Abstract Objectives Most giant cell tumors of bone (GCTs) occur in patients aged 20 to 40 years. We analyzed features of GCT in patients 55 years or older. Methods GCTs were examined for fibrosis, matrix, cystic change, histiocytes, mitoses, and necrosis. Clinical/radiologic data were collected. Results Thirty-four (5%) of 710 GCTs occurred in patients older than 55 years (14/20 male/female; 56-83 years) in long bones (n = 24), vertebrae (n = 6), pelvis (n = 3), and metacarpal (n = 1). Imaging was classic in 26 of 27 cases; one case appeared malignant. Morphologic patterns included fibrosis (n = 29), bone formation (n = 19), cystic change (n = 8), necrosis (n = 8), foamy histiocytes (n = 7), and secondary aneurysmal bone cyst formation (n = 1). Mitoses ranged from 0 to 18 per 10 high-power fields. Six recurred; one patient developed metastasis. Four of five cases harbored H3F3A mutations. Conclusions GCTs in patients 55 years or older share pathologic characteristics with those arising in younger adults. Fibrosis and reactive bone are common, potentially leading to diagnostic confusion in this population. No histologic features correlate with adverse outcome. Giant cell tumor of bone; Bone tumors; Elderly; H3F3A Giant cell tumor of bone (GCT) represents approximately 5% of primary bone tumors and most commonly occurs in skeletally mature individuals between the ages of 20 and 40 years.1-11 GCT primarily arises in the epiphyses of long bones, vertebral bodies, and flat bones such as the pelvis, in decreasing order of frequency. Histologically, this tumor is characterized by bland, round to ovoid mononuclear cells in a background of uniformly distributed osteoclast-like giant cells.12 Imaging typically reveals an osteolytic lesion with a variable degree of cortical thinning and expansion.7,11,13 GCT is uncommon in older adults, although its occurrence in this population is well documented within large population-based studies reporting the incidence of primary bone tumors.1-6,8-11,14-16 However, studies dedicated to the histologic features in this age group are lacking. Furthermore, differentiating GCT from other giant cell–rich entities, including giant cell reparative granuloma, Brown tumor of hyperparathyroidism, and giant cell–rich carcinomas and sarcomas, may be challenging without knowledge of the histologic spectrum of GCT. To our knowledge, we present the largest series to date specifically emphasizing the morphologic features of GCT presenting in patients 55 years of age and older. Materials and Methods This study was approved by the institutional review board at Mayo Clinic. Histologic sections representing tissue obtained from surgical procedures, including excisional biopsy specimens, curettage, and wide resection diagnosed as primary “giant cell tumor of bone” from January 1, 1940, to December 31, 2015, were retrieved from our institutional archives. All available H&E-stained slides from patients 55 years or older at the time of initial diagnosis were reviewed by bone pathologists (K.J.F., J.M.C., J.M.B., and C.Y.I.) to confirm the diagnosis. The histologic patterns queried for each case included reactive bone formation, hyaline cartilage, foamy histiocytes, cystic change, secondary aneurysmal bone cyst formation, infarct-like necrosis, cytologic atypia, and fibrosis. Fibrosis was morphologically subclassified into the following patterns, previously described in a series of pediatric GCT17: loose, pericellular, hyalinized, geographic, septal, and lace-like. Mitotic rates in the area of highest mitotic activity were recorded per 10 high-power fields (hpfs) for each case. The available radiologic images and imaging reports were reviewed by an experienced musculoskeletal radiologist (D.E.W.) focusing on location, imaging characteristics, extent of involvement, and pathologic fracture. Age, sex, and pertinent follow-up data were obtained from our institutional medical records. A subset of GCTs was tested for H3F3A mutations by polymerase chain reaction (PCR) and Sanger sequencing. Genomic DNA was extracted from formalin-fixed, paraffin-embedded (FFPE) tissue using the QIAamp DSP DNA FFPE Tissue Kit (Qiagen GmbH, Hilden, Germany). The PCR primers used in this study are as follows: 5ʹ-AAATCGACCGGTGGTAAAGC (forward) and 5ʹ-ATACAAGAGAGACTTTGTCCCA (reverse). After PCR amplification, the 145-bp PCR product of H3F3A underwent Sanger sequencing to detect mutations in the mutation hotspot in exon 2, codon 34. Patient and tumor characteristics were summarized with frequencies and percentages or with medians and ranges, as appropriate. Features were compared between categories with Fisher exact tests for categorical data and with Wilcoxon rank-sum tests for continuous data. All analyses were performed using SAS version 9.4 (SAS Institute, Cary NC) and R.18P values less than .05 were considered statistically significant. Results Clinical Features Thirty-four cases of GCT in patients 55 years and older were identified, from a total of 710 cases of GCT in our institutional archives Table 1 . GCT occurred in 14 (41%) men and 20 (59%) women, ranging in age from 56 to 83 years (median, 59.5 years). Anatomic sites of involvement included