Introduction
Primary malignant bone tumors (PMBTs) are rare worldwide, accounting for between 0.2% and 0.5% of all malignancies [ 1 – 4 ]. In a global study by Kumar and Gupta, PMBTs had a variable geographical incidence [ 2 ]. The latter suggests that the environment plays a role in health and tumor dynamics, in particular. Similarly, patient age and sex seem to play a role in the incidence of PMBT, especially in childhood [ 2 , 3 , 6 ]. In some studies, young and teenage boys during growth spurts are more likely to develop primary malignant bone tumors, with a male-to-female ratio of 2: 1 [ 2 , 3 , 6 , 7 ].
In our local South African orthopedic context, Pillay et al found osteosarcoma to be the most common PMBT treated at their university hospital [ 4 ]. They postulated socio-economic and/or environmental factors to be involved in the development of osteosarcoma [ 4 ]. However, they did not look at the environment as a risk factor in terms of the exposure to and role of agricultural pesticides and their association with PMBT. Other risk factors previously highlighted in the literature that are somewhat linked to geographical and environmental exposure are vitamin D deficiency and exposure to ionizing radiation [ 2 , 8 ]. In the pediatric population, osteosarcoma is the most common PMBT, followed by Ewing sarcoma, with the latter showing a predilection for children of Caucasian descent [ 1 – 4 , 7 , 8 ]. In a study by Ramdass et al, race was seen as a stronger risk factor for PMBT in mixed-race communities, as PMBTs had a racial predilection even for children living within the same island community [ 3 ]. In the adult population, chondrosarcoma is consistently the most common PMBT reported [ 2 – 4 , 8 ]. However, fibrosarcoma was found to be more common than chondrosarcoma in a study done by Ode et al in Nigeria on adult patients, further highlighting the influence that environmental factors have on PMBT development [ 8 ].
Osteosarcoma outcomes are well researched in the literature, with obvious improvement in overall survival and surgical outcomes over the past 3 decades [ 5 , 6 , 9 ]. Novel chemotherapy protocol success led to the possibility of limb salvage treatment algorithms [ 6 ]. Before this milestone, patients were destined for limb ablation surgery after diagnosis. Treatment reviews from developed countries are reporting a >60% treatment success rate, with survival rates of up to 82% at 5 years for chemotherapy-sensitive osteosarcoma [ 9 ]. The latter 2 figures have improved significantly from figures reported in the 1960s and 1970s [ 9 ]. Ewing sarcoma management protocols closely resemble and adapt, to an extent, from osteosarcoma management and overall outcomes, with poorer surgical outcomes and a 5-year survival rate of around 40% for both [ 10 ]. With chondrosarcoma, sensitivity to chemotherapy is generally poor; however, success rate and 5-year survival rates are reported around 75% after surgical resection and radiotherapy [ 11 ]. Multiple myeloma (MM) treatment success rates are better in young healthy individuals (<65 years) than in elderly patients with comorbidities [ 12 ]. Bone marrow ablation with autologous stem cell transplant remains the criterion standard in the former group of patients [ 12 ]. Unfortunately, Pillay et al, in their study, did not classify MM as a PMBT and rather they considered MM a hematological malignancy [ 4 ]. However, the WHO classification (2020) of tumors of the bone classifies plasmacytomas and MM as hemopoietic tumors of bone [ 13 ]. MM is defined as a primary malignant bone tumor in our study, together with other tumors that were analyzed. To a lesser extent, institutional variability in the definitions of PMBTs can also add to the role played by the environment when reporting on the epidemiology of PMBTs. To the best of our knowledge at the time of this study, PMBTs have not been studied locally concerning their geographical variation and their association with agricultural pesticides. Therefore, our study aimed to retrospectively study children and adults treated for PMBTs at our center between 2016 and 2021.
