J Cutan Pathol 2014: 41: 859–865 doi: 10.1111/cup.12396 John Wiley & Sons. Printed in Singapore
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Journal of Cutaneous Pathology
Fumarate hydratase immunohistochemical staining may help to identify patients with multiple cutaneous and uterine leiomyomatosis (MCUL) and hereditary leiomyomatosis and renal cell cancer (HLRCC) syndrome Aims: Multiple cutaneous and uterine leiomyomatosis (MCUL) also named as hereditary leiomyomatosis and renal cancer syndrome (HLRCC) is an autosomal dominant disorder caused by heterozygotic germline mutations in fumarate hydratase (FH) with incomplete penetrance and clinically challenging to diagnose. To test immunohistochemistry for FH as a potential marker for the detection of FH-deficiency. Methods and results: We have tested 42 smooth muscle neoplasms, 13 lesions of patients with suspicious or confirmed HLRCC, 20 sporadic piloleiomyomas, two angioleiomyomas and 7 leiomyosarcomas. FH staining grades from 1 to 3. Ten of the 13 lesions from the patients with HLRCC syndrome showed negative FH staining. Most sporadic piloleiomyomas presented grade 3 FH staining although five cases presented grade 1 FH staining. Sensitivity of FH staining in our series is 83.3% but specificity is 75%. Conclusions: This staining could indicate a high risk of HLRCC in most of the confirmed cases but it could also suggest the presence of a syndrome in up to 25% of sporadic cases. HLRCC syndrome should be rule out in FH negative piloleiomyomas after complete anamnesis if multiple lesions or positive familiar history is found.
M. Llamas-Velasco1 , L. Requena2 , H. Kutzner3 , L. Schärer3 , A. Rütten3 , M. Hantschke3 , B. E. Paredes3 and T. Mentzel3 1
Department of Dermatology, Hospital Infanta Cristina, Madrid, Spain, 2 Department of Dermatology, Fundación Jimenez Diaz, Madrid, Spain, and 3 Dermatopathologie Friedrichshafen, Friedrichshafen, Germany
Keywords: fumarase, immunocytochemistry, piloleiomyoma Llamas-Velasco M, Requena L, Kutzner H, Schärer L, Rütten A, Hantschke M, Paredes BE, Mentzel T. Fumarate hydratase immunohistochemical staining may help to identify patients with multiple cutaneous and uterine leiomyomatosis (MCUL) and hereditary leiomyomatosis and renal cell cancer (HLRCC) syndrome. J Cutan Pathol 2014; 41: 859–865. © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Mar Llamas-Velasco C/Avenida 9 de Junio, CP 28981, Parla, Spain Tel: +34911913000 Fax: +034915202435 e-mail: [email protected] Accepted for publication October 5, 2014
859
Llamas-Velasco et al. Cutaneous leiomyomas may be classified into piloleiomyoma (differentiating toward the arrector pili muscle), genital and nipple leiomyoma (differentiating toward the dartos and areolar smooth muscle) and angioleiomyoma (differentiating toward the muscular layer of vessel walls). Piloleiomyoma is the most common type. Autosomal dominant inheritance pattern in cutaneous piloleiomyomas was first described by Kloepfer et al. in 1958, and later, in 1973, Reed et al., described two families with cutaneous and uterine leiomyomas in different family members over three generations.1 – 3 Multiple cutaneous piloleiomyomas appear as smooth dome-shaped papules or nodules, ranging in size from 1 to 20 mm in diameter, which progressively increase in number and frequently group in clusters in an anatomical region or are arranged in a linear fashion. Cutaneous lesions often produce a sharp and shooting pain after physical stimuli. The cause of this pain is unknown, but cutaneous nerve compression and ischemic phenomena have been proposed as pathogenic mechanisms to explain it. Female patients with multiple cutaneous pilioleiomyomas often associate symptomatic uterine leiomyomas. Reed syndrome presents multiple cutaneous and uterine leiomyomatosis (MCUL), also having been named leiomyomatosis cutis et uteri (OMIM 150800). It has been found that up to 14% of MCUL’s patients present papillary type 2 renal cancer, with some authors having named this disorder ‘hereditary leiomyomatosis and renal cell cancer syndrome’ (HLRCC syndrome, OMIM 605839).4 HLRCC syndrome is an autosomal dominant hereditary disorder caused by heterozygotic germline mutations in a gene coding for fumarase (1q42.3-43). The mitochondrial isoform of fumarase, also known as fumarate hydratase, catalyzes the conversion of fumarate to malate. Fumarate hydratase mutations may be less frequently associated with Leydig cell tumors,5 ovarian mucinous cystadenoma and cerebral cavernoma.6 Fumarate hydratase mutations appear to cause tumor growth through the activation of the hypoxia pathway,5,6 although there are other proposed hypotheses to explain its role as a tumor suppressor, including the presence of hypermutability as a result from oxidative damage.7 The enzyme may also act as a competitive inhibitor of enzymes that use 2-oxoglutarate as co-substrate, such as histone demethylases,8,9 or it may alter functional characteristics of proteins.10 In any event, penetrance is incomplete in HLRCC syndrome and many cases are poorly symptomatic, making
860
the diagnosis challenging. Succinate dehydrogenase mutations, another enzyme with a role in oxidative phosphorylation that also behaves as a tumor suppressor,6 causes hereditary paragangliomas and pheochromocytomas (HPGL syndrome). Despite the closely related function of both enzymes, the tumor spectrum in both HPGL and HLRCC syndromes is entirely different, indicating that the pathway leading to tumor histogenesis in both disorders must be cell specific.6 Often, most families with HLRCC syndrome are not further evaluated because an easy and inexpensive diagnostic test to identify affected individuals is not yet available. In this sense, we have immunohistochemically studied 42 cutaneous smooth muscle neoplasms with an anti-fumarate hydratase antibody in order to determine whether immunohistochemistry investigation for fumarate hydratase is a helpful marker for detecting fumarate hydratase-deficiency in lesional skin of patients with familial and sporadic smooth muscle neoplasms. Material and methods We retrieved 13 cutaneous piloleiomyomas from 11 patients with multiple cutaneous leiomyomas, 10 of them with highly suspicious or genetically confirmed HLRCC syndrome (we studied three lesions of one confirmed case) and one with sporadic multiple piloleiomyomas: 20 sporadic piloleiomyomas, 2 angioleiomyomas and 7 cutaneous leiomyosarcomas from the files of the authors (TM, HK, LS, LR). Cases dated from 2005 to 2013. Cases in which the tissue block or necessary stains were not available for additional evaluation were excluded from the study. Clinical and follow-up information as well as information about genetic findings were provided by the referring clinicians and pathologists. We consider all cases with single skin lesions and without known suspicious personal history to be sporadic occurrences of piloleiomyoma. We consider all cases of multiple piloleiomyomas as suspicious, and we ask the referring clinicians for personal and familiar history to exclude sporadic multiple piloleiomyomas. A case of HLRCC is considered highly suspicious when more than two piloleiomyomas are present plus one of the following data: personal history of kidney cancer or uterine myomas or family history of cutaneous piloleiomyomas, uterine myomas (especially in young women) or kidney cancer.