radius (n = 7, 21%), femur (n = 6, 18%), tibia (n = 6, 18%), vertebral body (n = 6, 18%), humerus (n = 4, 12%), pelvis (n = 3, 9%), fibula (n = 1, 3%), and metacarpal (n = 1, 3%). Clinical features of hyperparathyroidism were not identified in any patients. Table 1 Clinical, Radiologic, and Histologic Features of Giant Cell Tumor of Bone Characteristic No. (%) of Cases Clinical variables (n = 34) Sex Male 14 (41) Female 20 (59) Site Long tubular bone 24 (71) Vertebral body 6 (18) Pelvis 3 (9) Metacarpal 1 (3) Radiologic features (n = 27) Extent of involvement of bones with an epiphysisa (n = 20) Epiphysis only 8 (40) Epiphysis and metaphysis and/or diaphysis 12 (60) Pathologic fracture 12 (60) Histologic findings (n = 34) Mitotic rate <4 per 10 hpfs 14 (41) 4-8 per 10 hpfs 12 (35) >8 per 10 hpfs 8 (24) Infarct-like necrosis 8 (24) Fibrosis 29 (85) Bone formation 19 (56) Cystic change 8 (24) Secondary aneurysmal bone cyst 1 (3) Foamy histiocytes 7 (21) Characteristic No. (%) of Cases Clinical variables (n = 34) Sex Male 14 (41) Female 20 (59) Site Long tubular bone 24 (71) Vertebral body 6 (18) Pelvis 3 (9) Metacarpal 1 (3) Radiologic features (n = 27) Extent of involvement of bones with an epiphysisa (n = 20) Epiphysis only 8 (40) Epiphysis and metaphysis and/or diaphysis 12 (60) Pathologic fracture 12 (60) Histologic findings (n = 34) Mitotic rate <4 per 10 hpfs 14 (41) 4-8 per 10 hpfs 12 (35) >8 per 10 hpfs 8 (24) Infarct-like necrosis 8 (24) Fibrosis 29 (85) Bone formation 19 (56) Cystic change 8 (24) Secondary aneurysmal bone cyst 1 (3) Foamy histiocytes 7 (21) hpfs, high-power fields. aOr an epiphyseal equivalent. View Large Radiologic Findings Radiologic images (n = 13) or reports (n = 14) were available for review in 27 cases (Table 1). For the 13 cases with imaging available (long or short tubular bone, n = 9; vertebral body, n = 2; and pelvis, n = 2), all demonstrated an osteolytic pattern of destruction, with 11 cases showing variable degrees of associated bony expansion. Most of the lytic lesions (11 of 13) had a narrow zone of transition Image 1 , with only two cases showing a peripheral rim of sclerosis. Of the nine cases located in the long or short tubular bones, all demonstrated involvement of the epiphysis. Two of the nine involved the epiphysis and metaphysis, while the remaining seven showed epiphyseal, metaphyseal, and diaphyseal involvement. Cortical destruction and an associated soft tissue mass were present in nine of the 13 cases Image 2 . A pathologic fracture was identified in five cases. Image 1 View largeDownload slide Anteroposterior (A) and lateral (B) radiographs and sagittal T1-weighted (C) and sagittal T2-weighted (D) magnetic resonance images of a 65-year-old woman with a giant cell tumor of bone in the distal radius demonstrate a purely osteolytic destructive lesion involving the distal radial epiphysis and metaphysis (asterisks) that extends to the subchondral endplate. The lesion has a narrow zone of transition without a sclerotic rim and is associated with marked volar expansion with preservation of a thin peripheral rim of bone (arrows) and a pathologic fracture (arrowhead). The lesion has classic radiographic imaging features of a giant cell tumor. The signal characteristics are nonspecific on magnetic resonance imaging, where it presents as a heterogeneous solid mass with an intermediate signal on T1-weighted images and mixed, but predominantly high, signal intensity on T2-weighted images. Image 1 View largeDownload slide Anteroposterior (A) and lateral (B) radiographs and sagittal T1-weighted (C) and sagittal T2-weighted (D) magnetic resonance images of a 65-year-old woman with a giant cell tumor of bone in the distal radius demonstrate a purely osteolytic destructive lesion involving the distal radial epiphysis and metaphysis (asterisks) that extends to the subchondral endplate. The lesion has a narrow zone of transition without a sclerotic rim and is associated with marked volar expansion with preservation of a thin peripheral rim of bone (arrows) and a pathologic fracture (arrowhead). The lesion has classic radiographic imaging features of a giant cell tumor. The signal characteristics are nonspecific on magnetic resonance imaging, where it presents as a heterogeneous solid mass with an intermediate signal on T1-weighted images and mixed, but predominantly high, signal intensity on T2-weighted images. Image 2 View largeDownload slide Axial computed tomography images with bone (A) and soft tissue (B) window settings of a 72-year-old man show an eccentrically positioned purely osteolytic destructive lesion involving the L4 vertebral body (asterisks). The lesion extends posteriorly to involve a portion of the pedicle and transverse process (curved arrows). It is also associated with destruction of the cortex along the lateral margin of the vertebral body, where there is an associated soft tissue mass (arrows). Although giant cell tumor of bone can be associated with cortical destruction and soft tissue mass, the differential diagnosis for a lesion with these aggressive features in a patient of this age would include metastatic disease, lymphoma, and multiple myeloma. Image 2 View largeDownload