Skeletal Radiology (2021) 50:1111–1116

[Link]

SCIENTIFIC ARTICLE

Bone biopsy results in patients with a history of malignancy:

a case series of 378 patients
Rhys Morris 1 & Karen Shepherd 1 & Gillian Cribb 1 & Jaspreet Singh 2 & Prudencia Tyrrell 2 & Paul Cool 1,3

Received: 13 July 2020 / Revised: 13 October 2020 / Accepted: 13 October 2020 / Published online: 23 October 2020
# ISS 2020

Abstract
Objective The value of a bone biopsy in patients who present with a bone lesion and past medical history of malignancy is
uncertain. The objective of this study was to evaluate the outcome of bone biopsies in patients with a history of a malignancy
undergoing bone biopsy of a lesion in a regional bone tumour unit. Secondary outcomes include the assessment of survival in the
different outcome groups.
Materials and methods This was a retrospective study of patients, with a previous malignancy and suspicious bone lesions, who
underwent bone biopsy for final diagnosis between March 2010 and September 2019. Results of the biopsy were summarized
into 3 groups: confirmed original malignancy, confirmed benign diagnosis, and confirmed new malignancy. Survival analysis of
each group was also undertaken.
Results A total of 378 patients met the inclusion criteria (mean age 64 years, 216 females (57%)). In 250 cases (66%), the original
malignancy was confirmed on the bone biopsy; in 128 cases, an alternative diagnosis was confirmed (benign diagnosis in 69
(18%)), and 59 had a new malignancy (16%). Survival was significantly greater for those in whom a benign diagnosis was
confirmed (logrank test p = 0.0100).
Conclusion This study shows that for patients presenting with a suspicious bone lesion, together with a history of malignancy, in
a third of cases, the bone biopsy will confirm an alternative diagnosis of a benign lesion or a new malignancy. Survival of these
patients will vary significantly depending on the biopsy outcome.

Keywords Metastatic bone disease . Bone biopsy . Bone lesion . Cancer

Introduction result in an increased burden of disease for the orthopaedic

and oncological communities [5].
Advances in the early detection and treatment of cancer One key dilemma facing orthopaedic surgeons is how to
have contributed to improved survival rates in many coun- manage patients with a history of cancer and a new finding
tries [1, 2]. However, this growing population of cancer of a suspicious bone lesion. UK guidance from British
survivors creates new challenges for clinicians including Orthopaedic Oncology Surgery (BOOS) [6] advises con-
orthopaedic surgeons. Bone is the third most common site sideration of a bone biopsy in cases of uncertainty.
of distant metastases [3], and around 80% of patients with However, there is a tendency to assume that the bony lesion
metastatic disease will have metastatic bone disease (MBD) is secondary to the previous malignancy [7–9] and manage
[4]. The increase in the prevalence of MBD will inevitably accordingly, often using scoring systems for guidance, for
example Mirels score [10].
Several studies in different populations have attempted
* Rhys Morris to quantify the frequency of an alternative diagnosis i.e.
rm502@[Link] benign lesion or a new malignancy compared with the orig-
inal cancer (Table 1) [3, 8, 9, 11–14]. These studies show
1
Montgomery Unit, The Robert Jones and Agnes Hunt Orthopaedic considerable variation with rates ranging from 2% [12] to
Hospital, Oswestry, Shropshire SY10 7AG, UK over a third [14]. Inaccurate diagnoses of the nature of new
2
Department of Radiology, The Robert Jones and Agnes Hunt bone lesions may have significant implications for the pa-
Orthopaedic Hospital, Oswestry, Shropshire SY10 7AG, UK tient, with potentially catastrophic consequences of inap-
3
Keele University, Newcastle, Staffordshire ST5 5BG, UK propriate management (intramedullary nailing of a primary
1112 Skeletal Radiol (2021) 50:1111–1116

Table 1 Summary of previous

studies assessing bone biopsy Author Country Year of Number of Alternative Benign New
results in patients with a previous of study Publication patients diagnosis (%) diagnosis malignancy
malignancy and new bone lesion (%) (%)

