INTRODUCTION
Adrenal pheochromocytoma (PH) and extra-adrenal-paraganglioma (PGL) are relatively rare endocrine neoplasms derived from chromaffin cells. The majority of PH/PGLs are benign; however, approximately 10% of pheochromocytomas and 15% to 35% of paragangliomas can be associated with malignant potential.[1] The current definition of malignancy by World Health Organization is the presence of metastasis in non-chromaffin tissues.[2,3] Distant metastasis occurs mainly due to hematogenous spread.[1] The 5-year survival rate of patients without metastasis is 89.3%, whereas the survival rate with metastasis is 30%.[1,4-6] Hence, it is important to predict the malignant potential of these tumors accurately for prognostication. As malignancy can develop years after removal of primary tumor, histological grading is the only means to predict the malignant potential, making the role of the pathologist crucial. Unlike most other tumors, no single histological parameter has been able to predict malignant potential of pheochromocytomas/paragnagliomas (PPGLs) accurately till date. Conventional histological indicators of high-grade tumor like mitosis, atypia, and necrosis alone have been proven to be unreliable. To overcome these difficulties and to make prognostication easier, a combination of multiple parameters (histological, biochemical, and immunohistochemical) have been tried, such as “Grading of Adrenal Pheochromocytoma and Paraganglioma” (GAPP) score and “Pheochromocytoma of Adrenal Gland Scaled (PASS) score”.[2,5-7]
PASS score, introduced by Thompson in 2002, uses 12 specific histological features each carrying one or two points, with 20 points altogether.[6] Although the PASS scoring system solved this prognostication dilemma to some extent, it has its own shortcomings. PASS score is not applicable on PGLs and has high inter-observer and even intra-observer variation, which decreases its reproducibility.[8] To overcome these drawbacks, GAPP score was developed by Kimura et al. in 2014,[7] which also incorporates immunohistochemistry and biochemical parameters along with four histological parameters. Although GAPP score showed better predictability of malignant potential, till date, very few studies have been conducted to validate the GAPP system, making its clinical use limited.[9] Moreover, SDHB mutation was not included in GAPP scoring, which we now know is strongly associated with metastasis.[10,11]
In this study, we tried to validate the risk stratification of adrenal and extra-adrenal paragangliomas using GAPP and PASS scoring systems. We also tried to analyze the association of individual GAPP/PASS parameters with adverse disease outcome. Correlation of SDHB expression with clinical outcome of PPGLs was also evaluated.
MATERIALS AND METHODS
This was a prospective single-institutional study and was approved by our Institutional Ethical Committee. Patients of any age undergoing PPGL removal in our institution, over the study period of 48 months, were included. Altogether, 30 patients were included in the study period. Institutional ethical clearance was obtained.
After taking informed consent, the chief complaint of the patient (hypertension, headache, palpitation, weakness, syncope, etc.), duration, thorough medical and surgical history, and family history of such tumors (to consider or negate syndromic association) were elucidated. Clinical and laboratory findings, specifically related to excess catecholamines and its metabolite, were recorded in a proforma. Current guidelines recommend testing of plasma-free metanephrines or urinary fractionated metanephrines initially. Computed tomography/magnetic resonance (CT/MR) imaging plates were scrutinized for sign of necrosis, hemorrhage, cystic changes, or calcification, which indicates increased malignant potential. Positron emission tomography (PET) CT scan was also done in cases of suspected metastasis. Patients were followed up periodically for any adverse disease outcome over the study period.
Physical parameters like diameter and weight of the surgically removed specimen were measured. Hematoxylin and eosin (H and E)-stained slides were prepared, and slides were assessed for GAPP/PASS parameters and scored accordingly. In PH cases, both GAPP and PASS score were determined, and for PGL cases, only GAPP was accessed.
PASS score incorporated 12 histological parameters each bearing a score of 1 to 2. The histological features of “large cell/diffuse growth”, “central tumor necrosis”, “high cellularity”, “cellular monotony”, “cell spindling”, “mitotic figures (>3/HPF)”, “atypical mitotic figures”, and “extension into adipose tissue” bear 2 points each and “vascular invasion” and “capsular invasion” bear 1 point each. A consolidated score of >=4 is considered high-grade PPGL conventionally.