slide Axial computed tomography images with bone (A) and soft tissue (B) window settings of a 72-year-old man show an eccentrically positioned purely osteolytic destructive lesion involving the L4 vertebral body (asterisks). The lesion extends posteriorly to involve a portion of the pedicle and transverse process (curved arrows). It is also associated with destruction of the cortex along the lateral margin of the vertebral body, where there is an associated soft tissue mass (arrows). Although giant cell tumor of bone can be associated with cortical destruction and soft tissue mass, the differential diagnosis for a lesion with these aggressive features in a patient of this age would include metastatic disease, lymphoma, and multiple myeloma. Of the 14 patients with only radiologic reports available for review, anatomic sites included long tubular bone (n = 11), vertebral body (n = 2), and pelvis (n = 1). All reports described an osteolytic pattern of destruction, with five cases also demonstrating osseous expansion. The zone of transition or presence of a sclerotic rim was not noted for this subset. Epiphyseal involvement was present in all 11 cases involving long tubular bones, while five cases also documented involvement of the metaphysis. Half of the cases exhibited evidence of cortical destruction, while three had soft tissue extension. Seven cases were associated with pathologic fracture. A single case involving the pedicle and body of the L3 vertebra of a 66-year-old woman was described specifically as having malignant features. Multifocal disease or features suggestive of Paget disease were not present in any case. Pathologic Features All tumors showed areas harboring classic GCT morphology, specifically round to oval mononuclear cells within a background of evenly spaced osteoclast-like giant cells (Table 1 and Image 3 ). However, only a single case was composed solely of conventional GCT morphology. At least one additional morphologic pattern was identified in the remaining cases, even though these patterns typically comprised less than 10% of the tumor volume available for review. The mitotic rate ranged from 0 to 18 mitoses per 10 hpfs (median, 5 mitoses per 10 hpfs). Atypical mitoses were not identified, and no case showed significant cytologic atypia. Image 3 View largeDownload slide All cases of giant cell tumor of bone showed areas of classic morphology consisting of round to oval mononuclear cells and evenly distributed osteoclast-like giant cells (A, H&E, ×10; B, H&E, ×20). Image 3 View largeDownload slide All cases of giant cell tumor of bone showed areas of classic morphology consisting of round to oval mononuclear cells and evenly distributed osteoclast-like giant cells (A, H&E, ×10; B, H&E, ×20). The most common alternative histologic pattern identified was fibrosis, which was present in 29 (85%) cases. Pericellular fibrosis, consisting of thin strands of fibrous tissue encasing cords, strips, and individual tumor cells, was the most common pattern of fibrosis, present in 12 (35%) cases Image 4A . Paucicellular, eosinophilic hyalinized fibrosis was seen in 11 (32%) cases Image 4B . Nine (26%) cases harbored thick bands of fibrosis, designated septal fibrosis Image 4C , and an equal number of cases exhibited loose fibrosis characterized by sparse fibroblasts and myofibroblasts deposited in a vascular stroma Image 4D . Seven (21%) cases contained lesional cells within a densely collagenous background (lace-like fibrosis) Image 4E . Geographic fibrosis was the least frequently encountered type of fibrosis, noted in six (18%) cases, and these cases harbored islands of conventional GCT demarcated by a serpiginous arrangement of fibrous septa Image 4F . Seventeen (50%) cases contained a single fibrosis pattern, while 12 cases exhibited two or more. Image 4 View largeDownload slide The spectrum of histologic patterns of fibrosis observed in giant cell tumor of bone included pericellular (A, H&E, ×10), hyalinized (B, H&E, ×10), septal (C, H&E, ×10), loose (D, H&E, ×10), lace-like (E, H&E, ×20), and geographic (F, H&E, ×4). Image 4 View largeDownload slide The spectrum of histologic patterns of fibrosis observed in giant cell tumor of bone included pericellular (A, H&E, ×10), hyalinized (B, H&E, ×10), septal (C, H&E, ×10), loose (D, H&E, ×10), lace-like (E, H&E, ×20), and geographic (F, H&E, ×4). Reactive-appearing bone was found in 19 (56%) cases. These foci contained thin, irregular trabeculae lined by a single layer of plump, cytologically bland osteoblasts Image 5 ) Eight of these cases were previously biopsied, and four patients had a pathologic fracture, although bone formation was not associated with recognizable biopsy site changes or other reactive changes in this cohort. Image 5 View largeDownload slide Reactive-appearing bone, comprising thin trabeculae lined by osteoblasts, was present in just over half of cases (H&E, ×20). Image 5 View largeDownload slide Reactive-appearing bone, comprising thin trabeculae lined by osteoblasts, was present in just over half of cases (H&E, ×20). Cystic change, composed of microscopic pools of hemorrhage within lesional stroma, was present in eight (24%) cases Image 6A One