Material and Methods
The Wits Human (Medical) Research Ethics Committee clearance (study number: M200758) was issued prior to data collection. Our study was a retrospective review of data from patients with PMBTs treated at our university hospital’s tumor and infection units. All tumor diagnoses followed clinical and radiological workup and were confirmed with tissue histological analysis of the biopsied tissue. Admission, surgery, and laboratory records were searched for data extraction. Demographic, medical, pathological, and treatment outcomes data were collected. All data were stored using Microsoft Excel.
Inclusion and Exclusion Criteria
Records of all pediatric and adult patients with a diagnosis of PMBTs during the study period were included in the study. We excluded all records for patients treated for secondary malignant bone, as well as those with primary soft tissue and secondary bone tumors. Records with inadequate and inconclusive histological results were also excluded.
Data Analysis
Microsoft Excel and Stata 18.0 software were used for descriptive statistical analysis.
Results
A total of 123 patient records met the inclusion criteria for our study. Our study population age ranged between 3 and 89 years; Figure 1 shows the age distribution curve. The mean age was 35.38 years, with a median age of 46 and a mode of 17 years, respectively. There was a slight female dominance over males, with a female: male percentage ratio of 58.5%: 41.8%. Figure 2 shows a pie chart depicting patient demographics by race. Our patient cohort was predominantly made up of patients from the Gauteng province, followed by the North-West province ( Figure 3 ). Only 1 individual in our cohort reported positive exposure to agricultural pesticides, with the rest denying knowledge or recollection of such exposure. Only 3.7% (7 of 107 respondents) had a positive family history of cancer.
Clinical Presentation
Pain was the most common symptom, presenting either in isolation or in combination with the development of a mass and/or a pathological fracture. Pain was reported in 93% (114/123) of patients. The development of a mass was reported in 64% (79/123) cases, while a fracture was diagnosed in 17% (21/123) cases only.
Body Part Affected
The lower limb bones were affected 3: 1 to the upper limb bones. A total of 99 lower limb bones were affected, and 33 upper limb bones were affected in our cohort. One case was recorded as unknown for the body part. There was a slight predilection for the right-hand side of the body, with 54% (71 out of 131 bones) of tumors affecting the right-hand side.
Tumor Diagnosis
Osteosarcoma was the most common PMBT diagnosis, accounting for 57.7% of all tumors. MM was the second most common, with 24.3% of patients having this tumor. The least common tumor in our cohort was synovial sarcoma, affecting only 1 patient. Table 1 depicts the PMBT by frequency.
Chemotherapy Response
Table 2 summarizes our findings on chemotherapy response. Twenty-one percent of all tumor types were classified as unknown (response not assessable). Ewing sarcoma had a 75% sensitivity rate, while all the others had poor sensitivity to chemotherapy.
Surgical Outcomes
Thirty-six percent of our participants received non-operative care, with the remaining undergoing definitive surgical intervention. Figures 4 – 7 show the categories and subcategories under both groups of intervention. At assessment, 54.7% of patients were deemed for palliative care, with patients receiving non-operative treatment or palliative amputation of a limb.
Discussion
PMBTs are rare and show a variable geographical incidence in each unique environment [ 1 – 5 , 8 ]. An association between agricultural pesticides and PMBTs has not been researched in our context. Agricultural pesticides have been shown to have a weak association with development of soft tissue tumors [ 14 ]. Freeman et al reported no association observed between lymphoma of bone, MM, and pesticide exposure [ 15 ]. In the present study, only 1 patient had prior knowledge of exposure to pesticides; therefore, we could not retrospectively establish any association between PMBTs and pesticide exposure in our cohort of patients.