Liang[3] China 2019 117 21.3 17.9 3.4

Monfardini[11] Italy 2014 290 15.5 15.5 0
Raphael[8] USA 2013 434 27.2 23.7 3.5
Toomayan[9] USA 2011 93 11.8 4.3 7.5
Cronin[12] Ireland 2009 43 2.3 0 2.3
Clayer[13] Australia 2006 50 22 4 18
Aoki[14] Japan 2005 35 34.3 34.3 0

bone tumour). Additionally, the prognosis for patients can clear history of malignancy, soft tissue biopsy only and
vary significantly depending on the diagnosis, which may non-diagnostic samples only.
influence other aspects of management. It is known, for
example, in breast cancer that a third of cases have altered Technical aspects
oestrogen (ER) or progesterone (PgR) receptor status and
15% change in Herceptin receptor (HER2) during the dis- All cases were discussed at a bone and soft tissue tumour
ease course [15]. multidisciplinary team (MDT) to plan biopsy and discuss dif-
This study aims to compare the diagnosis of new bone ferential diagnoses. All biopsies were performed under CT
lesions, using bone biopsy, in a cohort of patients with a his- guidance by a musculoskeletal (MSK) radiologist (consultant
tory of previous malignancy. The primary outcome recorded or senior fellow under consultant supervision) under general
is the rate with which an alternative diagnosis (benign or new anaesthesia (GA) or regional anaesthesia and using wide-bore
malignancy) occurs. Secondary outcomes include the assess- needles (11G). In a small number of cases, a needle biopsy or
ment of any differences in survival for patients where the bone open biopsy was performed in an operating theatre using X-
biopsy result differs from the original tumour. ray guidance by the surgical team.
Biopsy samples were reported by an MSK consultant his-
topathologist. Following further MDT discussion, the final
diagnosis was recorded.
Methods

This was a retrospective study utilizing prospectively collected Data analysis

data on the tumour database for the tumour unit in a regional
bone and soft tissue tumour unit. The study is part of a larger During data collection, the MDT diagnoses were summarized
project approved by the Health Research Authority (20/HRA/ into 3 groups: confirmed original malignancy, confirmed be-
4857). All patients between March 2010 and September 2019 nign diagnosis, or confirmed new malignancy. Further details
who underwent a biopsy were reviewed for possible inclusion in of the malignancy were collected where relevant. In a small
the study. number of cases, the biopsy confirmed a malignancy but the
Patients were reviewed for inclusion via the unit data- exact origin of the primary tumour was unspecified, for exam-
base and subsequently according to the electronic patient ple metastatic carcinoma of upper GI origin. In these cases, if
record. Inclusion criteria included a bone biopsy performed the report clearly demonstrated the malignancy to be different
in any patients with a suspicious bone lesion and a clear from the original, it was categorized as a new malignancy (and
history of malignancy. Bone lesions were regarded as sus- subclassified as unknown origin). Survival data were obtained
picious if after MDT discussion with MSK radiologists and from the tumour database and NHS Digital.
tumour surgeons there was concern regarding aggressive
features of a possible malignancy [16]. Malignancy defini- Statistical analysis
tion included carcinoma, sarcoma and haematological.
Also included in the analysis were cases where the initial Data was extracted from the database and collated on a
biopsy was non-diagnostic, but further bone biopsies re- Microsoft Excel® spreadsheet. Analysis was performed using
vealed a clear diagnosis. In any cases of uncertainty, patient R statistical software (R Foundation, Vienna). Categorical
notes and radiological and histology records were reviewed outcomes were evaluated using Chi-squared tests. Logrank
in further detail. Exclusion criteria were patients with no and Cox proportional hazards were used to analyse Kaplan-
Skeletal Radiol (2021) 50:1111–1116 1113

