Prognostic Role of RANK and RANKL Expression in Recurrent Giant Cell Tumor of Bone: A Retrospective Study
Received Date: May 18, 2018 / Accepted Date: Jun 11, 2018 / Published Date: Jun 21, 2018
Objective: Giant cell tumor of bone (GCTB) is a benign locally aggressive primary bone tumor with a tendency for recurrence which is one of the major problems in this disease. Recent data demonstrated the important role of the receptor activator of nuclear factor-κB (RANK)/RANK-ligand (RANKL) pathway in the pathogenesis. However, the roles of RANK and RANKL in predicting recurrence have never been proposed. Therefore, we aim to investigate the prognostic value of RANK and RANKL expression in predicting recurrence of GCTB which may change the treatment paradigm of the disease.
Method: 53 cases of GCTB were enrolled in the study. Data on patient demographics and clinical characteristics were reviewed. Immunohistochemistry was used to detect the expression of RANK and RANKL. Recurrence-free survival (RFS) analysis was performed by Kaplan–Meier method and the difference between survival curves were sought using the log-rank test. Cox’s proportional hazards model and binary logistic regression analyses were used to define the risk of recurrence.
Result: Of 53 cases, there were 8 patients (15.1%) had recurrent disease. The univariate analysis revealed that age (log rank 10.749, p=0.005), RANKL overexpression (log rank 5.187, p=0.023), RANK overexpression (log rank 4.055, p=0.044 for RANK) and co-overexpression of RANK/RANKL (log rank 7.541, p<0.006) were associated with recurrence of GCTB. Cox’s proportional hazards model emerged that the only signiÃ¯Â¬Âcant prognostic parameter capable of deÃ¯Â¬Âning the risk for recurrence was co-overexpression of RANK and RANKL (Hazard ratio 2.910; 95%CI 1.099-7.708; p= 0.032). Binary logistic regression multivariate analysis followed by ROC analysis conÃ¯Â¬Ârmed that cooverexpression of RANK/RANKL represented a signiÃ¯Â¬Âcant biological model to predict local recurrence (area under the curve=0.731±0.098; 95% CI 0.539–0.922, p=0.039).
Conclusion: RANK and RANKL co-overexpression increases the risk of recurrence of GCTB and could be a prognostic marker for recurrent disease.
Keywords: Giant cell tumour of bone; GCTB; RANK; RANKL; Recurrent; Prognostic factor
Giant cell tumour of bone; GCTB; RANK; RANKL; Recurrent; Prognostic factor
Giant cell tumor of bone (GCTB), also known as osteoclastoma, is accounting for 5% of all primary bone tumors . GCTB usually occurs in patients aged 20-45 years with equal frequency among male and female patients [2,3]. GCTB is typically considered as benign but locally aggressive lytic lesions. It is histologically characterized by osteoclast-like giant cells that express receptor activator of nuclear factor factor-κB (RANK), and mononuclear stromal cells that express RANK ligand (RANKL), a key mediator of osteoclast activation . The pathophysiological studies of GCTB show high concentrations of RANKL in the neoplastic stromal cells. It seems to be rational to use a potential RANKL inhibitor as a treatment option. Accordingly, there was a research on anti-RANKL treatment of patients with GCTB which found significantly effective .
Although considered to be benign tumors of bone, GCTB has a relatively high recurrence rate and continues to be the major problem [6,7]. Importantly, it is impossible to predict recurrence potential from the histological appearance alone. In spite of numerous attempts have been made to predict the clinical behavior of GCTB, there is no definite parameters to determine the risk of recurrence of GCTB [1,7,8]. As we already know the important role of RANK/ RANKL pathway in the pathogenesis of this disease, the effectiveness of RANK and RANKL expressions as prognostic markers in predicting the risk of recurrence of GCTB has never been reported. Despite there is an anti-RANKL therapy, surgery is still the primary treatment method for GCTB, especially in Srinagarind Hospital where Denosumab is prescribed only in some particular cases because of the expensive cost. So, we aimed to investigate the prognostic value of RANK and RANKL expression in giant cell tumor of bone in predicting recurrence which may change the treatment paradigm of the disease.