Fumarate hydratase and MCUL/HLRCC
Fig. 1. Clinical pictures of multiple piloleiomyomas. A) Smooth hemispheric papules of 1–5 mm grouped in left scapular area. B) Monomorphous pinkish papules located in left flank. C,D) Erythematous, well-defined, hemispheric to lineal papules located in right shoulder and left scapular area. E) More dispersed papules and slightly pinkish plaques are noted.
Cellular atypia was graded as absent, mild or moderate. Mild atypia was recorded when we found that few atypical cells were within the lesion, while moderate atypia was recorded if less than 10% of the nuclei of the cells showed atypical findings. Leiomyosarcomas in our study were graded according to the French Federation of Cancer Centers Sarcoma Group.11 Immunohistochemistry Four micrometersections were mounted onto positive charged slides. Sections were later dried overnight at 45∘ C. Slides were deparaffinized in xylene for 30 min, rehydrated using graded ethanol concentrations and incubated for 30 min at 95∘ C in ethylenediaminetetraacetic acid (EDTA) buffer (pH = 9.0). They were cooled to room temperature for 20 min. After quenching with alkaline phosphatase and biotin blocking using avidin, sections were incubated with a 1 : 200 dilution of the fumarate hydratase in phosphate-buffered saline (J-13 fumarate hydratase, Santa Cruz Biotechnology, Santa Cruz, CA, USA). This antibody is a mouse monoclonal antibody raised against recombinant
fumarate hydratase of human origin. Automatic staining was performed by a TechMate 500 (Biotech Solutions, Dako, Glostrup, Denmark) as a detection system and labeled with streptavidin-biotin (LSAB). Scoring of fumarate hydratase staining was determined using a semiquantitative system. Negative staining was considered when tumors presented no or minimal fumarate hydratase staining. Strong positive staining was considered when tumors presented strong cytoplasmic fumarate hydratase staining in most neoplastic cells. Weak positive staining was considered in cases where some degree of cytoplasmic staining was found, but with a weaker intensity than the internal positive controls (pre-existing arrector pili muscle fibers). All the cases were scored independently by two dermatopathologists (TM and MLV). Results Most patients with MCUL/HLRCC syndrome presented cutaneous lesions located on the trunk or upper extremities and one single case presented lesions located on the right leg (Fig. 1). HLRCC syndrome occurred in five
861
Llamas-Velasco et al.
Fig. 2. Histopathologic pictures of different fumarate hydratase staining grade. A–D show MCUL/HLRCC piloleiomyoma cases showing a) low and b) high power image of conventional hematoxylin/eosin staining as well as c) low and d) high power image of grade 1 fumarate hydratase staining, that is a fumarate hydratase negative neoplasm with epidermal positive control. E–H) show another MCUL/HLRCC case showing e) low and f) high power image of conventional hematoxylin/eosin staining and g) low and h) high power of grade 2 fumarate hydratase staining. Panels H–K show a i) low and j) high power image of conventional hematoxylin/eosin staining as well as k) low and l) high power image of grade 3 fumarate hydratase staining. Piloleiomyoma staining intensity is the same as the epidermis.
women and five men, with a median age of 41.0 years old (29–55). The median tumor-size was 1.07 cm (0.5–2 cm). Some tumors were closely grouped involving regional skin areas, up to 14 × 9 cm as the maximum size. Histopathologically, cutaneous lesions of HLRCC syndrome consisted of interweaving fascicles involving the dermis. The fascicles were composed of spindle cells, with elongated central nuclei and few perinuclear vacuoles and eosinophilic cytoplasm. In some cases we found larger nuclei with slightly lobulated contours but without pseudoinclusions without multinucleate cells or mitotic figures. No necrosis or hemorrhage was found. In all cases but one, a slight to moderate degree of nuclear atypia was found, which consisted of few enlarged and hyperchromatic nuclei. No mitotic figures or areas of necrosis were seen. All cases but two were confined to the upper dermis. In the other two cases, lesions involved the full-thickness of the dermis and were close to subcutaneous tissue, but no involvement of the hypodermis was found. Of the 10 patients with HLRCC syndrome, 10 of the 12 lesions showed entirely negative fumarate hydratase staining (Fig. 2A–D). Two cases showed weak positive immunoreactivity for fumarate hydratase, but with lower intensity than the internal control (Fig. 2E–H). We got genetic information about fumarate hydratase mutations in 5 of 10 suspected HLRCC cases. We found five different mutations: C1292_1334dup; p.(Asn446Arg*20); c139C>T (pGln47stop) heterozygosis; c1157A>G p Q386R; cG13T>Gp.L138R and C667-669 delins TAG prot p.Lys 222.