On the other hand, GAPP score incorporates four histological, one IHC, and one biochemical parameter as follows:
Histological pattern (Zellballen-0, large/irregular cell nest-1, pseudorosette-1), cellularity (low {<150 cells/HPF}-0, moderate {150–250 cells/HPF}-1, high {more than 250 cells/HPF}-2), comedo necrosis (absence-0, present-1), Ki-67 labeling index (%) (<1%-0, 1% to 3%-1, >3%-2), and catecholamine type (epinephrine-0, norepinephrine-1, non-functioning-0).
A PPGL bearing a score of 0–2 qualifies for a well-differentiated category, 3–6 qualifies for moderately differentiated, and 7–10 denotes a poorly differentiated tumor.
Finally, correlation between clinical outcome, histopathological features, and IHC findings were analyzed and GAPP/PASS score risk stratifications were tallied with disease course.
IHC for synaptophysin (PATHNSITU Synaptophysin rabbit monoclonal antibody, code-GR007), chromogranin (PATHNSITU chromogranin A rabbit monoclonal antibody, code-EP38), S100 (PATHNSITU S100 rabbit monoclonal antibody, code-EP32), Ki-67, and SDHB (ZETA rabbit monoclonal antibody anti-human SDHB, Code-EP288) were done on formalin-fixed paraffin-embedded (FFPE) tissue blocks. Whereas interpretation of synaptophysin, chromogranin and S100 was straight forward, SDHB IHC interpretation was cumbersome and was done as follows:[12]
- SDHB:Positive: Positive tumor cells show granular cytoplasmic staining in the same intensity as the positive internal control (endothelial cells, sustentacular cells, lymphocytes).
- Negative: Negative cells show completely absent of staining in the presence of positive internal control.
- Weak diffuse: Tumor cells show cytoplasmic blush without definite granularity contrasting the strong granular staining of positive internal control. In our study, we considered ‘weak diffuse’ as negative.
- Heterogenous: Cells show definite cytoplasmic granularity with a cytoplasmic blush.
- Non-informative: Completely absent staining without positive internal control.
Statistical analysis plan
Numerical variables were compared between groups using the unpaired t-test, and categorical variables were compared between groups using Fisher’s exact test. For each marker, the standard diagnostic indices were calculated, including sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Key findings have been presented with 95% confidence interval (CI). All the statistical analyses were done using MedCalc version 15.8. A P value of <0.05 was considered to be statistically significant.
RESULTS
During our study period of 48 months, a total of 30 PPGL cases were studied, of which 17 were male (56.67%) and 13 were female (43.3%) patients. Their ages ranged from 7 to 61 years with the mean being 37.33 years. The majority of the patients, 22, had PH (63.6%), and eight patients (36.4%) had PGL. The mean duration of symptoms was 5.2 months with the range of 1–12 months. Most common presenting symptoms were hypertension (46.7%), followed by headache, syncope, and abdominal pain. Non-functional PPGL cases presented with other non-specific symptoms and symptoms due to mass effect. One PH patient had known VHL syndrome with metachronous bilateral involvement of adrenals. The clinicopathological profile of the patients is listed in Table 1.
;)
Table 1: Baseline clinico-pathological profile of the patients
Patients were followed up over a period of 31–48 months. Three patients, 2 PH and 1 PGL, eventually showed metastasis, all metachronous. One malignant PH case with a PASS score of 16 and a GAPP score of 7 metastasized to liver and bone, and the patient succumbed to the disease 16 months after initial surgery. Another malignant PH case with a PASS score of 9 and a GAPP score of 6 developed liver metastasis, but the patient survived till the end of study with a follow-up period of 31 months. The malignant PGL case with GAPP score 6 involved the lungs, and the patient survived till the end of our study with a follow-up period of 34 months. The survival analysis (Kaplan–Meier curve) of these cases is shown in Figure 1. The rest of the patients achieved complete remission after surgery and showed no adverse outcome over the period of observation. These patients were considered benign in our study.