Materials and Methods
This was a retrospective case-control study. Of the cases referred to the Surgical Pathology Department of Srinagarind Hospital, Faculty of Medicine, Khon Kaen University between 2008 to 2017, 53 cases of GCTB were enrolled in the study. Hematoxylin-eosin (H&E) tissue sections of the resected tumors diagnosed with GCTB were re-evaluated. The paraffin blocks best representing the tumor were selected for immunohistochemical staining. Follow-up information was obtained from hospital files or by contacting with the patients. Data on age, gender, tumor location, tumor size, Campanacci Classification, surgical method, recurrence, primary metastasis, and lymphovascular invasion were reviewed. The potential relationship between the clinical and pathological findings and RANK and RANKL protein expressions were evaluated.
TMA immunohistochemical staining
Protein expression was evaluated by immunohistochemical (IHC) analysis on Tissue Microarray (TMA) of 53 paraffin-embedded GCTB samples with duplicated samples for each patient. H&E-stained slides were reviewed by the pathologist to choose representative areas for TMA construction. The sections were immunostained for RANK (clone ab13968, dilution 1 : 200, Abcam, Cambridge, United Kingdom), and RANKL (clone ab9957, dilution 1 : 2000, Abcam, Cambridge, United Kingdom) antibodies with the polymer DAB kit using the Ventana Benchmark XT automatic immunohistochemical stain device. As positive controls, tonsil was used for RANK, and lymph node was used for RANKL. No negative control was used in the study.
Staining was scored for intensity (0=no expression; 1= weak to moderate expression; 2=strong expression) and for percentage of positive cells (0<10%; 1=10–25%; 2=26–50%; 3>50). Cutoff levels for the sum of scores were applied as 0 for negative cases, 1–3 for weakly positive, and 4–5 for moderate to strong positivity that is considered as protein overexpression .
Recurrence-free survival (RFS) analysis was performed by Kaplan– Meier method and the difference between survival curves were calculated using the log-rank test. Cox’s proportional hazards model and binary logistic regression analyses were used to define the risk of recurrence. Follow-up was considered from the data of diagnosis to the first event (recurrence) or to the last follow-up in recurrence free patients. Statistical analysis was performed using SPSS version 22 software (SPSS Inc., Chicago, IL, USA), and p-values <0.05 were considered significant.
Patient demographics and clinical characteristics
Of 53 patients, there were 24 males (45.3%) and 29 females (54.7%) patients with a median age of 33 years (15-67 years). Most of the tumors were located in the non-axial region (86.8%). Surgical therapy consisted of curettage in 29 cases (54.7%) and resection in 24 cases (54.7%). 8 patients (15.1%) had recurrent disease with median recurrence interval of 34.5 months. All clinical data are listed in Table 1.
|Variables||Number of Patients (%)|
|Age (years old)||<20||5 (9.45%)|
|Tumor location||Axial||7 (13.2%)|
|Size in greatest dimension (cm)||<5 cm||10 (18.9%)|
|5-8 cm||32 (60.4%)|
|>8 cm||11 (20.8%)|
|Campannaci Classification||1||8 (15.1%)|
|Surgical method||Curettage||29 (54.7%)|
|Clinical course (first event)||Disease Free||45 (84.9%)|
|Local Recurrence||8 (15.1%)|
|Median Follow-up (months)||24|
|Median recurrence interval (months)||34.5|
Table 1: Patient demographics and clinical characteristics.
Histological features of human GCTB tissue sections
All tissue sections had high cellular density, with round or spindleshaped mononuclear stromal cells and numerous multinucleated osteoclast-like giant cells. Immunohistochemical analysis was performed on osteoclast-rich areas of GCTB TMA sections. RANK was predominantly stained on multinucleated cells, though immunostaining was also seen in mononuclear cells. RANKL was more uniformly expressed and more intense staining on mononuclear cells. Both RANK and RANKL were localized at cell membrane/cytoplasm (Figure 1).