additional case showed secondary aneurysmal bone cyst formation Image 6B . Aggregates of foamy histiocytes were found in seven (21%) cases Image 7A . Infarct-like necrosis, characterized by ghost outlines of mononuclear and giant cells with sharp demarcation from surrounding viable tissue, was identified in eight (24%) cases Image 7B . No case contained cartilage matrix. Image 6 View largeDownload slide Cystic changes observed included simple cystic change with associated hemorrhage (A, H&E, ×10) and secondary aneurysmal bone cyst formation in one case (B, H&E, ×4). Image 6 View largeDownload slide Cystic changes observed included simple cystic change with associated hemorrhage (A, H&E, ×10) and secondary aneurysmal bone cyst formation in one case (B, H&E, ×4). Image 7 View largeDownload slide Additional histologic features of giant cell tumor of bone (GCT) included aggregates of foamy histiocytes (A, H&E, ×10) and infarct-like necrosis with sharp demarcation from surrounding classic GCT (B, H&E, ×10). Image 7 View largeDownload slide Additional histologic features of giant cell tumor of bone (GCT) included aggregates of foamy histiocytes (A, H&E, ×10) and infarct-like necrosis with sharp demarcation from surrounding classic GCT (B, H&E, ×10). Molecular Genetics Mutational analysis of H3F3A was performed on five GCTs chosen at random Table 2 . Testing revealed H3F3A mutation in four tumors, while the remaining case was wild- type. Of the four cases with mutations, three had the more common p.G34W mutation, while one case harbored a p.G34V mutation Image 8 . Table 2 Clinical Characteristics of Cases Chosen for H3F3A Mutation Analysis Case No. Age, y/Sex Site Radiologic Extent of Involvement Follow-up Mutation Status 1 59/F Radius NA ANED p.G34W 2 57/F Metacarpal Epiphyseal/metadiaphyseal ANED p.G34W 3 58/F Radius Epiphyseal/metaphyseal, pathologic fracture ANED p.G34V 4 68/F Tibia NA ANED p.G34W 5 60/M Sacrum NA ANED WT Case No. Age, y/Sex Site Radiologic Extent of Involvement Follow-up Mutation Status 1 59/F Radius NA ANED p.G34W 2 57/F Metacarpal Epiphyseal/metadiaphyseal ANED p.G34W 3 58/F Radius Epiphyseal/metaphyseal, pathologic fracture ANED p.G34V 4 68/F Tibia NA ANED p.G34W 5 60/M Sacrum NA ANED WT ANED, alive with no evidence of disease; NA, not available; WT, wild-type. View Large Image 8 View large Download slide View large Download slide A-F, All cases of giant cell tumor of bone (GCT) with Sanger sequencing of H3F3A, codon 34, performed showed similar, usual GCT histology, regardless of mutational status. Representative photomicrograph (H&E, ×10) and results of Sanger sequencing: Case 1 with H3F3A mutation (p.G34W, GGG>TGG) (A, B). Case 3 with H3F3A mutation (p.G34V, GGG>GTG) (C, D). Case 5 with H3F3A wild-type (E, F). Image 8 View large Download slide View large Download slide A-F, All cases of giant cell tumor of bone (GCT) with Sanger sequencing of H3F3A, codon 34, performed showed similar, usual GCT histology, regardless of mutational status. Representative photomicrograph (H&E, ×10) and results of Sanger sequencing: Case 1 with H3F3A mutation (p.G34W, GGG>TGG) (A, B). Case 3 with H3F3A mutation (p.G34V, GGG>GTG) (C, D). Case 5 with H3F3A wild-type (E, F). Treatment Treatment data were available for 33 patients. Fifteen patients underwent intralesional surgery/curettage (with or without phenolization), and 18 patients underwent wide resection. Four cases treated with wide resection received postoperative external beam radiation. Two of these patients developed recurrent tumor, and both were treated again with radiation. A second recurrence in one of these patients was treated with chemotherapy. Outcomes Follow-up information was available for 33 patients with a range of 0 to 284 months (median, 115 months). Six tumors locally recurred, and one patient developed late metastases without local recurrence Table 3 . Table 3 Clinical Characteristics of Patients With Recurrence or Metastasis Age, y/Sex Site Treatment Follow-up, mo Event Time to Initial Event, mo Clinical Outcome 65/F Radius R 55 Recur 43 ANED 58/M Tibia C 285 Recur 16 DOOD 71/F Ilium R 166 Recur 163 DOOD 57/F T9 R 84 Recur 67 DOD 56/M Tibia C 83 Recur 13 ANED 72/M L4 R 105 Recur 13 DOOD 61/M Radius C 271 Met 53 DOOD Age, y/Sex Site Treatment Follow-up, mo Event Time to Initial Event, mo Clinical Outcome 65/F Radius R 55 Recur 43 ANED 58/M Tibia C 285 Recur 16 DOOD 71/F Ilium R 166 Recur 163 DOOD 57/F T9 R 84 Recur 67 DOD 56/M Tibia C 83 Recur 13 ANED 72/M L4 R 105 Recur 13 DOOD 61/M Radius C 271 Met 53 DOOD ANED, alive with no evidence of disease; C, curettage; DOD, dead of disease; DOOD, dead of other disease; R, resection. View Large Local recurrences occurred in the long bones (n = 3), vertebrae (n = 2), and pelvis (n = 1). Four recurrent cases were originally treated with wide resection and two with curettage. Time to local recurrence was 13 to 163 months (mean, 53 months; median, 29.5 months). One patient, a 57-year-old woman with GCT of the T9 thoracic vertebra, died of progressive local disease. She was initially treated with wide resection but the disease recurred locally at 67, 102, and 129 months, and she eventually received radiation and