Meier (KM) survival curves for the groups. A p value of less Table 3 Details of new
malignancy Diagnosis Frequency (%)
than 0.05 was deemed statistically significant.
Unknown primary 11 (19)
Myeloma 9 (15)
Results Lung 8 (14)
Chondrosarcoma 7 (12)
A total of 1517 bone biopsies were performed during the Osteosarcoma 6 (10)
study period. The inclusion criteria were met in 378 patients Renal Cell Ca 4 (7)
(350 with one previous malignancy, 24 with two malignancies Lymphoma 2 (3)
and four cases with three previous malignancies). There were Ewing’s sarcoma 2 (3)
216 female patients (mean age 61.8 years, range 23.8–84.7) Other 10 (17)
and 162 male patients (mean age 66.5 years, range 21.7–97).
Patients with more than one malignancy tended to be older.
Details of the primary malignancy are summarized in
Table 2, and Table 3 outlines details of the new malignancy Discussion
where this was detected. The most common site of the new
malignancy was an unknown primary (11 cases) followed by The presence of suspicious bone lesions in patients with a
myeloma (9 cases) (Table 3). Diagnostic details for benign history of malignancy is often assumed to be metastatic in
lesions are summarized in Table 4. origin secondary to this index malignancy [7–9]. However,
Overall, in 250 cases (66%) the original malignancy was our results suggest that around a third of these lesions will
confirmed on the bone biopsy (Fig. 1). In 128 cases (34%) an have an alternative diagnosis (18% benign and 16% new ma-
alternative diagnosis was confirmed; of these 69 had a benign lignancy). The prevalence of an alternative diagnosis could
diagnosis (18%) and 59 had a new malignancy (16%.) have a clinical impact on the management of these patients,
Examples of cases with an alternative benign and malignancy with our survival curves confirming significant differences in
diagnosis are shown in Figs. 2 and 3, respectively. survival if the lesion is benign or malignant.
For the secondary outcomes, the survival was significantly The challenge of managing MBD will increase in future for
better for those in whom the bone biopsy confirmed a benign several reasons; prolonged survival of cancer patients will
diagnosis (logrank test, p = 0.05, Fig. 4). There was no signif- increase the number of patients with MBD [5]. It has been
icant difference in survival between patients diagnosed with estimated that pathological fractures occur in 10–30% of these
the original malignancy and for those in whom a new malig- patients [17] thereby increasing the burden on trauma and
nancy was confirmed (logrank test, p = 0.4). oncological surgeons in particular, as these cases are often
Overall, a second bone biopsy was needed in 11 pa- complex and require extensive planning and MDT input.
tients (3%), which represents a diagnostic yield of 97%. Furthermore, increased utilization of advanced imaging tech-
In these patients, a benign diagnosis was confirmed on niques in clinical practice such as MRI, CT or PET will further
subsequent biopsy in 6 cases and recurrence of the known increase detection of incidental suspicious bone lesions in pa-
primary in 5 cases. tients with a previous cancer [3]. The combination of in-
creased prevalence and detection of MBD will provide con-
siderable challenges to orthopaedic surgeons working in both
Table 2 Details of initial
malignancy Diagnosis Frequency (%)

Breast 129 (34) Table 4 Details of benign diagnosis

Prostate 51 (13)
Diagnosis Frequency (%)
Lymphoproliferative 27 (7)
Colorectal 19 (5) Non-specific 38 (55)
Renal Cell Ca 15 (4) Chondroid lesion 7(10)
Lung 12 (3) Bone island 2(3)
Bladder 11 (3) Haemangioma 2(3)
Chondrosarcoma 10 (3) Cystic 2(3)
Endometrial 9 (2) Sarcoid 2(3)
Melanoma 8 (2) Reactive/sclerotic 2(3)
Osteosarcoma 8 (2) Periprosthetic tissue/wear debris 2(3)
Other 69 (18) Other 12 (17)
1114 Skeletal Radiol (2021) 50:1111–1116

Fig. 1 Case with a history of breast Ca and L2 lesion. Images from a 48- further axial CT slice is shown in panel c. A CT-guided needle biopsy
year-old female with a history of breast Ca. Panels a and b are T1- and (shown in panel d) confirmed that this was metastatic breast Ca
T2-weighted sagittal MRI images, respectively, showing a lesion in L2. A

elective and trauma settings. Another trend we have noted in alternative diagnosis from bone biopsies (Table 1) ranging
our unit is the increasing demand for bone biopsies to assess from 2 [12] to 34% [14]. Several theories may account for
the receptor status of bone metastases. Discordance in ER, this: primarily quite a few studies are very small (≤ 50
PgR and HER2 status between primary breast tumours and patients) [12–14] and are therefore likely to be underpow-
metastatic disease is well recognized [18]. Indeed, the highest ered to detect rare events. Furthermore, the published stud-
discordance rate for ER status in metastatic breast cancer is ies come from different populations which may have dif-
bone metastases [18]; knowledge of receptor status is critical fering background rates of cancer and have been conducted
when deciding on the use of hormone-based and targeted at different points in time during which the survival for
treatments in these patients. many cancers has improved. Moreover, image-guided bi-
The prevalence of metastatic disease in patients with a opsy is reported to have variable diagnostic yields [18].
previous malignancy and suspicious bone lesions is un- Interestingly, the largest study to date has a confirmed rate
clear [13]; one recent study suggests that more than one of alternative diagnosis of 27.2% [8], which is broadly
fifth of cases were unrelated to the primary tumour [3]. Our similar to our rate. Crucially, the authors acknowledge that
study suggests that the actual figure is nearer a third of their CT-guided bone biopsy technique yields non-
cases, with our series reporting 16% of cases having a diagnostic results in 30% of cases [19] and conclude that
new malignancy. It is worth noting that the literature on their results likely underestimate the prevalence of both
this area has demonstrated marked variation in rates of benign and malignant bone lesions [8].