Figure 1: A) H&E stain: Representative field of GCTB showing characteristic of multinucleated giant cells and round spindle shaped mononuclear cells (H&E, 200x) B) RANK was predominantly stained on multinucleated cells (RANK, 200x). C) RANKL was more uniformly expressed, with a higher percentage, and more intense staining of mononuclear cells (RANKL, 200x). Both RANK and RANKL were localized at cell membrane/cytoplasm.
Recurrence-free survival (RFS) analysis was performed by Kaplan– Meier method and the difference between survival curves were calculated using the log-rank test. The comparison between RFS curves based on age showed a statistically significant difference (log rank 10.749, p=0.005) with a maximum risk to develop recurrence within 96 months (Figure 2). Gender, tumor location, tumor size, Campannaci classification, surgical method (curettage versus resection) did not significantly associated with recurrence (Table 2).
|Variables||Clinical Course||Univariate analysisa||Multivariate analysisb|
|Without recurrence||With recurrence||Recurrent rate (%)||X2||p-values||HR (95% CI)||p-values|
|RANK expression||Negative-Weak||23||2||8||4.055||0.044*||7.025 (0.833-59.259)||0.073|
|RANKL expression||Negative-Weak||22||1||4.34||5.187||0.023*||0.019 (<0.001-9.825)||0.214|
|RANK/RANKL Co-overexpression||With Co-overexpression||13||6||31.58||7.541||0.006*||10.386 (1.246-86.553)||0.031|
|Note: alog-rank test, bCox’s proportional hazards model, HR- Hazard Ratio|
Table 2: Univariate and multivariate analysis of prognostic factors in GCTB.
Figure 2: Kaplan-Meier analysis: (A) The curves illustrates the recurrence-free survival according to age. The curves illustrate the metastasis-free survival according to (B) RANK and (C) RANKL expression. Range 0–3 indicates a low or absent immunoreactivity; range 4–5 indicates a moderate to strong immunoreactivity. (D) The curves illustrate the recurrence-free survival according to immunoreactivity of RANK/RANKL co-overexpression.
In the patients with recurrent disease, RANK and RANKL were strongly expressed (range 4–5) in 75% and 87.5% of cases, respectively. The rate for RANK and RANKL overexpression was 21.43% and 23.33% compared to 8% and 4.34% in non-recurrence cases, respectively. Differences between RFS curves based on their immunostains were statistically significant (log rank 4.055, p=0.044 for RANK and log rank 5.187, p=0.023 for RANKL) (Table 2).
Interestingly, 6 out of 8 recurrent tumors demonstrated a strong positivity for RANK/RANKL co-overexpression. The rate for RANK and RANKL co-overexpression was 31.58% compared to 5.88% in nonrecurrence cases (Table 2). We established different risk classifications showing a significant higher probability of recurrence in the group of patients with co-overexpression of both proteins (log rank 7.541, p=0.006) as compared to cases without co-overexpression (Figure 2).
When Cox’s proportional hazards model was used for those significant parameters to univariate analysis, it emerged that the only significant prognostic parameter capable of defining an increased risk for recurrence was co-overexpression of RANK/RANKL (hazard ratio 2.910; 95%CI 1.295-90.068; p=0.028) (Table 2). Binary logistic regression multivariate analysis followed by ROC analysis confirmed that co-overexpression of RANK/RANKL represented a significant biological model to predict recurrence (area under the curve=0.731±0.098; 95% CI 0.539–0.922, p=0.039) (Figure 3).