chemotherapy. She died of disease 150 months after initial diagnosis. All other cases recurred only once, based on available clinical follow-up data. A single patient, a 61-year-old man with a GCT of the radius initially treated by curettage, developed pulmonary metastasis without local recurrence 53 months after initial diagnosis. This patient had additional pulmonary metastases and died of pneumonia 271 months after initial diagnosis. The remaining 26 patients with follow-up, including the single patient with malignant-appearing imaging findings, did not experience recurrence or metastatic disease. Twenty-one were alive with no evidence of disease, and five died of other causes. Correlation of Morphologic Features and Clinical Characteristics No statistically significant correlation was found between morphologic patterns (fibrosis, reactive bone formation, cystic change, infarct-like necrosis, foamy histiocytes, secondary aneurysmal bone cyst formation) and anatomic site or patient age in this series. Similarly, mitotic rate and the presence of necrosis also did not correlate with age, site, or outcome. However, compared with a similar cohort of 63 pediatric GCTs,17 GCTs in older adults were more likely to show fibrosis (85% vs 49%, P = .0005) and involve the epiphysis (100% vs 79%, P = .036). Discussion GCT is a rare primary bone tumor, classically involving the epiphysis or epiphyseal equivalent, with a peak incidence in early to middle adulthood (20-40 years of age). Its incidence tapers off in the sixth decade and beyond.1-6,8-11,14,16,19 The occurrence of GCT in patients 55 years of age and older is well documented, comprising approximately 1.6% to 35% of case series and international population-based studies.1-6,8-11,14-16 However, series of GCT that include older adults do not delineate the clinical, radiologic, and pathologic characteristics that may be unique to this population. Only one publication specifically describing GCT in older adults has been published to our knowledge. McCarthy and Weber20 described a series of 10 cases comprising patients ranging in age from 62 to 78 years. One case showed an extensive fibrohistiocytic reaction, and another showed extensive necrosis, but no other distinct histologic features were noted.20 We sought to better define the histologic spectrum of GCT in patients 55 years and older to avoid misdiagnosis as other giant cell–rich lesions seen in this population. Of 710 cases of primary GCT seen at our institution over a span of approximately 100 years, 34 (5%) cases were diagnosed in patients 55 years of age or older. The clinical features found in the current series are concordant with those features documented in classic populations with GCT. Similar to other large series noting a slight female predominance,6,8,10,11 the female to male ratio in our series of older patients with GCT was approximately 1.5:1. The anatomic distribution of GCT in our study is also similar to prior series, with most cases involving the long bones (71%), including radius (21%), followed by the femur and tibia (18% each), and then humerus (12%). The imaging findings in our GCT cohort were comparable to those arising in the conventional age range. All tumors except one showed classic features on imaging, including an eccentric expansile osteolytic lesion. One case with features available by report only had aggressive imaging findings, although specific details were lacking. Malignant features were not identified histologically in this case Image 9 , and this patient did not experience recurrence or metastasis. Associated pathologic fracture, normally present in 5% to 10% of all cases of GCT,6,10,11 was present in 44% (12 of 34) of our cases. This higher incidence of pathologic fracture may be due to the inherent increased risk of fracture in this older population, compounded by an expansile osteolytic bone tumor. Not surprisingly, all GCTs arising in tubular bones or bones with an epiphyseal equivalent involved the epiphysis (20 of 20, 100%), which was statistically significant when comparing this group with our recent study in pediatric patients, which showed epiphyseal involvement in approximately 80% of cases (P = .036).17 Image 9 View largeDownload slide Histologic examination of giant cell tumor of bone involving the pedicle and L3 vertebral body of a 66-year-old woman with malignant features described on radiologic evaluation shows conventional morphology without cytologic atypia. Image 9 View largeDownload slide Histologic examination of giant cell tumor of bone involving the pedicle and L3 vertebral body of a 66-year-old woman with malignant features described on radiologic evaluation shows conventional morphology without cytologic atypia. All cases in this series showed areas of histologically conventional GCT, including bland round or ovoid mononuclear cells in a background of evenly spaced osteoclast-like giant cells, although only one case consisted entirely of this morphology. Additional histologic patterns identified in the remaining cases included fibrosis, reactive bone formation, infarct-like necrosis, cystic changes, foamy histiocytes, and secondary aneurysmal bone cyst, in descending order