Fig. 2 Case with a history of melanoma and medial tibial lesion. Images confirmed the lesion, and a CT-guided needle biopsy (shown in panel c)
from a 40-year-old female with a history of melanoma and a medial tibial demonstrated that this was actually a Giant cell tumour (GCT)
plateau lesion initially seen on plain radiograph in panel a. An MRI (b)
Skeletal Radiol (2021) 50:1111–1116 1115

Fig. 3 Case with multiple previous malignancies and proximal humeral scan in panel b demonstrated that this was a solitary lesion, and CT in
lesion. Images from an 81-year-old male with a history of bladder Ca, panel c showed that the lesion was lytic with loss of the anterior cortex.
prostate Ca and hepatocellular Ca presenting with a proximal humeral The biopsy showed a chondrosarcoma, and excision surgery was per-
lesion and a humeral IM nail in situ from previous trauma (a). The bone formed with an endoprosthesis inserted as shown in panel d

Strengths specific region within the lesion most likely to be diagnostic.

Biopsies are undertaken using CT guidance on a regular GA
This study has several key strengths; notably it is the largest list by radiologist experienced in this technique.
European study of this type (Table 1), and our unit is a very Furthermore, our subgroup analysis confirms a significant
well-established bone tumour unit with extensive experience difference in prognosis depending on the diagnosis, thus fur-
collecting and analysing bone biopsies. Indeed, our diagnostic ther underlining the need to consider a biopsy to confirm the
yield from biopsies was very good with only 11 cases (3%) diagnosis in these cases.
requiring multiple biopsies. This compares well with other
units previously mentioned where the diagnostic yield of ini-
tial biopsies was around 30% [19]. Indeed, one recent US Limitations
study reviewing over 900 CT-guided musculoskeletal biop-
sies showed a non-diagnostic rate of 15% on initial biopsy There are several limitations to this study which should be
[20]. Reasons for our high diagnostic yield are multifactorial; considered when interpreting the results. As our unit is based
planned biopsy cases are discussed in an MDT allowing dis- in an orthopaedic hospital, we had limited access to original
cussion regarding further imaging with MSK pathologists and notes detailing the original diagnosis and management of the
radiologists. This allows collective decision-making regarding malignancy. For that reason, it was not always possible to
patient selection and further imaging when necessary in plan- deduce an accurate timescale between original cancer and
ning an intervention. This enables identification of the optimal subsequent bone biopsies.
lesion to biopsy (if more than one is present) and to target the Furthermore, being based in a tertiary referral centre could
introduce selection bias in the case mix of patients undergoing
bone biopsy, which has been highlighted in similar studies [3,
8, 13]. We have been unable to account for patients who have
not been referred to us or are not well enough to undergo
biopsy. It is feasible, the pattern of results in these patients
may differ slightly from our sample and may skew our sur-
vival results.
We have also not considered the impact of solitary versus
multiple bone lesions in this study. Recent authors have noted
that this was not a significant factor influencing correlation
between bone lesions and primary malignancy [3]. Other au-
thors have emphasized the need to biopsy single lesions in
patients with a history of cancer [7], and this is a viewpoint
with which we strongly agree. Interestingly, some previous
authors have gone as far as to recommend that bone biopsy
Fig. 4 Kaplan-Meier survival curves comparing survival according to should be mandatory in cases of uncertainty and lamented
bone biopsy diagnosis guidelines for not being forceful enough on this issue [21].
1116 Skeletal Radiol (2021) 50:1111–1116

Conclusion 8. Raphael B, Hwang S, Lefkowitz RA, Landa J, Sohn M, Panicek

DM. Biopsy of suspicious bone lesions in patients with a single
known malignancy: prevalence of a second malignancy. AJR Am
Overall, this study shows that for patients with a history of J Roentgenol. 2013;201:1309–14.
malignancy and suspicious bone lesions, in a third of cases the 9. Toomayan GA, Major NM. Utility of CT-guided biopsy of suspi-
bone biopsy will confirm another diagnosis i.e. benign or a cious skeletal lesions in patients with known primary malignancies.
new malignancy. This has considerable implications in the AJR Am J Roentgenol. 2011;196(2):416–23.
10. Mirels H. Metastatic disease in long bones. A proposed scoring
management of these cases, and our study confirms that the system for diagnosing impending pathologic fractures. Clin
survival of these patients varies significantly depending on Orthop Relat Res. 1989;249:256–64.
whether this diagnosis is benign or malignant. We feel that 11. Monfardini L, Preda L, Aurilio G. CT-guided bone biopsy in cancer
in these cases discussion with an appropriate bone tumour unit patients with suspected bone metastases: retrospective review of
is essential to consider the merits of biopsy. 308 procedures. Radiol Med. 2014;119(11):852–60.
12. Cronin CG, Cashell T, Mhuircheataigh JN, et al. Bone biopsy of
new suspicious bone lesions in patients with primary carcinoma:
Compliance with ethical standards prevalence and probability of an alternative diagnosis. AJR Am J
Roentgenol. 2009;193(5):W407–10.
Conflict of interest The authors declare that they have no conflict of 13. Clayer M, Duncan W. Importance of new bone lesions in patients
interest. with previous carcinoma. Clin Orthop Rel Res. 2006;451:208–11.
14. Aoki J, Koyama Y, Morita H, et al. Clinical background and its
Ethical approval All procedures performed in studies involving human relation to results of percutaneous needle biopsy of suspected bone
participants were in accordance with the ethical standards of the institu- metastasis under guidance with CT fluoroscopy. Radiat Med.
tional and/or national research committee and with the 1964 Helsinki 2005;23(7):463–7.
declaration and its later amendments or comparable ethical standards. 15. Lindstrom LS, Karlsson E, Wilking UM, et al. Clinically used
breast markers such as estrogen receptor, progesterone receptor,
and human epidermal growth factor receptor 2 are unstable
throughout tumour progression. J Clin Oncol. 2012;30:2601–8.
References 16. Didolkar MM, Andersson ME, Hochman MG, et al. Image guided
core needle biopsy of musculoskeletal lesions: are nondiagnostic
1. Storm HH, Engholm G, Hakulinen T, et al. Survival of patients results clinically useful? Clin Orthop Relat Res. 2013;471:3601–9.
diagnosed with cancer in the Nordic countries up to 1999–2003 17. Macedo F, Ladeira K, Pinho F, et al. Bone metastases: an overview.
followed to the end of 2006. A critical overview of the results. Oncol Rev. 2017;11:321.
Acta Oncol. 2010;49(5):532–44. 18. Yeung C, Hilton J, Clemons M, et al. Estrogen, progesterone and
2. De Moor JS, Mariotto AB, Parry C, et al. Cancer survivors in the HER2/neu receptor discordance between primary and metastatic
United States: prevalence across the survivorship trajectory and breast tumours- a review. Cancer Metastasis Rev. 2016;35:427–37.
implications for care. Cancer Epidemiol Biomark Prev. 19. Hwang S, Lefkowitz RA, Landa J, et al. Percutaneous CT-guided
2013;22(4):561–70. bone biopsy: diagnosis of malignancy in lesions with initially inde-
3. Zhang L, Wang Y, Gu Y, Yingyong H, Zixian C. The need for bone terminate biopsy results and CT features associated with diagnostic
biopsies in the diagnosis of new bone lesions in patients with a or indeterminate results. AJR Am J Roentgenol. 2011;197:1417–
known primary malignancy: a comparative review of 117 biopsy 25.
cases. J Bone Oncol. 2019;14:100213.
20. Chang CY, Huang AJ, Bredella MA, et al. Percutaneous CT-guided
4. Coleman RE. Clinical features of metastatic bone disease and risk
needle biopsies of musculoskeletal tumors: a 5-year review analysis
of skeletal morbidity. Clin Cancer Res. 2006;12:6243s–9s.
of non-diagnostic biopsies. Skelet Radiol. 2015;44:1795–803.
5. Harvie P, Whitwell D. Metastatic bone disease. Have we improved
after a decade of guidelines? Bone Joint Res. 2013;2(6):96–101. 21. Schiavon G, Tait DM, Briggs TWR, Smith IE. A solitary bone
6. British Orthopaedic Oncology Society and British Orthopaedic lesion in a patient with previous breast Cancer and the importance
Association. Metastatic Bone disease: A guide to Good Practice. of bone biopsy: a case report and a review of the guidelines. Clin
2015 Revision. Breast Cancer. 2013;13:77–80.
7. Patton JT, Sommerville SMM, Grimer RJ. Primary malignant tu-
mours of bone following previous malignancy. Sarcoma. Publisher’s note Springer Nature remains neutral with regard to jurisdic-
2008;418697. tional claims in published maps and institutional affiliations.