862
Analysis of the clinical history of the only patient of our series with sporadic multiple piloleiomyomas revealed a 58-year-old male with progressive appearance of painful papules in the left scapular area and ipsilateral arm. There are no other family members affected. There is no personal or family history of kidney cancer or other fumarate hydratase-related neoplasms. Female members of the family did not present uterine myomas. We could not perform a genetic study in this case. Fumarate hydratase staining was strongly positive. Sporadic piloleiomyomas included in this study occurred in 8 women and in 12 men, with a median age of 59 years old (ages ranging from 30 to 78). The most frequently involved sites were the lower extremities,8 upper extremities,4 trunk2 and face,1 whereas in five cases the location was unknown. Median tumor-size was 1.1 cm (range from 0.4 to 2 cm). The histopathologic characteristics of these sporadic tumors were identical to those previously described in piloleiomyomas from patients with HLRCC syndrome, although only six cases with very subtle nuclear atypia was found in these cases. Most sporadic piloleiomyomas presented strong positive fumarate hydratase staining (Fig. 2H–K), although five cases presented negative fumarate hydratase staining. A table compiling the fumarate hydratase staining results as well as the sensitivity/specificity for every tested group is included (Table 1). Some clinical and histopathologic features of the patients with HLRCC syndrome and with sporadic solitary or multiple piloleiomyomas are summarized in Table 2.
Fumarate hydratase and MCUL/HLRCC Table 1. Compilation of fumarate hydratase staining results within the different studied groups, including sensitivity and specificity of fumarate hydratase staining in genetically confirmed HLRCC or all highly suspicious HLRCC cases
Negative fumarate staining Positive fumarate staining
Genetically confirmed HLRCC cases
All suspected HLRCC cases
Sporadic single/multiple piloleiomyomas
Angioleiomyomas
Leiomyosarcomas
4 1 (1 weak)
8 2 (2 weak)
5 16 (8 weak)
0 2 (1 weak)
0 7 (2 weak)
Only genetically confirmed
All suspected HLRCC cases
80% 76.2%
80% 65.4%
Sensitivity Specificity
HLRCC, hereditary leiomyomatosis and renal cell cancer. Table 2. Cases of HLRCC and sporadic piloleiomyomas
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
Gender/age
Type
Location
FH antibody
FH genetic study
Nuclear pleomorphism
M/55 M/29 F/47 F/48 F/30 F/40 M/58 M/37 F/53 M/35 M/36 F/75 M/66 F/78 M/75 F/56 M/46 M/64 M/43 F/53 M/76 M/61 M/60 F/37 F/49 M/57 F/48 M/30 M/46 M/53 F/75
HLRCC HLRCC HLRCC HLRCC HLRCC HLRCC Sp. mul. HLRCC HLRCC HLRCC HLRCC Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol.
Left arm/neck/interscapular area Right leg Right arm Left arm Right arm Right scapular area Left scapular area Thorax Left scapular area Right deltoid area Left pectoral area Right leg Chin Right shoulder Non-reported Left knee Right buttock Left knee Left knee Left back Non-reported Left arm Non-reported Non-reported Right arm Right leg Right shoulder Non-reported Right thigh Back Right lower leg
Negative Negative Negative Weak positive Negative Weak positive Strong Positive Negative Negative Negative Negative Weak positive Weak positive Negative Strong positive Strong positive Strong positive Weak positive Strong positive Weak positive Weak positive Negative Negative Negative Negative Weak positive Weak positive Weak positive Strong positive Negative Strong positive
C1292_1334dup; p.(Asn446Arg*20) Not found. c139C>T (pGln47stop) Not available c1157A>G p Q386R cG13T>Gp.L138R (CTC>CGC Not available Not available Not available Cc. 667-669 delin TAG prot p. Lys 223. Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available
Slight Slight Mild Slight Mild Slight Slight Slight None Slight Slight Slight None Slight None None None None None None None Slight None None Slight None None None None Slight Slight
M, male; F, female; FH, fumarate hydratase; HLRCC, hereditary leiomyomatosis and renal cell cancer; Sp. sol., sporadic solitary; Sp. mul., sporadic multiple.
We also evaluated seven well-differentiated cutaneous leiomyosarcomas, two of them infiltrating subcutaneous tissue. Five of those lesions consisted of grade 1 leiomyosarcomas, and the other two were grade 2. No areas of
massive necrosis of neoplastic cells were seen in these tumors. Five of the seven cutaneous leiomyosarcomas studied showed strong positive of fumarate hydratase staining, whereas weak positive immunostaining for fumarate
863
Llamas-Velasco et al.
Fig. 3. Grade I leiomyosarcoma (A) low power and (B) high power hematoxylin/eosin staining as well as (C,D) desminpositive staining and (E,F) fumarate hydratase grade 3 staining.