Osteosarcoma was by far the most common PMBT in our cohort. The literature consistently reports osteosarcoma as the most common PMBT [ 1 – 5 , 8 , 16 , 17 ]. MM was the second most common PMBT in our cohort and in another African study by Sakala et al [ 16 ]. We defined and included MM as a PMBT in our study. Based on our definition, MM was the most common PMBT in our older population group. Pillay et al and Palmerini et al excluded MM in reports, suggesting they defined MM as a hematological malignancy [ 4 , 18 ]. Ewing sarcoma, although rare in Africans, was the second most common PMBT in our pediatric population group; however, it accounted for only 4 cases. Sofulu et al had 30 cases of Ewing sarcoma over 8 years [ 19 ]. The lower rates of Ewing sarcoma reported in Africa could be due to the lack of Musculoskeletal Tumor Registry data on the continent and the predominant dedication of resources for osteosarcoma care [ 1 – 5 , 7 , 8 ]. Primary chondrosarcoma was the second most common PMBT in the adult population in our study and in a meta-analysis on bone tumors in Nigeria by Koyejo et al [ 20 ]. However, chondrosarcoma typically ranks as the second most common PMBT in adulthood, typically when MM is defined as a PMBT, and the most common PMBT in adults when MM is considered a hematological malignancy [ 17 , 18 ]. Deka et al included metastatic bone disease in their review of malignant bone tumors, and chondrosarcoma was the third most common PMBT after metastatic bone disease and osteosarcoma. Surprisingly, MM was the fourth most common PMBT in that study. The latter highlights the role and effect that definitions and geography/environment have on the epidemiological picture of PMBTs.
Pain was the most reported symptom either in isolation or in combination with a mass or pathological fracture in our cohort. Pain was reported commonly in our patients, probably due to presentation with large tumor sizes and associated pathological fractures. The appearance of a swelling (mass) was the second most common symptom, and a pathological fracture was the least common symptom in isolation. Oyedemade et al found results similar to those in our study, with 85% of their patients presenting with a painful mass, while only 16.5% had a pathological fracture at presentation. However, Lingerih et al, in their study in Ethiopia, found a mass/swelling to be the most frequent symptom, followed by pain [ 21 ]. In another study by Sakala et al, pain as a standalone symptom accounted for 27% of symptoms at presentation. Swelling alone was noted in 17% of cases, and pathological fracture was noted in 20% of cases, although it was always associated with pain and/or swelling [ 16 ]. The variation in presentation findings is likely due to a difference in time to presentation by patients as well as variation in study population demographics and the variable tumor diagnoses unique to each community. Additionally, Papagelopoulos et al showed that tumor diagnosis and the anatomical location affected by the tumor influenced pathological fracture development [ 22 ].
Numerous factors influence a tumor’s chemotherapy response evaluation. Our cohort of PMBTs was documented to be chemosensitive in 33% of cases. Chemoresistance was noted in 45% of that cases. The latter, together with patients who discontinued and those with no documented response, accounted for a substantial proportion of our cohort. This poor response was also likely related to the late time of presentation seen in our patient population. Equally, large tumors are prone to requiring emergency surgery, disrupting the chemotherapy treatment protocols before chemosensitivity assessment is possible. Late presentation and larger tumor size have been associated with a poor chemotherapy response, especially for osteosarcoma and Ewing sarcoma, with a tumor size >10 cm being a predictor of poor response and, therefore, outcomes [ 6 , 19 , 23 ]. Our patients with osteosarcoma were sensitive to chemotherapy in 38% of cases, a low figure as compared with the 70% chemosensitivity in the literature [ 24 ]. However, a dedifferentiation percentage >20% in chondrosarcoma is more predictive of chemotherapy response than tumor size [ 25 ]. Notwithstanding that, chondrosarcoma generally has a poor response to chemotherapy and radiotherapy, with both modalities reserved mainly for palliation [ 11 , 26 , 27 ]. Effectively, only Ewing sarcoma was sensitive to chemotherapy, in keeping with the literature [ 6 , 9 , 10 ]. All other PMBTs in our study had a poor response to chemotherapy.
