doi:10.1215/S1152851704001115.
Kimura N, Watanabe T, Noshiro T, Shizawa S & Miura Y 2005 Histological grading of adrenal and extra-adrenal pheochromocytomas and relationship to prognosis: a clinicopathological analysis of 116 adrenal pheochromocytomas and 30 extra-adrenal sympathetic paragangliomas including 38 malignant tumors. Endocrine Pathology 16 23–32 doi:10.1385/EP:16:1:023.
Komminoth P, Perren A, van Nederveen FH & Krijger RR 2009 Familial endocrine tumours: phaeochromocytomas and extra-adrenal paragangliomas. Diagnostic Histopathology 15 61–68 doi:10.1016/j.mpdhp.2009.01.003.
Lee S, Nakamura E, Yang H, Wei W, Linggi MS, Sajan MP, Farese RV, Freeman RS, Carter BD & Kaelin WG Jr et al. 2005 Neuronal apoptosis linked to EglN3 prolyl hydroxylase and familial pheochromocytoma genes: developmental culling and cancer. Cancer Cell 8 155–167 doi:10.1016/j.ccr.2005.06.015.
Lenders JW, Eisenhofer G, Mannelli M & Pacak K 2005 Phaeochromocytoma. Lancet 366 665–675 doi:10.1016/S0140-6736(05)67139-5.
Maier-Woelfle M, Brandle M, Komminoth P, Saremaslani P, Schmid S, Locher T, Heitz PU, Krull I, Galeazzi RL & Schmid C et al. 2004 A novel succinate dehydrogenase subunit B gene mutation, H132P, causes familial malignant sympathetic extraadrenal paragangliomas. Journal of Clinical Endocrinology and Metabolism 89 362–367 doi:10.1210/jc.2003-031236.
Maxwell PH 2005 A common pathway for genetic events leading to pheochromocytoma. Cancer Cell 8 91–93 doi:10.1016/j.ccr.2005.07.012.
Maxwell PH, Wiesener MS, Chang GW, Clifford SC, Vaux EC, Cockman ME, Wykoff CC, Pugh CW, Maher ER & Ratcliffe PJ 1999 The tumour suppressor protein VHL targets hypoxia-inducible factors for oxygen-dependent proteolysis. Nature 399 271–275 doi:10.1038/20459.
McWhinney SR, Pilarski RT, Forrester SR, Schneider MC, Sarquis MM, Dias EP & Eng C 2004 Large germline deletions of mitochondrial complex II subunits SDHB and SDHD in hereditary paraganglioma. Journal of Clinical Endocrinology and Metabolism 89 5694–5699 doi:10.1210/jc.2004-0769.
van Nederveen FH, Gaal J, Favier J, Korpershoek E, Oldenburg RA, de Bruyn EM, Sleddens HF, Derkx P, Riviere J & Dannenberg H et al. 2009 An immunohistochemical procedure to detect patients with paraganglioma and phaeochromocytoma with germline SDHB, SDHC, or SDHD gene mutations: a retrospective and prospective analysis. Lancet Oncology 10 764–771 doi:10.1016/S1470-2045(09)70164-0.
Neumann HP, Bausch B, McWhinney SR, Bender BU, Gimm O, Franke G, Schipper J, Klisch J, Altehoefer C & Zerres K et al. 2002 Germ-line mutations in nonsyndromic pheochromocytoma. New England Journal of Medicine 346 1459–1466 doi:10.1056/NEJMoa020152.
Pollard PJ, El-Bahrawy M, Poulsom R, Elia G, Killick P, Kelly G, Hunt T, Jeffery R, Seedhar P & Barwell J et al. 2006 Expression of HIF-1alpha, HIF-2alpha (EPAS1), and their target genes in paraganglioma and pheochromocytoma with VHL and SDH mutations. Journal of Clinical Endocrinology and Metabolism 91 4593–4598 doi:10.1210/jc.2006-0920.
Qing G & Simon MC 2009 Hypoxia inducible factor-2alpha: a critical mediator of aggressive tumor phenotypes. Current Opinion in Genetics & Development 19 60–66 doi:10.1016/j.gde.2008.12.001.
Schmitt AM, Schmid S, Rudolph T, Anlauf M, Prinz C, Kloppel G, Moch H, Heitz PU, Komminoth P & Perren A 2009 VHL inactivation is an important pathway for the development of malignant sporadic pancreatic endocrine tumors. Endocrine-Related Cancer 16 1219–1227 doi:10.1677/ERC-08-0297.
Selak MA, Armour SM, MacKenzie ED, Boulahbel H, Watson DG, Mansfield KD, Pan Y, Simon MC, Thompson CB & Gottlieb E 2005 Succinate links TCA cycle dysfunction to oncogenesis by inhibiting HIF-alpha prolyl hydroxylase. Cancer Cell 7 77–85 doi:10.1016/j.ccr.2004.11.022.
Selak MA, Duran RV & Gottlieb E 2006 Redox stress is not essential for the pseudo-hypoxic phenotype of succinate dehydrogenase deficient cells. Biochimica et Biophysica Acta 1757 567–572 doi:10.1016/j.bbabio.2006.05.015.
Strong VE, Kennedy T, Al-Ahmadie H, Tang L, Coleman J, Fong Y, Brennan M & Ghossein RA 2008 Prognostic indicators of malignancy in adrenal pheochromocytomas: clinical, histopathologic, and cell cycle/apoptosis gene expression analysis. Surgery 143 759–768 doi:10.1016/j.surg.2008.02.007.
Thompson LD 2002 Pheochromocytoma of the adrenal gland scaled score (PASS) to separate benign from malignant neoplasms: a clinicopathologic and immunophenotypic study of 100 cases. American Journal of Surgical Pathology 26 551–566 doi:10.1097/00000478-200205000-00002.
Timmers HJ, Kozupa A, Eisenhofer G, Raygada M, Adams KT, Solis D, Lenders JW & Pacak K 2007 Clinical presentations, biochemical phenotypes, and genotype–phenotype correlations in patients with succinate dehydrogenase subunit B-associated pheochromocytomas and paragangliomas. Journal of Clinical Endocrinology and Metabolism 92 779–786 doi:10.1210/jc.2006-2315.
Tischler AS 2008 Pheochromocytoma and extra-adrenal paraganglioma: updates. Archives of Pathology & Laboratory Medicine 132 1272–1284.
van Uden P, Kenneth NS & Rocha S 2008 Regulation of hypoxia-inducible factor-1alpha by NF-kappaB. Biochemical Journal 412 477–484 doi:10.1042/BJ20080476.
Wu D, Tischler AS, Lloyd RV, DeLellis RA, de Krijger R, van Nederveen F & Nose V 2009 Observer variation in the application of the pheochromocytoma of the adrenal gland scaled score. American Journal of Surgical Pathology 33 599–608 doi:10.1097/PAS.0b013e318190d12e.
(A Blank and A M Schmitt contributed equally to this work)