of frequency. Fibrosis was identified in 85% of our cases, manifesting in a variety of morphologic patterns, including pericellular, hyalinized, septal, loose, lace-like, and geographic. McCarthy and Weber20 suggested that a prominent fibrohistiocytic reparative reaction may be an indicator of chronicity in their series in elderly patients. Our data seem to support this hypothesis, especially when comparing the presence of fibrosis in our recent series of pediatric patients (aged <18 years) with GCT, which was 49.2% (P = .0005). Direct comparison of the prevalence of fibrosis in our study with other studies is largely precluded by a lack of systematic recording of fibrosis in the latter. Even though the fibrosis identified in these tumors is typically only focal, this morphologic pattern is important to recognize, as its presence raises the possibility of other giant cell–rich entities such as Brown tumor of hyperparathyroidism, giant cell reparative granuloma, and solid aneurysmal bone cyst. When ample tissue is available for review, differentiating these entities is often aided by the identification of areas classic for GCT. However, on biopsy specimens, the distinction is made more challenging when fibrosis is present. In differentiating GCT from Brown tumor of hyperparathyroidism, laboratory measurement of calcium, phosphorus, and parathyroid hormone levels is critical.21,22 Furthermore, Brown tumor may be multifocal on imaging, a feature rarely seen in GCT and not appreciated in our series. Giant cell reparative granuloma almost exclusively arises in the craniofacial bones, a site uncommonly affected by GCT.23,24 Solid aneurysmal bone cyst usually affects individuals in the first two decades of life and occurs in the metaphysis of long bones, small bones of the hands and feet, and posterior elements of vertebrae, rather than the epiphysis and vertebral bodies, respectively. Assay for USP6 rearrangement, present in up to 70% of primary aneurysmal bone cysts25,26 but not GCT, may also be employed. GCT with fibrosis, particularly in combination with foamy histiocytes (a combination occurring in six cases in our series), may mimic nonossifying fibroma. Histologically, nonossifying fibroma consists of bland spindle cells in a prominent storiform arrangement admixed with unevenly spaced multinucleated giant cells, foamy histiocytes, and hemosiderin-laden macrophages. While both GCT and nonossifying fibroma share some morphologic similarities, the latter presents with a peak incidence in the second decade of life as an eccentric cortically based lesion with elongated morphology, narrow zone of transition, and a well-defined peripheral rim of sclerosis. Bone formation was found in 56% of our cases, and its presence often raises concern for osteosarcoma. While most osteosarcomas arise in adolescents and young adults in a metaphyseal location, there is a second incidence peak in the sixth and seventh decades of life. Careful histologic inspection will show the bony component of GCT to be reactive, consisting of woven bone lined by a single layer of plump osteoblasts within a vascularized background. Osteosarcomas often show more extensive bone formation, although in some cases, the fine lace-like pattern of fibrosis seen in GCT may simulate the “lace-like” pattern of malignant osteoid deposition characteristic of osteosarcoma. While the presence of matrix deposition is often distracting, focus on the cellular elements is paramount in differentiating these tumors. High-grade osteosarcomas harbor significant cytologic atypia and mitotic activity, including atypical mitoses, which are lacking in GCT. Clinical findings may provide additional clues, as osteosarcoma in the elderly is more often associated with prior radiation and Paget disease.27 The etiology of bone formation in GCT remains unclear but may be associated with prior biopsy or pathologic fracture; however, bone formation was also identified in a subset of patients in our study without previous biopsy or fracture. Other sarcomas primary to bone, including undifferentiated pleomorphic sarcoma and primary and secondary malignancy arising in GCT, may also enter the differential diagnosis of GCT, particularly in those cases with readily identifiable mitotic activity and necrosis. Consequently, it is imperative for pathologists to be aware that GCT may harbor these findings. Morphologic features incompatible with conventional GCT include cytologic pleomorphism and atypia, nuclear hyperchromasia, and atypical mitoses. Review of imaging is essential, as most conventional GCTs appear as well-defined eccentric osteolytic lesions, in contrast to the typically aggressive, destructive appearance of sarcoma. However, as illustrated by one of our cases, GCT can rarely appear malignant on imaging in the absence of histologic or clinical features of malignancy. Metastatic carcinoma is the most common bone neoplasm in the elderly population.28 Although differentiation of GCT from carcinoma is not usually difficult, the ubiquity of carcinoma in this population requires its consideration in the differential diagnosis. Furthermore, numerous carcinomas, including those of renal and pancreatic