Our mortality rate, when considering the number of discontinued and unknown participants, respectively, was “functionally” 37% (45/123) at data collection, with a 5-year survival of 63%. Sofulu et al also had an overall 5-year survival of 57%, while Wang et al reported a 77% 5-year survival rate [ 19 , 28 ]. The latter figures are in keeping with our findings on the cohort. Lisenda et al also reported a 20% absconding rate in their study, although with patients with osteosarcoma [ 29 ]. Irresectable pulmonary metastasis at presentation is a poor predictor of outcomes in osteosarcoma [ 27 ]. No study evaluated the combined mortality of all PMBTs locally. The latter may be due to poor infrastructure, the recent expansion of orthopedic oncology services and, more importantly, a lack of established registry databases for PMBTs in our context. Our study was limited by poor medical records, which is typical of retrospective studies in our context. The years 2020 and 2021 affected by COVID-19 restrictions may have influenced our findings, as services were halted. A longer follow-up study period or a controlled prospective study could yield different results.
Conclusions
PMBTs are rare. The most common PMBT at our hospital is osteosarcoma. There were variations in the incidence of the other PMBTs, compared with those in the literature. We had a poor patient chemotherapy response for most PMBTs, due to death, discontinuation, and late presentation by patients. The majority of our patients had no knowledge of prior exposure to agricultural pesticides.
Acknowledgements
We thank Dr Machuene Ananias Poopedi for statistical guidance and overview.
Conflict of interest: None declared
Declaration of Figures’ Authenticity: All Figures submitted have been created by the authors, who confirm that the images are original with no duplication and have not been previously published in whole or in part.
Financial support: None declared
References
1 Deka KM Talukdar A A retrospective analysis of bone tumours and tumour-like lesions: A hospital based study of 76 cases Int J Res Med Sci 2017 5 11 4915 18
2 Kumar M Gupta B Global incidence of primary malignant bone tumours 2016 Available from: www.c-orthopaedicpractice.com last accessed: 05/04/2025
3 Ramdass MJ Mooteerm J Beharry A Mencia M An 8-year Analysis of bone tumours in a Carribean island Ann Med Surg 2015 4 414 16 10.1016/j.amsu.2015.10.009 PMC4720716 26904191
4 Pillay Y Ferreira N Marais LC Primary malignant bone tumours: Epidemiological data from an Orthopaedic Oncology Unit in South Africa SA Orthop J 2016 14 4 12 16
5 Oyemade GAA Abioye AA Primary malignant bone tumors of bone: Incidence in Ibadan, Nigeria J Natl Med Assoc 1982 74 1 65 68 6956750 PMC2552825
6 Xin S Wei G Prognostic factors in osteosarcoma: A study level meta-analysis and systematic review of current practice J Bone Oncol 2020 21 100281 32140401 10.1016/j.jbo.2020.100281 PMC7047183
7 Rodvall Y Dich J Wiklund K Cancer risk in offspring of male pesticide applicators in agriculture in Sweden Occup Environ Med 2003 60 798 801 14504372 10.1136/oem.60.10.798 PMC1740388
8 Ode MB Misauno MA Nwadiaro HC Pattern and distribution of primary bone tumours in Jos Nigeria J Dent Med Sci 2014 13 12 9 12
9 Misaghi A Goldin A Awad M Kulidjian AA Osteosarcoma: A comprehensive review SICOT J 2018 4 12 29629690 10.1051/sicotj/2017028 PMC5890448
10 Bosma SE Ayu O Fiocco M Prognostic factors for survival in Ewing sarcoma: A systematic review Surg Oncol 2018 27 4 603 10 30449479 10.1016/j.suronc.2018.07.016