hydratase was seen in two cases (Fig. 3). The two angioleiomyomas studied showed respectively weak and strong positive fumarate hydratase immunostaining. Discussion Histopathologically, all piloleiomyomas included in this study were poorly demarcated neoplasms composed of interweaving fascicles of spindle cells, with elongated central nuclei and few perinuclear vacuoles, and an eosinophilic cytoplasm, which are the usual histopathologic characteristics of these neoplasms. In our experience, however, it seems to be that piloleiomyomas associated with suspected or confirmed HLRCC syndrome present a higher degree of nuclear atypia compared with sporadic piloleiomyomas, as we found a slight or mild degree of nuclear atypia in 90% of them. There were no other distinctive histopathologic features, as the large nuclei with a prominent inclusion-like eosinophilic nucleolus surrounded by a clear halo, as those described by Merino et al. in the renal tumors associated
864
with this syndrome,12 were found in the cutaneous piloleiomyomas of our study. In any case, slight nuclear atypia can be also be found in sporadic solitary leiomyomas. Cutaneous piloleiomyomas showed a higher penetrance among men, although women significantly developed more numerous lesions, without differences in their distribution.13 Clustered lesions were most common on the trunk, lower limbs, upper limbs, head and neck. Unilateral or bilateral symmetrically distributed lesions were infrequent.13 As we only included patients with cutaneous leiomyomas, gender differences of penetrance of skin lesions could not be further analyzed. Fumarate hydratase immunoreactivity is preserved in most sporadic piloleiomyomas, and thus although their molecular basis is not well known, our results suggest that they maintain fumarate hydratase activity. Up to 54% of the cutaneous leiomyosarcomas studied by Barker et al. showed losses in chromosome one
Fumarate hydratase and MCUL/HLRCC involving 1q42-4314 but all cutaneous leiomyosarcomas included in our study showed preserved fumarate hydratase immunoreactivity. We also found that immunohistochemical positive expression for fumarate hydratase is usually preserved in sporadic angioleiomyomas. A peculiar finding of our study consisted of the presence of well-preserved fumarate hydratase immunoexpression in the smooth muscle fibers of the pre-existing arrector pili muscle seen in some of the biopsies of the piloleiomyomas from patients with suspected or confirmed HLRCC syndrome, whereas it was clearly decreased or absent in neoplastic smooth muscle cells. Because patients with HLRCC syndrome present a germinal heterozygotous mutation, this finding supports the idea that an additional loss of heterozygosity is necessary in order to elicit tumor growth in piloleiomyomas of these patients. Bardella et al. proposed a polyclonal S-(2-succinyl) cysteine (2SC) antibody to immunohistochemically identify an increase in succinated proteins, as they show that immunohistochemical detection of fumarate hydratase was unreliable due to the fact that many fumarate hydratase mutations are missense changes, which lead to the production of a stable but inactive enzyme.10 This hypothesis, the production of a stable but inactive enzyme, could explain the weak positive staining found in two of our genetically confirmed cases of HLRCC syndrome, as they presented weaker staining than the sporadic piloleiomyomas. In conclusion, the results of fumarate hydratase immunostaining should be interpreted in a
clinicopathologic correlation context, because a weak positive result for fumarate hydratase immunostaining does not exclude a diagnosis of HLRCC syndrome. The main limitation of our study is that we lack additional clinical data to further classify the five cases of ‘suspected sporadic leiomyomas’ as truly sporadic leiomyomas, and we cannot completely rule out that some of that cases could be actually be HLRCC cases. The sensitivity and specificity of fumarate hydratase staining in our series therefore varies depending on how we define HLRCC cases on a clinical or on a genetic basis (Table 2). These results suggest that this staining could indicate a high risk of HLRCC in most of the cases but it could also suggest the presence of a syndrome in some of the sporadic cases. In summary, cutaneous leiomyomas associated with HLRCC syndrome may present some degree of nuclear atypia and usually show negative or weak positive immunostaining for fumarate hydratase. Therefore, as genetic testing is not easily available due to its high cost, and direct assay of intracellular fumarate levels is technically demanding, the immunohistochemical investigation for fumarate hydratase may provide a helpful tool for screening and stratifying the risk of HLRCC syndrome. Cases with strong fumarate hydratase staining can be excluded from further testing, while cases with negative or weak immunostaining may require further analysis regarding personal and family clinical history to classify risk of an underlying genetic syndrome, such as HLRCC.
References 1. Reed WB, Walker R, Horowitz R. Cutaneous leiomyomata with uterine leiomyomata. Acta Derm Venereol 1973; 53: 409. 2. Stewart L, Glenn GM, Stratton P, et al. Association of germline mutations in the fumarate hydratase gene and uterine fibroids in women with hereditary leiomyomatosis and renal cell cancer. Arch Dermatol 2008; 144: 1584. 3. Kloepfer HW, Krafchuk J, Derbes V, Burks J. Hereditary multiple leiomyoma of the skin. Am J Hum Genet 1958; 10: 48. 4. Alam NA, Olpin S, Rowan A, et al. Missense mutations in fumarate hydratase in multiple cutaneous and uterine leiomyomatosis and renal cell cancer. J Mol Diagn 2005; 7: 437. 5. Carvajal-Carmona LG, Alam NA, Pollard PJ, et al. Adult leydig cell tumors of the testis caused by germline fumarate hydratase mutations. J Clin Endocrinol Metab 2006; 91: 3071.
6. Bayley JP, Launonen V, Tomlinson IP. The fumarate hydratase mutation database: an online database of fumarate hydratase mutations involved in the MCUL (HLRCC) tumor syndrome and congenital fumarase deficiency. BMC Med Genet 2008; 9: 20. 7. Tomlinson IP, Alam NA, Rowan AJ, et al. Germline mutations in fumarate hydratase predispose to dominantly inherited uterine fibroids, skin leiomyomata and papillary renal cell cancer. Nat Genet 2002; 30: 406. 8. Pollard PJ, Ratcliffe PJ. Cancer. Puzzling patterns of predisposition. Science 2009; 324: 192. 9. Maxwell PH. Seeing the smoking gun: a sensitive and specific method to visualize loss of the tumour suppressor, fumarate hydratase, in human tissues. J Pathol 2011; 225: 1. 10. Bardella C, El-Bahrawy M, Frizzell N, et al. Aberrant succination of proteins in
11.
12.
13.
14.
fumarate hydratase-deficient mice and HLRCC patients is a robust biomarker of mutation status. J Pathol 2011; 225: 4. Coindre JM, Trojani M, Contesso G, et al. Reproducibility of a histopathologic grading system for adult soft tissue sarcoma. Cancer 1986; 58: 306. Merino MJ, Torres-Cabala C, Pinto P, Linehan WM. The morphologic spectrum of kidney tumors in hereditary leiomyomatosis and renal cell carcinoma (HLRCC) syndrome. Am J Surg Pathol 2007; 31: 1578. Alam NA, Barclay E, Rowan AJ, et al. Clinical features of multiple cutaneous and uterine leiomyomatosis: an underdiagnosed tumor syndrome. Arch Dermatol 2005; 141: 199. Barker KT, Bevan S, Wang R, et al. Low frequency of somatic mutations in the fumarate hydratase/multiple cutaneous leiomyomatosis gene in sporadic leiomyosarcomas and uterine leiomyomas. Br J Cancer 2002; 87: 446.