origin, have giant cell–rich variants.29 However, metastatic disease rarely affects the epiphysis of long bones. The presence of multiple lesions also strongly favors metastases over GCT. Histologically, carcinoma is characterized by infiltrative sheets and clusters of atypical cells, lacking the monotony of the mononuclear constituent of GCT. Although mitotic figures are readily identifiable in GCT and carcinoma, the presence of atypical mitoses argues against the former. Finally, the presence of cytokeratin expression should help confirm the diagnosis of carcinoma. Four of five cases analyzed harbored H3F3A mutations: three cases with G34W (GGG>TGG) and one case with G34V (GGG>GTG). The fifth case was wild-type for H3F3A. Despite differences in molecular alterations, all of these cases exhibited similar histology (Image 8). Even though a limited number of cases were analyzed molecularly, GCT in this age group seems to share a similar genetic profile with their younger counterparts. Amary and colleagues30 recently demonstrated H3F3A G34W immunohistochemistry to be a robust marker to differentiate GCT from its mimics; this immunostain may be used in difficult cases. Conventional GCT may be locally aggressive, recurring in approximately 15% to 50% of cases,6,7,9-11,31-34 with pulmonary metastases occurring overall in about 2% of patients,35 although some series report a higher rate.32,33 Six (18%) patients in our study experienced local recurrence, concordant with previous recurrence rates reported in the literature. Even though McCarthy and Weber20 found no recurrences in their series of 10 elderly patients, suggesting that GCT may behave less aggressively in this age group, our results show that GCT in patients older than 55 years is capable of locally aggressive behavior, warranting careful follow-up. One case of pulmonary metastasis was recorded in our set, again comparable to prior studies. Unfortunately, long-term clinical follow-up in older adults is also hampered by increasing mortality from other causes. Conclusion GCT in patients 55 years and older shares clinical, radiologic, histologic, and genetic features with their more typical counterparts in young to middle-aged adults. However, fibrosis is particularly prominent in GCT arising in older patients, potentially resulting in diagnostic challenges with limited material. Furthermore, some morphologic features of GCT in this age group such as mitotic activity and necrosis may raise concern for malignancy. Consequently, as with all bone lesions, correlation with clinical and imaging data is critical. Ancillary testing for H3F3A and USP6 mutations or H3F3A G34W immunohistochemistry is a useful adjunct in the workup of giant cell–rich bone tumors. References 1. Amelio JM, Rockberg J, Hernandez RKet al. Population-based study of giant cell tumor of bone in Sweden (1983-2011). Cancer Epidemiol . 2016; 42: 82- 89. Google Scholar CrossRef Search ADS PubMed 2. Lin F, Hu Y, Zhao Let al. The epidemiological and clinical features of primary giant cell tumor around the knee: a report from the multicenter retrospective study in China. J Bone Oncol . 2016; 5: 38- 42. Google Scholar CrossRef Search ADS PubMed 3. Niu X, Xu H, Inwards CYet al. Primary bone tumors: epidemiologic comparison of 9200 patients treated at Beijing Ji Shui Tan Hospital, Beijing, China, with 10165 patients at Mayo Clinic, Rochester, Minnesota. Arch Pathol Lab Med . 2015; 139: 1149- 1155. Google Scholar CrossRef Search ADS PubMed 4. Rockberg J, Bach BA, Amelio Jet al. Incidence trends in the diagnosis of giant cell tumor of bone in Sweden since 1958. J Bone Joint Surg Am . 2015; 97: 1756- 1766. Google Scholar CrossRef Search ADS PubMed 5. Baena-Ocampo Ldel C, Ramirez-Perez E, Linares-Gonzalez LMet al. Epidemiology of bone tumors in Mexico City: retrospective clinicopathologic study of 566 patients at a referral institution. Ann Diagn Pathol . 2009; 13: 16- 21. Google Scholar CrossRef Search ADS PubMed 6. Campanacci M, Baldini N, Boriani Set al. Giant-cell tumor of bone. J Bone Joint Surg Am . 1987; 69: 106- 114. Google Scholar CrossRef Search ADS PubMed 7. McDonald DJ, Sim FH, McLeod RAet al. Giant-cell tumor of bone. J Bone Joint Surg Am . 1986; 68: 235- 242. Google Scholar CrossRef Search ADS PubMed 8. Dahlin DC. Caldwell lecture. Giant cell tumor of bone: highlights of 407 cases. AJR Am J Roentgenol . 1985; 144: 955- 960. Google Scholar CrossRef Search ADS PubMed 9. Larsson SE, Lorentzon R, Boquist L. Giant-cell tumor of bone: a demographic, clinical, and histopathological study of all cases recorded in the Swedish cancer registry for the years 1958 through 1968. J Bone Joint Surg Am . 1975; 57: 167- 173. Google Scholar CrossRef Search ADS PubMed 10. Goldenberg RR, Campbell CJ, Bonfiglio M. Giant-cell tumor of bone: an analysis of two hundred and eighteen cases. J Bone Joint Surg Am . 1970; 52: 619- 664. Google Scholar CrossRef Search ADS PubMed 11. Mnaymneh WA, Dudley HR, Mnaymneh LG. Giant-cell tumor of bone: an analysis and follow-up study of the forty-one cases observed at the Massachusetts General Hospital between 1925 and 1960. J Bone Joint Surg Am . 1964; 46: 63- 75. Google Scholar CrossRef Search ADS PubMed 12. Czerniak B. Dorfman and Czerniak’s Bone Tumors . 