11 Amer KM Munn M Congiusta D Survival and prognosis of chondrosarcoma subtypes: SEER database analysis J Orthop Res 2020 38 2 311 19 31498474 10.1002/jor.24463
12 Turesson I Bjorkholm M Blimark CH Rapidly changing myeloma epidemiology in the general population: Increased incidence, older patients, and longer survival Eur J Haematol 2018 101 237 44 10.1111/ejh.13083 PMC6195866 29676004
13 Choi JH Ro JY The 2020 WHO classification of tumors of bone: An updated review Adv Anat Pathol 2021 28 119 38 33480599 10.1097/PAP.0000000000000293
14 Odes EJ Randolph-Quinney PS Steyn M Earliest hominin cancer: 1.7-million-year-old osteosarcoma from Swartkrans Cave, South Africa S Afr J Sci 2016 112 7/8 1 5
15 Freeman LEB Rusiecki JA Hoppin JA Atrazine and cancer incidence among pesticide applicators in the agricultural health study Environ Health Perspect 2011 119 9 1253 59 21622085 10.1289/ehp.1103561 PMC3230407
16 Sakala D Munthali JC Mulla Y Primary malignant bone tumours at the University Teaching Hospital in Lusaka Zambia Med J Zambia 2016 43 1 24 30
17 Redondo A Bagué S Bernabeu D Malignant bone tumours (other than Ewing’s): Clinical practice guidelines for diagnosis, treatment and follow-up by Spanish Group for Research on Sarcomas (GEIS) Cancer Chemother Pharmacol 2017 80 1113 31 29038849 10.1007/s00280-017-3436-0 PMC5686259
18 Palmerini E Righi A Staals EL Rare primary malignant bone sarcomas Cancers (Basel) 2020 12 11 3092 33114111 10.3390/cancers12113092 PMC7690832
19 Sofulu Ö Erol B Evaluation of factors affecting survival rate in primary bone sarcomas with extremity and pelvis involvement Acta Orthop Traumatol Turc 2020 54 3 234 44 32544060 10.5152/j.aott.2020.03.79 PMC7586774
20 Koyejo TT Olusunmade OI Olufemi OT Epidemiology of primary bone tumours in Nigeria: A systematic review SA Orthop J 2022 21 3 167 71
21 Lingerih T Yeshiwas S Mohamedsaid A Arega G Patterns and treatment outcomes of primary bone tumors in children treated at tertiary referral hospital, Ethiopia BMC Cancer 2024 24 1 394 38549062 10.1186/s12885-024-12169-x PMC10976724
22 Papagelopoulos PJ Mavrogenis AF Savvidou OD Pathological fractures in primary bone sarcomas Injury 2008 39 4 395 403 18061185 10.1016/j.injury.2007.07.018
23 Prabowo Y Iwan S Kamal AF Correlation between prognostic factors and the histopathological response to neoadjuvant chemotherapy in osteosarcoma: A retrospective study Int J Surg Oncol 2021 3 1 10 10.1155/2021/8843325 PMC8096583 33996154
24 Aoki Y Han Q Kubota Y Super methotrexate-resistant osteosarcoma cells retain their sensitivity to recombinant methioninase: Targeting methionine addiction to overcome extreme cancer-chemotherapy resistance Anticancer Res 2025 45 3 929 34 40037887 10.21873/anticanres.17480
25 Hung YP Chebib I Bredella M Progonostic significance of percentage and size of dedifferentiation in dedifferentiated chondrosarcoma Mod Pathol 2023 36 3 100069 36788104 10.1016/j.modpat.2022.100069
26 Lohberger B Glaenzer D Eck N Effects of a combined therapy of bortezomib and ionizing radiation on chondrosarcoma three-dimensional spheroid cultures Oncol Lett 2021 21 6 428 33868466 10.3892/ol.2021.12689 PMC8045153
27 Rozeman LB Cleton-Jansen AM Hogendoorn PCW Pathology of primary malignant bone and cartilage tumours Int Orthop 2006 30 6 437 44 16944143 10.1007/s00264-006-0212-x PMC3172744
28 Wang Z Li S Li Y Prognostic factors for survival among patients with primary bone sarcomas of small bones Cancer Manag Res 2018 1191 99 29795990 10.2147/CMAR.S163229 PMC5958943
29 Lisenda L Linda ZA Snyman FPJ Osteosarcoma outcomes at a South African tertiary hospital S Afr Med J 2017 107 9 754 57 28875882 10.7196/SAMJ.2017.v107i9.11424