865
© 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Journal of Cutaneous Pathology
Fumarate hydratase immunohistochemical staining may help to identify patients with multiple cutaneous and uterine leiomyomatosis (MCUL) and hereditary leiomyomatosis and renal cell cancer (HLRCC) syndrome Aims: Multiple cutaneous and uterine leiomyomatosis (MCUL) also named as hereditary leiomyomatosis and renal cancer syndrome (HLRCC) is an autosomal dominant disorder caused by heterozygotic germline mutations in fumarate hydratase (FH) with incomplete penetrance and clinically challenging to diagnose. To test immunohistochemistry for FH as a potential marker for the detection of FH-deficiency. Methods and results: We have tested 42 smooth muscle neoplasms, 13 lesions of patients with suspicious or confirmed HLRCC, 20 sporadic piloleiomyomas, two angioleiomyomas and 7 leiomyosarcomas. FH staining grades from 1 to 3. Ten of the 13 lesions from the patients with HLRCC syndrome showed negative FH staining. Most sporadic piloleiomyomas presented grade 3 FH staining although five cases presented grade 1 FH staining. Sensitivity of FH staining in our series is 83.3% but specificity is 75%. Conclusions: This staining could indicate a high risk of HLRCC in most of the confirmed cases but it could also suggest the presence of a syndrome in up to 25% of sporadic cases. HLRCC syndrome should be rule out in FH negative piloleiomyomas after complete anamnesis if multiple lesions or positive familiar history is found.
M. Llamas-Velasco1 , L. Requena2 , H. Kutzner3 , L. Schärer3 , A. Rütten3 , M. Hantschke3 , B. E. Paredes3 and T. Mentzel3 1
Department of Dermatology, Hospital Infanta Cristina, Madrid, Spain, 2 Department of Dermatology, Fundación Jimenez Diaz, Madrid, Spain, and 3 Dermatopathologie Friedrichshafen, Friedrichshafen, Germany
Keywords: fumarase, immunocytochemistry, piloleiomyoma Llamas-Velasco M, Requena L, Kutzner H, Schärer L, Rütten A, Hantschke M, Paredes BE, Mentzel T. Fumarate hydratase immunohistochemical staining may help to identify patients with multiple cutaneous and uterine leiomyomatosis (MCUL) and hereditary leiomyomatosis and renal cell cancer (HLRCC) syndrome. J Cutan Pathol 2014; 41: 859–865. © 2014 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd
Mar Llamas-Velasco C/Avenida 9 de Junio, CP 28981, Parla, Spain Tel: +34911913000 Fax: +034915202435 e-mail: [email protected] Accepted for publication October 5, 2014
859
Llamas-Velasco et al. Cutaneous leiomyomas may be classified into piloleiomyoma (differentiating toward the arrector pili muscle), genital and nipple leiomyoma (differentiating toward the dartos and areolar smooth muscle) and angioleiomyoma (differentiating toward the muscular layer of vessel walls). Piloleiomyoma is the most common type. Autosomal dominant inheritance pattern in cutaneous piloleiomyomas was first described by Kloepfer et al. in 1958, and later, in 1973, Reed et al., described two families with cutaneous and uterine leiomyomas in different family members over three generations.1 – 3 Multiple cutaneous piloleiomyomas appear as smooth dome-shaped papules or nodules, ranging in size from 1 to 20 mm in diameter, which progressively increase in number and frequently group in clusters in an anatomical region or are arranged in a linear fashion. Cutaneous lesions often produce a sharp and shooting pain after physical stimuli. The cause of this pain is unknown, but cutaneous nerve compression and ischemic phenomena have been proposed as pathogenic mechanisms to explain it. Female patients with multiple cutaneous pilioleiomyomas often associate symptomatic uterine leiomyomas. Reed syndrome presents multiple cutaneous and uterine leiomyomatosis (MCUL), also having been named leiomyomatosis cutis et uteri (OMIM 150800). It has been found that up to 14% of MCUL’s patients present papillary type 2 renal cancer, with some authors having named this disorder ‘hereditary leiomyomatosis and renal cell cancer syndrome’ (HLRCC syndrome, OMIM 605839).4 HLRCC syndrome is an autosomal dominant hereditary disorder caused by heterozygotic germline mutations in a gene coding for fumarase (1q42.3-43). The mitochondrial isoform of fumarase, also known as fumarate hydratase, catalyzes the conversion of fumarate to malate. Fumarate hydratase mutations may be less frequently associated with Leydig cell tumors,5 ovarian mucinous cystadenoma and cerebral cavernoma.6 Fumarate hydratase mutations appear to cause tumor growth through the activation of the hypoxia pathway,5,6 although there are other proposed hypotheses to explain its role as a tumor suppressor, including the presence of hypermutability as a result from oxidative damage.7 The enzyme may also act as a competitive inhibitor of enzymes that use 2-oxoglutarate as co-substrate, such as histone demethylases,8,9 or it may alter functional characteristics of proteins.10 In any event, penetrance is incomplete in HLRCC syndrome and many cases are poorly symptomatic, making
860
the diagnosis challenging. Succinate dehydrogenase mutations, another enzyme with a role in oxidative phosphorylation that also behaves as a tumor suppressor,6 causes hereditary paragangliomas and pheochromocytomas (HPGL syndrome). Despite the closely related function of both enzymes, the tumor spectrum in both HPGL and HLRCC syndromes is entirely different, indicating that the pathway leading to tumor histogenesis in both disorders must be cell specific.6 Often, most families with HLRCC syndrome are not further evaluated because an easy and inexpensive diagnostic test to identify affected individuals is not yet available. In this sense, we have immunohistochemically studied 42 cutaneous smooth muscle neoplasms with an anti-fumarate hydratase antibody in order to determine whether immunohistochemistry investigation for fumarate hydratase is a helpful marker for detecting fumarate hydratase-deficiency in lesional skin of patients with familial and sporadic smooth muscle neoplasms. Material and methods We retrieved 13 cutaneous piloleiomyomas from 11 patients with multiple cutaneous leiomyomas, 10 of them with highly suspicious or genetically confirmed HLRCC syndrome (we studied three lesions of one confirmed case) and one with sporadic multiple piloleiomyomas: 20 sporadic piloleiomyomas, 2 angioleiomyomas and 7 cutaneous leiomyosarcomas from the files of the authors (TM, HK, LS, LR). Cases dated from 2005 to 2013. Cases in which the tissue block or necessary stains were not available for additional evaluation were excluded from the study. Clinical and follow-up information as well as information about genetic findings were provided by the referring clinicians and pathologists. We consider all cases with single skin lesions and without known suspicious personal history to be sporadic occurrences of piloleiomyoma. We consider all cases of multiple piloleiomyomas as suspicious, and we ask the referring clinicians for personal and familiar history to exclude sporadic multiple piloleiomyomas. A case of HLRCC is considered highly suspicious when more than two piloleiomyomas are present plus one of the following data: personal history of kidney cancer or uterine myomas or family history of cutaneous piloleiomyomas, uterine myomas (especially in young women) or kidney cancer.
Fumarate hydratase and MCUL/HLRCC
Fig. 1. Clinical pictures of multiple piloleiomyomas. A) Smooth hemispheric papules of 1–5 mm grouped in left scapular area. B) Monomorphous pinkish papules located in left flank. C,D) Erythematous, well-defined, hemispheric to lineal papules located in right shoulder and left scapular area. E) More dispersed papules and slightly pinkish plaques are noted.
Cellular atypia was graded as absent, mild or moderate. Mild atypia was recorded when we found that few atypical cells were within the lesion, while moderate atypia was recorded if less than 10% of the nuclei of the cells showed atypical findings. Leiomyosarcomas in our study were graded according to the French Federation of Cancer Centers Sarcoma Group.11 Immunohistochemistry Four micrometersections were mounted onto positive charged slides. Sections were later dried overnight at 45∘ C. Slides were deparaffinized in xylene for 30 min, rehydrated using graded ethanol concentrations and incubated for 30 min at 95∘ C in ethylenediaminetetraacetic acid (EDTA) buffer (pH = 9.0). They were cooled to room temperature for 20 min. After quenching with alkaline phosphatase and biotin blocking using avidin, sections were incubated with a 1 : 200 dilution of the fumarate hydratase in phosphate-buffered saline (J-13 fumarate hydratase, Santa Cruz Biotechnology, Santa Cruz, CA, USA). This antibody is a mouse monoclonal antibody raised against recombinant
fumarate hydratase of human origin. Automatic staining was performed by a TechMate 500 (Biotech Solutions, Dako, Glostrup, Denmark) as a detection system and labeled with streptavidin-biotin (LSAB). Scoring of fumarate hydratase staining was determined using a semiquantitative system. Negative staining was considered when tumors presented no or minimal fumarate hydratase staining. Strong positive staining was considered when tumors presented strong cytoplasmic fumarate hydratase staining in most neoplastic cells. Weak positive staining was considered in cases where some degree of cytoplasmic staining was found, but with a weaker intensity than the internal positive controls (pre-existing arrector pili muscle fibers). All the cases were scored independently by two dermatopathologists (TM and MLV). Results Most patients with MCUL/HLRCC syndrome presented cutaneous lesions located on the trunk or upper extremities and one single case presented lesions located on the right leg (Fig. 1). HLRCC syndrome occurred in five
861
Llamas-Velasco et al.
Fig. 2. Histopathologic pictures of different fumarate hydratase staining grade. A–D show MCUL/HLRCC piloleiomyoma cases showing a) low and b) high power image of conventional hematoxylin/eosin staining as well as c) low and d) high power image of grade 1 fumarate hydratase staining, that is a fumarate hydratase negative neoplasm with epidermal positive control. E–H) show another MCUL/HLRCC case showing e) low and f) high power image of conventional hematoxylin/eosin staining and g) low and h) high power of grade 2 fumarate hydratase staining. Panels H–K show a i) low and j) high power image of conventional hematoxylin/eosin staining as well as k) low and l) high power image of grade 3 fumarate hydratase staining. Piloleiomyoma staining intensity is the same as the epidermis.
women and five men, with a median age of 41.0 years old (29–55). The median tumor-size was 1.07 cm (0.5–2 cm). Some tumors were closely grouped involving regional skin areas, up to 14 × 9 cm as the maximum size. Histopathologically, cutaneous lesions of HLRCC syndrome consisted of interweaving fascicles involving the dermis. The fascicles were composed of spindle cells, with elongated central nuclei and few perinuclear vacuoles and eosinophilic cytoplasm. In some cases we found larger nuclei with slightly lobulated contours but without pseudoinclusions without multinucleate cells or mitotic figures. No necrosis or hemorrhage was found. In all cases but one, a slight to moderate degree of nuclear atypia was found, which consisted of few enlarged and hyperchromatic nuclei. No mitotic figures or areas of necrosis were seen. All cases but two were confined to the upper dermis. In the other two cases, lesions involved the full-thickness of the dermis and were close to subcutaneous tissue, but no involvement of the hypodermis was found. Of the 10 patients with HLRCC syndrome, 10 of the 12 lesions showed entirely negative fumarate hydratase staining (Fig. 2A–D). Two cases showed weak positive immunoreactivity for fumarate hydratase, but with lower intensity than the internal control (Fig. 2E–H). We got genetic information about fumarate hydratase mutations in 5 of 10 suspected HLRCC cases. We found five different mutations: C1292_1334dup; p.(Asn446Arg*20); c139C>T (pGln47stop) heterozygosis; c1157A>G p Q386R; cG13T>Gp.L138R and C667-669 delins TAG prot p.Lys 222.
862
Analysis of the clinical history of the only patient of our series with sporadic multiple piloleiomyomas revealed a 58-year-old male with progressive appearance of painful papules in the left scapular area and ipsilateral arm. There are no other family members affected. There is no personal or family history of kidney cancer or other fumarate hydratase-related neoplasms. Female members of the family did not present uterine myomas. We could not perform a genetic study in this case. Fumarate hydratase staining was strongly positive. Sporadic piloleiomyomas included in this study occurred in 8 women and in 12 men, with a median age of 59 years old (ages ranging from 30 to 78). The most frequently involved sites were the lower extremities,8 upper extremities,4 trunk2 and face,1 whereas in five cases the location was unknown. Median tumor-size was 1.1 cm (range from 0.4 to 2 cm). The histopathologic characteristics of these sporadic tumors were identical to those previously described in piloleiomyomas from patients with HLRCC syndrome, although only six cases with very subtle nuclear atypia was found in these cases. Most sporadic piloleiomyomas presented strong positive fumarate hydratase staining (Fig. 2H–K), although five cases presented negative fumarate hydratase staining. A table compiling the fumarate hydratase staining results as well as the sensitivity/specificity for every tested group is included (Table 1). Some clinical and histopathologic features of the patients with HLRCC syndrome and with sporadic solitary or multiple piloleiomyomas are summarized in Table 2.
Fumarate hydratase and MCUL/HLRCC Table 1. Compilation of fumarate hydratase staining results within the different studied groups, including sensitivity and specificity of fumarate hydratase staining in genetically confirmed HLRCC or all highly suspicious HLRCC cases
Negative fumarate staining Positive fumarate staining
Genetically confirmed HLRCC cases
All suspected HLRCC cases
Sporadic single/multiple piloleiomyomas
Angioleiomyomas
Leiomyosarcomas
4 1 (1 weak)
8 2 (2 weak)
5 16 (8 weak)
0 2 (1 weak)
0 7 (2 weak)
Only genetically confirmed
All suspected HLRCC cases
80% 76.2%
80% 65.4%
Sensitivity Specificity
HLRCC, hereditary leiomyomatosis and renal cell cancer. Table 2. Cases of HLRCC and sporadic piloleiomyomas
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31
Gender/age
Type
Location
FH antibody
FH genetic study
Nuclear pleomorphism
M/55 M/29 F/47 F/48 F/30 F/40 M/58 M/37 F/53 M/35 M/36 F/75 M/66 F/78 M/75 F/56 M/46 M/64 M/43 F/53 M/76 M/61 M/60 F/37 F/49 M/57 F/48 M/30 M/46 M/53 F/75
HLRCC HLRCC HLRCC HLRCC HLRCC HLRCC Sp. mul. HLRCC HLRCC HLRCC HLRCC Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol. Sp. sol.
Left arm/neck/interscapular area Right leg Right arm Left arm Right arm Right scapular area Left scapular area Thorax Left scapular area Right deltoid area Left pectoral area Right leg Chin Right shoulder Non-reported Left knee Right buttock Left knee Left knee Left back Non-reported Left arm Non-reported Non-reported Right arm Right leg Right shoulder Non-reported Right thigh Back Right lower leg
Negative Negative Negative Weak positive Negative Weak positive Strong Positive Negative Negative Negative Negative Weak positive Weak positive Negative Strong positive Strong positive Strong positive Weak positive Strong positive Weak positive Weak positive Negative Negative Negative Negative Weak positive Weak positive Weak positive Strong positive Negative Strong positive
C1292_1334dup; p.(Asn446Arg*20) Not found. c139C>T (pGln47stop) Not available c1157A>G p Q386R cG13T>Gp.L138R (CTC>CGC Not available Not available Not available Cc. 667-669 delin TAG prot p. Lys 223. Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available Not available
Slight Slight Mild Slight Mild Slight Slight Slight None Slight Slight Slight None Slight None None None None None None None Slight None None Slight None None None None Slight Slight
M, male; F, female; FH, fumarate hydratase; HLRCC, hereditary leiomyomatosis and renal cell cancer; Sp. sol., sporadic solitary; Sp. mul., sporadic multiple.
We also evaluated seven well-differentiated cutaneous leiomyosarcomas, two of them infiltrating subcutaneous tissue. Five of those lesions consisted of grade 1 leiomyosarcomas, and the other two were grade 2. No areas of
massive necrosis of neoplastic cells were seen in these tumors. Five of the seven cutaneous leiomyosarcomas studied showed strong positive of fumarate hydratase staining, whereas weak positive immunostaining for fumarate
863
Llamas-Velasco et al.
Fig. 3. Grade I leiomyosarcoma (A) low power and (B) high power hematoxylin/eosin staining as well as (C,D) desminpositive staining and (E,F) fumarate hydratase grade 3 staining.