Giant cell tumor of bone (GCTB) is a benign bone tumor. Although classified as a benign tumor, it has a tendency for recurrence. The recurrence rate differs between 5-56% and continues to be the major problem . In our study, there were 15.1% of GCTB cases with recurrence disease. However, the risk of recurrence could not be predicted by histology alone. Cheng et al. (2015) has suggested that Ki-67 (p=0.042), CD147 (p=0.010), pathological fracture (p=0.002), Campanacci grade (P<0.001) and surgical method (p=0.030) were the factors influencing the recurrence of GCTB . On the other hand, Yalcinkaya et al. identified that the Ki-67 index was higher in cases with recurrence, but this difference was not statistically significant (p= 0.731) . Pan Hu et al. (2016) mentioned that the recurrence rate was the highest in patients treated with intralesional curettage (53.4%), and the lowest in those treated with resection (4.9%). They found a higher risk of tumor recurrence in the proximal fibula compared to the distal femur (hazard ratio: 28.52, 95% confidence interval: 5.88–138.39; p<0.001), and in patients treated with curettage compared to those treated with resection (hazard ratio: 12.07, 95% confidence interval: 4.99–29.18; p<0.0001) . However, our study showed that sex, age, tumor location, tumor size, Campannaci classification, and surgical method were not considered as the prognostic factors for GCTB recurrence from multivariate analysis.
To date, no definite prognostic marker for the risk of recurrence has been proposed. As we already know that RANK and RANKL have important role in GCTB pathogenesis. It is logical to seek for significant prognostic markers of recurrence from this pathway. To identify the prognostic markers for the risk of GCTB recurrence, we detected the expression of RANK and RANKL using immunohistochemistry in 53 cases. Univariate analysis revealed that age, RANK and RANKL overexpression and the co-overexpression of RANK/RANKL were statistically significantly associated with recurrence. However, the multivariate analysis suggested that only the co-overexpression of RANK/RANKL could be the prognostic marker for predicting recurrence.
RANKL is a membrane-bound TNF-related factor expressed by osteoblast-like stromal cells. Osteoclast precursors and mature osteoclasts express RANK, a membrane-bound TNF receptor that recognizes RANKL through direct cell-to-cell interaction with osteoblast-like stromal cells. Finally, osteoprotegerin (OPG), a secreted decoy receptor that belongs to the TNF receptor family and lacks the transmembrane domain of RANK, recognizes RANKL, thereby blocking the RANK-RANKL interaction and inhibiting osteoclast differentiation and activation. Dysregulation of RANKL, RANK, and osteoprotegerin (OPG) signaling cascade induces the imbalance between bone formation and bone resorption, which leads to the changes in bone mass and increases osteoclast-mediated bone destruction. In giant cell tumors of bone, the mesenchymal stromal cells are the neoplastic cells and induce recruitment and formation of osteoclasts. Osteoclasts typically are present in large numbers in GCTB, suggesting that these tumors may contain cells expressing factors that stimulate osteoclast precursor recruitment and differentiation. The pathophysiological studies of GCTB show high concentrations of RANKL in the neoplastic stromal cells [2,4,10-12]. In 2002, Roux et al. provided the first immunohistochemical evidence that RANK and RANKL were expressed in these tumors at the protein level, with RANK being expressed by osteoclast-like giant cells, and RANKL by the mononuclear cells that form the mesenchymal component of GCTBs. These findings were consistent with the involvement of RANKL and RANK expression in the pathogenesis of GCTs, as the overexpression of RANKL by tumor stromal cells could be responsible for the formation of numerous osteoclasts within the tumor . Therefore, RANK and RANKL could be associated with recurrent disease of GCTB. With the recognition of RANKL in the pathogenesis, investigations began into the poten¬tial curative effect of suppression of this molecule. The result was a human monoclonal IgG2 antibody against RANKL, Denosumab, which is a human monoclonal antibody that binds to RANKL and inhibits its binding to RANK, and preventing RANK activation [2,4,5,14].
Now-a-days, surgery is the primary treatment method for GCTB. When GCTB is unresectable, or when surgery is likely to result in severe morbidity, Denosumab, a RANKL inhibitor can be used . However, recurrence is still a serious problem. According to our study, anti-RANKL may help preventing local recurrence in the cases with RANK/RANKL co-overexpression as we identified that the cooverexpression of RANK/RANKL could be the prognostic marker for the risk of recurrence. The median survival time compared to cases without co-overexpression was 96 and 57 months, respectively.
The limitation of our study may be counted as the following: limitation of the number of patients and heterogeneous distribution of cases. To evaluate medical treatment efficiency of this tumor, multicenter studies with more homogenous groups or prospective design should be considered.