Figure 1: Kaplan-Meier survival analysis curve of all pheochromocytoma/paraganglioma cases
The most common site of PGL was the paravertebral region (5 out of 8 PGL cases). A right sided predilection was noted in PH cases. Of the 22 cases, 13 were right-sided, 7 were left-sided, and 2 cases, including the one with VHL syndrome, were bilateral.
The size of the PPGLs ranged from 0.4 to 20 cm with a mean size of 6.39 cm. The mean size of benign PPGLs was 5.7 cm, whereas the mean size of malignant PPGLs was 16 cm. The weight of the tumors ranged from 0.5 to 315.6 g with a mean of 46.78 g. The mean weight of benign PPGLs was 34.62 g, whereas malignant PPGLs weighed 217 g [Table 1].
The PASS scoring system was applied, and 15 out of 22 cases were categorized as high-risk category (score ≥4). The PASS score was ≥4 in 13 out of 20 benign PH cases; the rest of the 7 benign cases showed a PASS score of <4. Two malignant PH cases also showed a score of ≥4. In PH, the sensitivity of the PASS SCORE was 100% and the specificity was 35%. The PPV was 13.3%; the NPV was 100%.
One malignant PH case (PASS score of 16) showed all the features of the PASS scoring system, except for ‘cellular monotony’ and ‘periadrenal adipose tissue invasion’. Another malignant PH (PASS score of 9) showed features of ‘mitotic figure >3/10 HPF’, ‘capsular invasion’, ‘vascular invasion’, ‘central tumor necrosis’, ‘large nests/diffuse growth pattern >10%’, and ‘nuclear pleomorphism’. Out of 20 benign PH cases, six cases showed ‘large cell nest/diffuse growth pattern >10%’. Eight benign cases showed ‘confluent necrosis’. None showed ‘high cellularity’. Only one case showed “cell monotony”. None of the benign PH showed ‘tumor cell spindling’ (P value 0.001). Six cases showed ‘mitotic figure >3/10 HPF’. ‘Atypical mitotic figure’ was seen in only one case. ‘Extension into adipose tissue’ was not found in any cases. ‘Vascular invasion’ was detected in 14 cases. Eighteen cases ‘showed capsular invasion’. Ten cases showed ‘profound nuclear pleomorphism’. ‘Nuclear hyperchromasia’ was seen in six cases [Table 2].
Table 2: Assessment of individual parameters using PASS scoring system (n=22)
Twelve benign PPGL cases showed a GAPP score of ≥3 (3–6) and belonged to “intermediate grade”. Fifteen PPGL showed a GAPP score of <3 (“low grade”). Two malignant PPGL (1 PH and 1 PGL) cases showed a score of ≥3 or more specifically >6 (“high grade”). One PH case with a GAPP score of 6 metastasized. The sensitivity of GAPP score was 100%. The specificity of GAPP score was 55.6%. The PPV was 20%. The NPV was 100%. In our study, one malignant PH case (PASS score 16, GAPP score 7) showed ‘large irregular nests’, ‘high cellularity’, presence of ‘vascular/capsular invasion’, ‘secretion of adrenaline’, and a ‘Ki-67 labelling index’ of >3%. Another malignant PH (PASS score 9, GAPP score 6) showed all the previously mentioned features except for having ‘intermediate cellularity’. The malignant PGL cases showed ‘large irregular nest’, ‘intermediate cellularity’, presence of ‘vascular/capsular invasion’, ‘non-adrenaline’ secretion, and a ki67 index of >3%. Out of 27 benign PPGL cases, 13 cases showed a zellballen pattern, 14 showed large irregular nests, and none showed a pseudorossette pattern. Ten cases showed low cellularity, 17 cases showed intermediate cellularity, and none showed high cellularity. Vascular invasion was present in 17 cases; the rest of the 10 cases did not have any vascular invasion. Fourteen PPGLs were noradrenaline-secreting, four were adrenaline-secreting, and nine were non-functioning. Fifteen cases scored 1–3% in Ki-67 IHC, 12 cases scored <3%, and none scored <1% [Table 3].
Table 3: Assessment of individual parameters using GAPP scoring system (n=30)
IHC for SDHB was done in only 15 cases. Among them, 12 were PH cases and three were PGL case. Of the 12 PH cases, two cases show loss of SDHB, of which one case metastasized. Of the three PGL cases, one showed SDHB loss and turned out to be malignant later. None were ‘weak positive’ or ‘heterogeneous’. The rest of the 10 cases showed retained SDHB staining [Figures 2 and 3].