hydratase was seen in two cases (Fig. 3). The two angioleiomyomas studied showed respectively weak and strong positive fumarate hydratase immunostaining. Discussion Histopathologically, all piloleiomyomas included in this study were poorly demarcated neoplasms composed of interweaving fascicles of spindle cells, with elongated central nuclei and few perinuclear vacuoles, and an eosinophilic cytoplasm, which are the usual histopathologic characteristics of these neoplasms. In our experience, however, it seems to be that piloleiomyomas associated with suspected or confirmed HLRCC syndrome present a higher degree of nuclear atypia compared with sporadic piloleiomyomas, as we found a slight or mild degree of nuclear atypia in 90% of them. There were no other distinctive histopathologic features, as the large nuclei with a prominent inclusion-like eosinophilic nucleolus surrounded by a clear halo, as those described by Merino et al. in the renal tumors associated
864
with this syndrome,12 were found in the cutaneous piloleiomyomas of our study. In any case, slight nuclear atypia can be also be found in sporadic solitary leiomyomas. Cutaneous piloleiomyomas showed a higher penetrance among men, although women significantly developed more numerous lesions, without differences in their distribution.13 Clustered lesions were most common on the trunk, lower limbs, upper limbs, head and neck. Unilateral or bilateral symmetrically distributed lesions were infrequent.13 As we only included patients with cutaneous leiomyomas, gender differences of penetrance of skin lesions could not be further analyzed. Fumarate hydratase immunoreactivity is preserved in most sporadic piloleiomyomas, and thus although their molecular basis is not well known, our results suggest that they maintain fumarate hydratase activity. Up to 54% of the cutaneous leiomyosarcomas studied by Barker et al. showed losses in chromosome one
Fumarate hydratase and MCUL/HLRCC involving 1q42-4314 but all cutaneous leiomyosarcomas included in our study showed preserved fumarate hydratase immunoreactivity. We also found that immunohistochemical positive expression for fumarate hydratase is usually preserved in sporadic angioleiomyomas. A peculiar finding of our study consisted of the presence of well-preserved fumarate hydratase immunoexpression in the smooth muscle fibers of the pre-existing arrector pili muscle seen in some of the biopsies of the piloleiomyomas from patients with suspected or confirmed HLRCC syndrome, whereas it was clearly decreased or absent in neoplastic smooth muscle cells. Because patients with HLRCC syndrome present a germinal heterozygotous mutation, this finding supports the idea that an additional loss of heterozygosity is necessary in order to elicit tumor growth in piloleiomyomas of these patients. Bardella et al. proposed a polyclonal S-(2-succinyl) cysteine (2SC) antibody to immunohistochemically identify an increase in succinated proteins, as they show that immunohistochemical detection of fumarate hydratase was unreliable due to the fact that many fumarate hydratase mutations are missense changes, which lead to the production of a stable but inactive enzyme.10 This hypothesis, the production of a stable but inactive enzyme, could explain the weak positive staining found in two of our genetically confirmed cases of HLRCC syndrome, as they presented weaker staining than the sporadic piloleiomyomas. In conclusion, the results of fumarate hydratase immunostaining should be interpreted in a
clinicopathologic correlation context, because a weak positive result for fumarate hydratase immunostaining does not exclude a diagnosis of HLRCC syndrome. The main limitation of our study is that we lack additional clinical data to further classify the five cases of ‘suspected sporadic leiomyomas’ as truly sporadic leiomyomas, and we cannot completely rule out that some of that cases could be actually be HLRCC cases. The sensitivity and specificity of fumarate hydratase staining in our series therefore varies depending on how we define HLRCC cases on a clinical or on a genetic basis (Table 2). These results suggest that this staining could indicate a high risk of HLRCC in most of the cases but it could also suggest the presence of a syndrome in some of the sporadic cases. In summary, cutaneous leiomyomas associated with HLRCC syndrome may present some degree of nuclear atypia and usually show negative or weak positive immunostaining for fumarate hydratase. Therefore, as genetic testing is not easily available due to its high cost, and direct assay of intracellular fumarate levels is technically demanding, the immunohistochemical investigation for fumarate hydratase may provide a helpful tool for screening and stratifying the risk of HLRCC syndrome. Cases with strong fumarate hydratase staining can be excluded from further testing, while cases with negative or weak immunostaining may require further analysis regarding personal and family clinical history to classify risk of an underlying genetic syndrome, such as HLRCC.
References 1. Reed WB, Walker R, Horowitz R. Cutaneous leiomyomata with uterine leiomyomata. Acta Derm Venereol 1973; 53: 409. 2. Stewart L, Glenn GM, Stratton P, et al. Association of germline mutations in the fumarate hydratase gene and uterine fibroids in women with hereditary leiomyomatosis and renal cell cancer. Arch Dermatol 2008; 144: 1584. 3. Kloepfer HW, Krafchuk J, Derbes V, Burks J. Hereditary multiple leiomyoma of the skin. Am J Hum Genet 1958; 10: 48. 4. Alam NA, Olpin S, Rowan A, et al. Missense mutations in fumarate hydratase in multiple cutaneous and uterine leiomyomatosis and renal cell cancer. J Mol Diagn 2005; 7: 437. 5. Carvajal-Carmona LG, Alam NA, Pollard PJ, et al. Adult leydig cell tumors of the testis caused by germline fumarate hydratase mutations. J Clin Endocrinol Metab 2006; 91: 3071.
6. Bayley JP, Launonen V, Tomlinson IP. The fumarate hydratase mutation database: an online database of fumarate hydratase mutations involved in the MCUL (HLRCC) tumor syndrome and congenital fumarase deficiency. BMC Med Genet 2008; 9: 20. 7. Tomlinson IP, Alam NA, Rowan AJ, et al. Germline mutations in fumarate hydratase predispose to dominantly inherited uterine fibroids, skin leiomyomata and papillary renal cell cancer. Nat Genet 2002; 30: 406. 8. Pollard PJ, Ratcliffe PJ. Cancer. Puzzling patterns of predisposition. Science 2009; 324: 192. 9. Maxwell PH. Seeing the smoking gun: a sensitive and specific method to visualize loss of the tumour suppressor, fumarate hydratase, in human tissues. J Pathol 2011; 225: 1. 10. Bardella C, El-Bahrawy M, Frizzell N, et al. Aberrant succination of proteins in
11.
12.
13.
14.
fumarate hydratase-deficient mice and HLRCC patients is a robust biomarker of mutation status. J Pathol 2011; 225: 4. Coindre JM, Trojani M, Contesso G, et al. Reproducibility of a histopathologic grading system for adult soft tissue sarcoma. Cancer 1986; 58: 306. Merino MJ, Torres-Cabala C, Pinto P, Linehan WM. The morphologic spectrum of kidney tumors in hereditary leiomyomatosis and renal cell carcinoma (HLRCC) syndrome. Am J Surg Pathol 2007; 31: 1578. Alam NA, Barclay E, Rowan AJ, et al. Clinical features of multiple cutaneous and uterine leiomyomatosis: an underdiagnosed tumor syndrome. Arch Dermatol 2005; 141: 199. Barker KT, Bevan S, Wang R, et al. Low frequency of somatic mutations in the fumarate hydratase/multiple cutaneous leiomyomatosis gene in sporadic leiomyosarcomas and uterine leiomyomas. Br J Cancer 2002; 87: 446.
865