Our results suggested that RANK and RANKL co-overexpression increases the risk of recurrence of GCTB and could be a prognostic marker for recurrent disease. This could provide useful information for new treatment paradigm of the disease. Further studies with a larger number of cases may result in a better understanding and more information regarding this matter.
The authors have no conflicts of interest to this work.
- Cheng DD, Hu T, Zhang HZ, Huang J, Yang QC (2015) Factors affecting the recurrence of giant cell tumor of bone after surgery: A clinicopathological study of 80 cases from a single center. Cell Physiol Biochem 36: 1961-1970.
- Dufresne A, Derbel O, Cassier P, Vaz G, Decouvelaere AV (2012) Giant-cell tumor of bone, anti-RANKL therapy. Bonekey Rep 1: 149.
- Fletcher CD, Bridge JA, Hogendoorn P, Mertens F (2013) WHO classification of tumors of soft tissue and bone, 4th Edition. Lyon Cedex, France: IARC.
- Ueda T, Morioka H, Nishida Y, Kakunaga S, Tsuchiya H, et al. (2015) Objective tumor response to denosumab in patients with giant cell tumor of bone: A multicenter phase II trial. Ann Oncol 26: 2149-2154.
- Branstetter DG, Nelson SD, Manivel JC, Blay JY, Chawla S, et al. (2012) Denosumab induces tumor reduction and bone formation in patients with giant-cell tumor of bone. Clin Cancer Res 18: 4415-4424.
- Deveci MA, Paydaş S, Gönlüşen G, Özkan C, Biçer ÖS, et al. (2016) Clinical and pathological results of denosumab treatment for giant cell tumors of bone: Prospective study of 14 cases. Acta Orthop Traumatol Turc 51: 1-6.
- Hu P, Zhao L, Zhang H, Yu X, Wang Z, et al. (2016) Recurrence rates and risk factors for primary giant cell tumors around the knee: A multicentre retrospective study in China. Sci Rep 6: 36332.
- Yalcinkaya U, Ugras N, Kabul S, Ocakoglu G, Bilgen MS (2015) Prognostic value of p53 protein expression in giant cell tumor of bone. Pol J Pathol 66: 389-396.
- Quattrini I, Pollino S, Pazzaglia L, Conti A, Novello C, et al. (2015) Prognostic role of nuclear factor/IB and bone remodeling proteins in metastatic giant cell tumor of bone: A retrospective study. J Orthop Res 33: 1205-1211.
- Demirsoy U, Karadogan M, Selek Ö, Anik Y, Aksu G, et al. (2014) Golden bullet-denosumab: Early rapid response of metastatic giant cell tumor of the bone. J PediatrHematol Oncol 36: 156-158.
- Wada T, Nakashima T, Hiroshi N, Penninger JM (2006) RANKL-RANK signaling in osteoclastogenesis and bone disease. Trends Mol Med 12: 17-25.
- Singh AS, Chawla NS, Chawla SP (2015) Giant-cell tumor of bone: treatment options and role of denosumab. Biologics 9: 69-74.
- Roux S, Amazit L, Meduri G, Guiochon-Mantel A, Milgrom E, et al. (2002) RANK (receptor activator of nuclear factor kappa B) and RANK ligand are expressed in giant cell tumors of bone. Am J Clin Pathol 117: 210-216.
- Goldschlager T, Dea N, Boyd M, Reynolds J, Patel S, et al. (2015) Giant cell tumors of the spine: Has denosumab changed the treatment paradigm? J Neurosurg Spine 22: 526-533.
Citation: Atsawaphidsawat N, Ungarreevittaya P, Sumananont C (2018) Prognostic Role of RANK and RANKL Expression in Recurrent Giant Cell Tumor of Bone: A Retrospective Study. J Mol Biomark Diagn 9: 394. DOI: 10.4172/21559929.1000394
Copyright: © 2018 Atsawaphidsawat N, et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Select your language of interest to view the total content in your interested language
Share This Article
- Total views: 794
- [From(publication date): 0-2018 - Dec 18, 2018]
- Breakdown by view type
- HTML page views: 761
- PDF downloads: 33