Figure 2: (a) Gross of pheochromocytoma (b) Low magnification of pheochromocytoma showing typical zellballen pattern (100× H and E) (c) High power view of pheochromocytoma with zellballen pattern (400× H and E) (d) Benign pheochromocytoma showing stromal hyalinisation (100× H and E) (e) Benign pheochromocytoma showing fibrosis (100× H and E) (f) IHC of a benign pheochromocytoma showing retained SDHB expression (100×)
Figure 3: (a) Low magnification of a malignant PPGL showing loss of zellballen pattern (100× H and E) (b) Low magnification of a malignant PPGL showing hypercellularity, tumor cell spindling (c) Low magnification showing necrosis in malignant PPGL (100× H and E) (d) Low magnification showing capsular invasion in a malignant PPGL (100× H and E) (e) High magnification of a PPGL case showing vascular invasion (400× H and E) (f) Immunohistochemistry of a malignant PPGL showing loss of SDHB expression (100×)
DISCUSSION
PPGL is characterized by a wide range of clinical presentation and genetic profiling. Surgical resection is the mainstay of treatment till now with satisfactory outcome if the surgical margin is clean and evidence of metastasis at the time of surgery is absent. Presence of metastasis, which may take up to 20 years to appear, drastically affects the survival.[13,14] To be able to pinpoint cases with future risk of malignancy has a direct clinical impact, and therefore, efforts had been made to pinpoint such cases histologically. To standardize the histological diagnostic approach, GAPP and PASS scores were introduced, which reduces the interobserver variation of disease prediction to some extent. Both these scoring systems have their own pros and cons. PASS system takes only the histological parameters into account and has high interobserver variability. Its applicability is also limited to PH only. Whereas GAPP system is relatively new, there are very few studies to validate its reproducibility. WHO 2022 also did not include GAPP and PASS scoring systems in the tumor classification.[14] More recently, SDHB, carbonic anhydrase9 (CA9), and S100 have been found to predict the malignant potential of PPGL, which have not yet been included in either scoring systems.
In our study, the majority of the PPGL cases comprised PH and most of them (70%) were functional (catecholamine secreting), leading to the common symptom of hypertensive episodes. The patient with syndromic association presented with disease at a much earlier age of 7 years. As indicated in several studies, the syndromic patient also presented with bilateral PH.[10] No gender predilection in either benign or malignant PPGL was noted.
For reasons unknown, PH was found to affect right adrenal more, a finding consistent with other clinical series.[15,16] Increased weight was found to be associated with histologically malignant tumor. Although association of increased tumor weight with poor prognosis is well known, there is not enough evidence in the literature to associate a large tumour size with poor outcome.[17]
The true incidence of metastasis is hard to determine, but according to the literature, approximately 10% pheochromocytoma, approximately 25% paraganglioma, and about 40% of sympathetic paraganglioma metastasize in a lifetime.[5,18] In our study 9% PH and 12.5% PGL metastasized in 37 months follow-up. The lower-than-expected rate of malignancy can be attributed to the short follow-up period.
After application of PASS score, 15 PH cases were categorized as high risk. Surprisingly, only two cases turned out to be malignant, making the specificity of the scoring system high and sensitivity low. Cases categorized as low risk (7 out of 22 cases) had uneventful remission. PASS system had a high NPV of 100%, but PPV was low, 13.3%. Therefore, the true value of PASS system lies in its ability to segregate cases with exceptionally low risk of malignancy rather than to identify cases with a high chance of future dissemination. However, a low specificity score can again be attributable to the short observation period as malignancy may develop years after the resection of the primary tumor.[12]
We also weighed each of the PASS parameters and tried to find out the most relevant parameter in predicting malignancy. ‘High cellularity’, ‘tumor cell spindling’, and ‘extension into adipose tissue’ were most relevant in this respect, followed by ‘cellular monotony’, ‘atypical mitotic figure’. ‘capsular invasion’, ‘vascular invasion’, ‘nuclear pleomorphism’, and ‘nuclear hyperchromasia’, which were weakly associated with malignancy. These findings corroborate the PASS scoring system, where these four weakly associated categories bear 1 score each. The only exception is ‘confluent necrosis’, which bears 2 points but is found to be weakly associated with malignancy in our study.[6] Regardless, it is important to mention that no single parameters should be overemphasized.
After application of GAPP score, 9 PPGL cases were categorized as ‘low risk’ (score 0–2) and 19 cases were categorized as ‘intermediate risk’ category. One PH case belonging to ‘intermediate category’ showed metastasis during the follow-up period. Two PPGL cases were categorized ‘high risk’ with a score of >=7 and eventually metastasized. GAPP score showed 100% sensitivity. The specificity was 45% and 50% for PH and PG, respectively. The PPV was 15.4% in PH cases and 25% in PGL cases. The NPV was 100%. This result is almost similar to that of PASS score, and like PASS score, GAPP score is better at ruling out rather than ruling in malignant potential.
Both GAPP and PASS systems showed high NPV and low PPV, which is appreciable in the sense that for a scoring system, it is important not to under-call a tumor with malignant potential which may deprive the patient of regular follow-ups with disastrous consequence. Out of all GAPP parameters, ‘cellularity’ is most strongly associated with malignancy, which is similar to PASS system. The specificity of GAPP is slightly higher than that of PASS score, supporting the previous notion of PASS system’s high inter-observer variability and the fact that GAPP score uses one biochemical and one immunohistochemical parameter along with other histologic parameters. Incorporating other molecular and IHC parameters like SDHB may further increase the specificity of this scoring system like composite pheochromocytoma/paraganglioma prognostic score (COPPS), modified GAPP (M-GAPP), age, size, extra-adrenal location, and secretory type (AGES) score.[19] We performed SDHB IHC on 15 cases, out of which three cases showed loss of SDHB. Out of those three cases, two cases turned out to be malignant. The other one case with loss of SDHB expression continued a benign course till the end of observation period. So, it can be concluded that addition of SDHB IHC to the existing system will increase its specificity, thereby reducing unnecessary follow-up and screening for metastasis in a potentially benign PPGL case.[20] A few studies also recommend inclusion of sustentacular cell counts by S100 to the current algorithms.[21] Probably in future, we will see more such objective parameters being added in the scoring system.
CONCLUSION
We feel that GAPP and PASS scores can pinpoint cases with exceedingly low risk of future metastasis, thereby reserving the more aggressive treatment approach for those with high risk of dissemination. Degree of predictability can be further enhanced by adding more ‘objective’ parameters like SDHB and S100, which can also increase the reproducibility of the systems.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
REFERENCES
1. Chrisoulidou A, Kaltsas G, Ilias I, Grossman AB. The diagnosis and management of malignant Pheochromocytoma and Paraganglioma. Endocr Relat Cancer 2007;14:569–85.
2. Tischler AS, De Krijger RR, Gill A, Kawashima A, Kimura N, Komminoth P, et al. Phaeochromocytoma Lloyd RV, Osamura RY, Kioppel G, Rosai J. WHO Classification of Tumours of Endocrine Organs. 4th ed. Lyon:IARC Press;2017 183–9.
3. Plouin PF, Fitzgerald P, Rich T, Ayala-Ramirez M, Perrier ND, Baudin E, et al. Metastatic pheochromocytoma and paraganglioma:Focus on therapeutics. Horm Metab Res 2012;44:390–9.
4. Trimmers HJ, Brouwers FM, Hernus AR, Sweep FC, Verhofstad AA, Verbeek AL, et al. Metastases but not cardiovascular mortality reduces life expectancy following surgical resection of apparently benign Pheochromocytoma. Endocr Relat Cancer 2008;15:1127–33.
5. Turkova H, Prodanov T, Maly M, Martucci V, Adams K, Widimsky J Jr, et al. Characteristics and outcomes of metastatic SDHB and sporadic pheochromocytoma/paraganglioma:An National Institutes of Health Study. Endocr Pract 2016;22:302–14.
6. Thompson LD. Pheochromocytoma of the Adrenal gland Scaled Score (PASS) to separate benign from malignant neoplasms:A clinicopathologic and immunophenotypic study of 100 cases. Am J Surg Pathol 2002;26:551–66.
7. Kimura N, Takayanagi R, Takizawa N, Itagaki E, Katabami T, Kakoi N, et al. Phaeochromocytoma Study Group in Japan. Pathological grading for predicting metastasis in phaeochromocytoma and paraganglioma. Endocr Relat Cancer 2014;21:405–14.
8. Wu D, Tischler AS, Lloyd RV, DeLellis RA, De Krijger R, Van Nederveen F, et al. Observer variation in the application of Pheochromocytoma of the Adrenal Gland Scaled Score. Am J Surg Pathol 2009;33:599–608.
9. Medeiros LJ, Wolf BC, Balogh K, Federman M. Adrenal pheochromocytoma:A clinicopathologic review of 60 cases. Hum Pathol 1985;16:580–9.
10. Modlin IM, Farndon JR, Shepherd A, Johnston ID, Kennedy TL, Montgomery DA, et al. Phaeochromocytomas in 72 patients:Clinical and diagnostic features, treatment and long term results. Br J Surg 1979;66:456–65.
11. Agarwal A, Mehrotra PK, Jain M, Gupta SK, Mishra A, Chand G, et al. Size of the tumor and pheochromocytoma of the adrenal gland scaled score (PASS):Can they predict malignancy?. World J Surg 2010;34:3022–8.
12. Ding CC, Chan S, Mak J, Umetsu SE, Simko JP, Ruiz-Cordero R, et al. An exploration in pitfalls in interpreting SDHB immunohistochemistry. Histopathology 2022;81:264–9.
13. Fishbein L, Ben-Maimon S, Keefe S, Cengel K, Pryma DA, Loaiza-Bonilla A, et al.
SDHB mutation carriers with malignant pheochromocytoma respond better to CVD. Endocr Relat Cancer 2017;24:L51–5.
14. Hamidi O, Young WF Jr, Gruber L, Smestad J, Yan Q, Ponce OJ, et al. Outcomes of patients with metastatic phaeochromocytoma and paraganglioma:A systematic review and meta-analysis. Clin Endocrinol (Oxf) 2017;87:440–50.
15. Papathomas TG, Oudijk L, Persu A, Gill AJ, van Nederveen F, Tischler AS, et al. SDHB/SDHA immunohistochemistry in pheochromocytomas and paragangliomas:A multicenter interobserver variation analysis using virtual microscopy:A Multinational Study of the European Network for the Study of Adrenal Tumors (ENS@T). Mod Pathol 2015;28:807–21.
16. van Nederveen FH, Gaal J, Favier J, Korpershoek E, Oldenburg RA, de Bruyn EM, et al. An immunohistochemical procedure to detect patients with paraganglioma and phaeochromocytoma with germline SDHB, SDHC, or SDHD gene mutations:A retrospective and prospective analysis. Lancet Oncol 2009;10:764–71.
17. Lewi HJ, Reid R, Mucci B, Davidson JK, Kyle KF, Macpherson SG, et al. Malignant phaeochromocytoma. Br J Urol 1985;57:394–8.
18. Zeigler RG, Weinstein JA, Fears TR. Nutritional and genetic inefficiencies in one-carbon metabolism and cervical cancer risk. J Nutr 2002;132 8 Suppl:2345S–9S.
19. Pierre C, Agopiantz M, Brunaud L, Battaglia-Hsu SF, Max A, Pouget C, et al. COPPS, a composite score integrating pathological features, PS100 and SDHB losses, predicts the risk of metastasis and progression-free survival in pheochromocytomas/paragangliomas. Virchows Arch 2019;474:721–34.
20. Choi HR, Koo JS, Lee CR, Lee JD, Kang SW, Jo YS, et al. Efficacy of immunohistochemistry for SDHB in the Screening of Hereditary Pheochromocytoma-Paraganglioma. Biology (Basel) 2021;10:677.
21. Unger P, Hoffman K, Pertsemlidis D, Thung S, Wolfe D, Kaneko M. S100 protein-positive sustentacular cells in malignant and locally aggressive adrenal pheochromocytomas. Arch Pathol Lab Med 1991;115:484–7.