2nd ed. Philadelphia, PA: Elsevier; 2015. 13. Hodges FJ, Latourette HB, Macintyre RS. Radiologic aspects of giant-cell tumor of bone. Clin Orthop . 1956; 7: 82- 92. Google Scholar PubMed 14. Werner M. Giant cell tumour of bone: morphological, biological and histogenetical aspects. Int Orthop . 2006; 30: 484- 489. Google Scholar CrossRef Search ADS PubMed 15. McGrath PJ. Giant-cell tumour of bone: an analysis of fifty-two cases. J Bone Joint Surg Br . 1972; 54: 216- 229. Google Scholar PubMed 16. Liede A, Bach BA, Stryker Set al. Regional variation and challenges in estimating the incidence of giant cell tumor of bone. J Bone Joint Surg Am . 2014; 96: 1999- 2007. Google Scholar CrossRef Search ADS PubMed 17. Al-Ibraheemi A, Inwards CY, Zreik RTet al. Histologic spectrum of giant cell tumor (GCT) of bone in patients 18 years of age and below: a study of 63 patients. Am J Surg Pathol . 2016; 40: 1702- 1712. Google Scholar CrossRef Search ADS PubMed 18. R Core Team. R: A Language and Environment for Statistical Computing . Vienna, Austria: R Foundation for Statistical Computing; 2015. 19. Unni KK, Inwards CY. Dahlin’s Bone Tumors: General Aspects and Data on 11087 Cases . Philadelphia, PA: Lippincott Williams & Williams; 2010: 179- 183, 310-316. 20. McCarthy EF, Weber KL. Giant cell tumor of bone in elderly patients: a study of ten patients. Iowa Orthop J . 2009; 29: 79- 82. Google Scholar PubMed 21. Phulsunga RK, Parghane RV, Kanojia RKet al. Multiple Brown tumors caused by a parathyroid adenoma mimicking metastatic bone disease from giant cell tumor. World J Nucl Med . 2016; 15: 56- 58. Google Scholar PubMed 22. Bandeira F, Cassibba S. Hyperparathyroidism and bone health. Curr Rheumatol Rep . 2015; 17: 48. Google Scholar CrossRef Search ADS PubMed 23. Roy S, Joshi NP, Sigamani Eet al. Clival giant cell tumor presenting with isolated trigeminal nerve involvement. Eur Arch Otorhinolaryngol . 2013; 270: 1167- 1171. Google Scholar CrossRef Search ADS PubMed 24. Wolfe JT III, Scheithauer BW, Dahlin DC. Giant-cell tumor of the sphenoid bone: review of 10 cases. J Neurosurg . 1983; 59: 322- 327. Google Scholar CrossRef Search ADS PubMed 25. Oliveira AM, Perez-Atayde AR, Inwards CYet al. USP6 and CDH11 oncogenes identify the neoplastic cell in primary aneurysmal bone cysts and are absent in so-called secondary aneurysmal bone cysts. Am J Pathol . 2004; 165: 1773- 1780. Google Scholar CrossRef Search ADS PubMed 26. Oliveira AM, Hsi BL, Weremowicz Set al. USP6 (tre2) fusion oncogenes in aneurysmal bone cyst. Cancer Res . 2004; 64: 1920- 1923. Google Scholar CrossRef Search ADS PubMed 27. Savage SA, Mirabello L. Using epidemiology and genomics to understand osteosarcoma etiology. Sarcoma . 2011; 2011: 548151. Google Scholar CrossRef Search ADS PubMed 28. Hage WD, Aboulafia AJ, Aboulafia DM. Incidence, location, and diagnostic evaluation of metastatic bone disease. Orthop Clin North Am . 2000; 31: 515-5 28, vii. Google Scholar CrossRef Search ADS PubMed 29. Berzal Cantalejo F, Sabater Marco V, Alonso Hernández Set al. Syncytial giant cell component: review of 55 renal cell carcinomas. Histol Histopathol . 2004; 19: 113- 118. Google Scholar PubMed 30. Amary F, Berisha F, Ye Het al. H3F3A (Histone 3.3) G34W immunohistochemistry: a reliable marker defining benign and malignant giant cell tumor of bone. Am J Surg Pathol . 2017; 41: 1059- 1068. Google Scholar CrossRef Search ADS PubMed 31. Júnior RC, Pereira MG, Garcia PBet al. Epidemiological study on giant cell tumor recurrence at the Brazilian National Institute of Traumatology and Orthopedics. Rev Bras Ortop . 2016; 51: 459- 465. Google Scholar CrossRef Search ADS PubMed 32. Klenke FM, Wenger DE, Inwards CYet al. Recurrent giant cell tumor of long bones: analysis of surgical management. Clin Orthop Relat Res . 2011; 469: 1181- 1187. Google Scholar CrossRef Search ADS PubMed 33. Klenke FM, Wenger DE, Inwards CYet al. Giant cell tumor of bone: risk factors for recurrence. Clin Orthop Relat Res . 2011; 469: 591- 599. Google Scholar CrossRef Search ADS PubMed 34. Kivioja AH, Blomqvist C, Hietaniemi Ket al. Cement is recommended in intralesional surgery of giant cell tumors: a Scandinavian sarcoma group study of 294 patients followed for a median time of 5 years. Acta Orthop . 2008; 79: 86- 93. Google Scholar CrossRef Search ADS PubMed 35. Siebenrock KA, Unni KK, Rock MG. Giant-cell tumour of bone metastasising to the lungs: a long-term follow-up. J Bone Joint Surg Br . 1998; 80: 43- 47. Google Scholar CrossRef Search ADS PubMed Footnotes This article was originally presented as a poster at the 2017 meeting of the USCAP; March 8, 2017; San Antonio, TX. © American Society for Clinical Pathology, 2018. All rights reserved. For permissions, please e-mail: email@example.com
American Journal of Clinical Pathology – Oxford University Press
Published: Mar 1, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.
All for just $49/month
Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.
Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.
It’s easy to organize your research with our built-in tools.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera