Pheochromocytomas (PHEOs) and paragangliomas (PGLs) (collectively, PPGLs), are rare neuroendocrine tumors (NETs) derived from chromaffin cells of the adrenal medulla (PHEOs) and extra-adrenal paraganglia (paragangliomas) (1, 2). All PPGLs are considered to possess metastatic potential, with metastases occurring in approximately 10-15% of PHEOs and 15-40% of PGLs (3, 4).

Bone metastases (BM) are frequently identified in patients with solid malignancies, with lung, prostate, and breast cancers being the most common sources of skeletal involvement (5). In metastatic pheochromocytomas and paragangliomas (mPPGL), bone is also one of the principal sites of metastasis, with reported prevalence varying across studies (6, 7).

BM in patients with PPGLs are a significant clinical concern, leading to Skeletal-related Events (SREs) that diminish both quality of life (QoL) and survival. Complications include pain, fractures, neurological issues like spinal cord compression, rare hypercalcemia, and interventions such as surgery or radiation therapy (6). These patients also endure reduced mobility, loss of autonomy, impaired QoL, and lower overall survival (OS) rates (7), highlighting the impact of BM on patient outcomes. Early diagnosis and timely intervention are crucial to reducing the risk of morbidity and complications associated with BM (7).

Data derived primarily from case reports, retrospective studies, and, less frequently, clinical trials have demonstrated the efficacy of various treatments for managing mPPGL. However, treatment efficacy at specific anatomical sites, such as bone, has not been extensively addressed due to the rarity and heterogeneity of these tumors (6).

BM represent thereby a significant clinical challenge in patients with PPGL. Taking all the above into consideration, this narrative review aims to provide a comprehensive overview of BM in PPGL. It outlines the general characteristics, clinical presentation, and complications of these lesions, and discusses current diagnostic and therapeutic approaches. Key challenges in management and areas requiring further research are also highlighted.

This narrative review was based on a comprehensive literature search conducted in PubMed/MEDLINE and Embase databases, with Web of Science additionally consulted to cross-check and confirm the bibliographic completeness of the retrieved references. The search covered publications from database inception through July 15, 2025, and the last search was performed on July 15, 2025, prior to manuscript submission.

Only articles published in English were included, encompassing original studies, case series, reviews, and clinical guidelines. Additional references were identified through manual screening of the bibliographies of retrieved articles and relevant reviews, to ensure inclusion of studies potentially missed by the electronic search.

A summary of the detailed search strings for each database, the Boolean term combinations, and the approximate number of records screened is provided in Supplementary Appendix 1, Table 1 .

Given the narrative design of this review, no formal systematic review process or quantitative synthesis was performed.

Population-based studies confirm that PPGL are rare, with incidence rates ranging from 1.9 to 6.6 cases per million person-years across regions. Reported estimates include 3.7–5.7 per million in the Netherlands (8), 6.6 per million in Canada (9), and up to 6.6 per million in Denmark (10), while a global meta-analysis reported 1.9 per million person-years (11). Regarding prevalence, a recent meta-analysis including only patients with PHEO reported a global pooled estimate of 19.8 cases per 1,000,000 individuals (≈2 per 100,000; 95% CI: 9.6–40.8) (11). In contrast, a Danish nationwide study encompassing both PHEO and PGL found a point prevalence of 64.4 per million inhabitants as of 2015 (10). Importantly, PPGLs display a strong genetic background, with germline mutations identified in approximately 26–41% of patients across contemporary mixed PPGL cohorts (12–15), particularly in younger patients (16, 17), those with bilateral or multifocal tumors (18, 19), and in extra-adrenal disease (17, 19, 20). The most frequently involved genes include Succinate Dehydrogenase Complex Subunit B (SDHB), Succinate Dehydrogenase Complex Subunit D (SDHD), Von Hippel-Lindau (VHL), RET, Succinate Dehydrogenase Complex Subunit A (SDHA), Succinate Dehydrogenase Complex Subunit C (SDHC), Succinate Dehydrogenase Complex Subunit AF 2 (SDHAF2), Neurofibromatosis type 1 (NF1), TMEM127, and MAX also contributing but at lower frequence (12, 15, 21, 22). Germline SDHB mutations confer the highest metastatic risk in PPGL, with metastatic rates ranging from 12–41% across cohorts (21) and an adjusted odds ratio of 5.68 versus non-SDHB genotypes (23). Cohort data confirm a ~29% metastatic rate among SDHB carriers (24). In a Saudi cohort, 28.6% of SDHB carriers developed metastases, markedly exceeding rates observed in other genes (22), further confirming SDHB as the predominant driver of metastatic behavior in PPGL (25). Metastasic risk varies among non-SDHB genes: SDHD ≈4%, SDHA ≈16%, SDHC ≈23%, and SDHAF2 none (21); VHL and MEN2A show moderate risk (26),NF1, TNEM127, MAX, FH and EPAS1 rarely metastatic (27). These data highlight the importance of genetic testing for all PPGL patients, as specific mutations, particularly in SDHB, are closely linked to metastatic potential, including bone dissemination (21, 28–33).

Bone is one of the primary sites of metastasis in patients with PPGLs (7, 34–36), and in up to 20% of cases, the skeleton represents the sole site of metastatic spread (7). mPPGLs are among the solid tumors with a particularly high tropism for bone, with involvement rates comparable to those observed in other osteotropic malignancies such as breast cancer (6, 7, 37). However, data from reported cohorts indicate that the prevalence of BM is highly variable and influenced by differences in study populations. The epidemiological findings across studies are summarized in Figure 1 and in Table S2 (Supplementary Appendix) . Direct evidence shows that germline SDHB and SDHA mutations confer predisposition to BM in PPGL. SDHB mutations involve bone in up to 80% of metastatic cases (38), while SDHA mutations also show marked bone tropism when metastases occur (14, 39).

Further insights into the epidemiology of BM in PPGLs have been provided by retrospective institutional series. At MD Anderson Cancer Center, a study involving 128 patients reported that the median time from primary tumor diagnosis to the development of BM was 3.4 years (7). In the same study, the presence of bone-limited disease was associated with a significantly longer median OS (12 years) compared to patients with both osseous and non-osseous metastatic involvement (5 years) (7).

A retrospective review from the National Hospital Organization Kyoto Medical Center, including 40 patients with mPPGL, identified a significant association between extraskeletal extension of spinal lesions and the occurrence of SREs (37). In this same study, the median time from the diagnosis of BM and the onset of the first SRE was 11.9 months (37).

In a larger cohort of 294 patients with bone-metastatic PPGL, 31% developed SREs, with a median time of 4.4 months (range: 0–246.6) after BM diagnosis. Among these patients, 22% underwent surgical treatment and 46% received radiotherapy (36).

Additionally, a multi-institutional study that included 100 patients with PHEO or sympathetic PGL among a broader NET population reported BM in 25% of cases (40). However, it is important to note that this estimate was not limited to individuals with confirmed metastatic disease, and therefore may not be directly comparable to studies focused exclusively on mPPGL cohorts.

Regarding the primary anatomical locations of PPGLs that give rise to BM it has been reported that these lesions can originate from both adrenal and extra-adrenal sites. In a recent large multicenter cohort of 294 patients with PPGL-related BM (36), the most common primary tumor sites were adrenal PHEOs (41%) and abdominal or retroperitoneal PGLs (30%), followed by head and neck PGLs (14%), with pelvic and thoracic locations being less frequent (5% each). Interestingly, BM have also been reported to arise from uncommon sites, such as PGLs located in the sellar region (41). These findings highlight the anatomical diversity of PPGLs associated with skeletal dissemination.

Due to their rarity, the pathophysiology of PPGL-related BM remains poorly understood; however, it is likely to reflect mechanisms observed in other malignancies with skeletal involvement. Bone is considered a preferential site for metastasis due to the extensive vascularization of the bone marrow. Tumor cells can express adhesion molecules that facilitate their attachment to marrow stromal cells, and the subsequent release of growth factors supports tumor cell proliferation and expansion within the bone microenvironment (42).

In the specific context of PPGL, it has been hypothesized that tumor cells strongly express the chemokine receptors CXCR4 and CCR9, while bone marrow stromal cells and osteoblasts secrete their corresponding ligands, CXCL12 and CCL25, respectively. This ligand-receptor interaction may facilitate the recruitment and homing of mPPGL cells to bone tissue (7, 42, 43).

Once established in the bone, tumor-induced skeletal damage may result from both catecholamine-dependent and catecholamine-independent mechanisms. Among the latter, tumor cells may disrupt the physiological balance between bone resorption and formation by promoting osteoclast activation and bone matrix degradation. This results in the development of lytic or mixed lesions, potentially exacerbated by the secretion of tumor-derived mediators that alter the bone microenvironment and favor metastatic progression (44).

Regarding catecholamine-mediated mechanisms, these hormones are known to play a significant role in bone metabolism by inducing imbalanced remodeling through activation of β-adrenergic receptors (2). In patients with PPGL, catecholamine excess may lead to increased bone turnover, with bone resorption surpassing formation (45). One study suggests that catecholamine excess and sympathetic overstimulation in PHEO can impair trabecular bone microarchitecture, reduce bone mass, and increase bone resorption, thus positioning PHEO as a potential secondary cause of osteoporosis (46). Indirect evidence further supports this hypothesis, as studies have shown that individuals treated with beta-blockers (a class of drugs that inhibit β-adrenergic signaling) exhibit a lower risk of fractures and improved bone microarchitecture compared to non-users (47, 48). Additional insights from neuroendocrine neoplasms in general suggest that bone metastases arise from a “vicious cycle,” in which tumor cells stimulate osteoclast activation primarily through RANKL overexpression and suppression of osteoprotegerin, leading to excessive bone resorption, release of growth factors from the bone matrix, and further tumor proliferation within a hypoxic, acidic, and calcium-rich microenvironment (49).

BM in PPGLs, although variably prevalent, can result in significant clinical manifestations and complications that substantially impair patients’ QoL (7, 34–36). BM most commonly affect the vertebrae, spine, pelvis, long bones, ribs, and skull (7, 50), and are predominantly osteolytic (92%) and multiple (88%) in nature (37). These lesions compromise skeletal integrity and predispose patients to a range of debilitating complications, including severe pain, pathological fractures, spinal cord compression, hypercalcemia, and other SREs, all of which negatively impact QoL and functional status (7, 36, 42, 51). Notably, it has been reported that in 31% of patients with SREs, BM was the initial clinical manifestation that led to the diagnosis of malignant PPGL, underscoring that in a substantial proportion of cases, skeletal involvement can precede recognition of the underlying tumor (7).

PPGLs may also contribute to secondary osteoporosis, potentially increasing the risk of fragility fractures, particularly of the vertebrae (52). Notably, extraskeletal invasion of the spine has been identified as a significant risk factor for SREs, such as vertebral fractures and spinal cord compression (37). In one study, extraskeletal spinal invasion was significantly associated with the development of SREs (p = 0.001), suggesting that this radiological finding may serve as a valuable clinical risk marker (37).

Skeletal manifestations in this population can be severe and may require urgent multidisciplinary evaluation by neurosurgical and/or orthopedic teams, especially in cases involving spinal instability or neurological compromise (7).

Evaluation of bone health (including bone mineral density (BMD) and trabecular bone score [TBS]) is important in patients with PPGL, as both parameters have been shown to improve following surgical resection of the tumor (2, 52). Additionally, mild bone marrow suppression is a frequent finding in patients with BM from malignant PPGLs (53).

The clinical outcomes of skeletal involvement in PPGL have been described in several cohorts, whose findings are summarized in Table 1 to illustrate the frequency and nature of bone-related complications (7, 36, 37). SREs were common but showed variability across studies. In the study by Ayala-Ramirez et al. (7), 72% of evaluable patients experienced at least one true SRE, with severe pain, pathological fractures, and spinal cord compression occurring in 33%, 27%, and 25% of cases, respectively. In contrast, Laganà et al. (36) reported a lower overall frequency of SREs (31%) in a larger cohort (n = 294), with bone pain (49%), pathological fractures (19%), and spinal cord compression (16%) being the most common manifestations. Hypercalcemia was rare in both studies. Similar outcomes were observed in a Japanese cohort of 24 patients with BM, where 50% developed at least one SRE (Yokomoto-Umakoshi et al., 2018 (37)). In this group, the most frequent SRE was radiation to bone (42%), followed by spinal cord compression (21%), severe bone pain (33%), pathological fracture (8%), and bone surgery (8%). No cases of hypercalcemia were reported, and the median time to the first SRE was 11.9 months. These findings, though derived from a smaller cohort, are consistent with the heterogeneity observed in the larger studies and highlight the need for prospective evaluation of skeletal complications in mPPGL.

A multi-institutional study that included 100 patients with PHEO or sympathetic PGL within a broader NET cohort reported that, in the PPGL subgroup, 20% developed spinal cord compression—consistent with previous reports. Additionally, pathological fractures and hypercalcemia were observed in 8% and 12% of PPGL patients with BM, respectively (40).

Metastatic manifestations of PPGLs can be classified as synchronous, occurring concurrently with the initial diagnosis, a scenario less frequently observed but associated with poorer prognosis (36, 54). Alternatively, they may present as metachronous metastases, which develop subsequent to the primary diagnosis and represent the more prevalent pattern (36, 55). In the study by Ayala-Ramirez et al., the median time to the first SRE was 4.4 months. This time frame is comparable to that observed in other cancers that often metastasize to the bone, such as breast cancer (56). In the MD Anderson Cancer Center study, 38% of patients presented with synchronous BM, while 63% developed metachronous BM, with a median onset of 3.4 years after the diagnosis of the primary tumor (7). Similarly, in the cohort reported by Laganà et al. (2024), 33% of patients had synchronous and 67% metachronous BM, with a longer median latency of 5.7 years (range, 0–48 years), underscoring the need for long-term surveillance in patients with PPGL (36).

Metachronous cases illustrate that skeletal metastases can occur even decades after initial diagnosis (57, 58), with intervals of up to 46 years (59) and even 52 years after the initial diagnosis in a reported case of PGL (51). This underscores the need for long-term follow-up in patients with PPGLs. The mechanisms driving the prolonged dormancy of certain metastatic PGL remain unclear. Bone-residing metastatic cells can stay quiescent for extended periods, ranging from months to decades. Reactivation of these dormant cells is hypothesized to result from alterations in chromatin structure and shifts in the tumor microenvironment (60).

In the study from MD Anderson Cancer Center, SREs often occurred after BM diagnosis and, in some cases, were the first sign of malignancy (7). See Figure 2 which graphically shows the SREs in patients with mPPGL.

Risk factors for SREs in patients with mPPGL have been evaluated in both single-center and multicenter cohorts, revealing complementary insights. In a smaller Japanese series, Yokomoto-Umakoshi et al. (37) identified extraskeletal invasion of the spine as a strong predictor of SREs, with affected patients showing a significantly higher likelihood of developing one or more SREs compared to those without this radiological feature (73% vs. 0%, p = 0.001). This anatomical marker of local aggressiveness also correlated with the occurrence of multiple SREs during follow-up (75% vs. 7%, p = 0.002), while no associations were found with clinical variables such as age, tumor size, functional type, or extent of metastases. In contrast, the large multicenter study by Laganà et al. (36) identified the presence of liver metastases—a marker of systemic disease aggressiveness—as an independent risk factor for SREs in multivariable analysis (p=0.046), reinforcing the link between disseminated tumor burden and skeletal complications. Notably, treatment with ¹³¹I-Meta-iodobenzylguanidine (¹³¹I-MIBG) radionuclide therapy was associated with a reduced risk of SREs in both studies (significantly so in Laganà et al.), suggesting a potential protective effect. While Yokomoto et al. emphasized the importance of local skeletal invasion, Laganà et al. highlighted the prognostic impact of systemic dissemination. Together, these findings underscore the multifactorial nature of SRE risk in mPPGL, involving both local anatomical features and overall disease burden, and support a tailored approach to surveillance and bone-targeted therapy.

QoL in patients with mPPGL has not been systematically evaluated using standardized instruments. Existing studies have assessed QoL in PGL patients more broadly, without focusing specifically on those with BM. In a cohort of 174 patients with PGLs, van Hulsteijn et al. employed three validated questionnaires (HADS, MFI-20, and SF-36) and found that individuals with mPPGL reported significantly greater mental fatigue and lower general health perception compared to those with benign disease (61). Furthermore, patients with PGLs (regardless of metastatic status) report worse QoL compared to the general population, with increased fatigue, poorer physical functioning, and psychological and social impairments. These effects are more pronounced in individuals experiencing disease-related symptoms (61).

Although formal QoL assessments are lacking in studies focused exclusively on bone-metastatic PPGL, the available evidence strongly suggests a considerable impact. Retrospective studies have consistently highlighted the physical and functional burden associated with SREs, including complications such as bone pain, pathological fractures, and spinal cord compression (7, 36, 37). Metastatic bone lesions are particularly detrimental to QoL, as they can severely impair patients’ ability to carry out daily activities due to the significant pain and risk of fractures they cause (62). These events frequently result in reduced mobility, loss of independence, hospitalizations, and the need for intensive interventions, factors that are widely recognized as detrimental to QoL.

Further supporting this, von Moos et al., in a comprehensive review of metastatic bone pain across various advanced cancers (not limited to PPGL) (63), emphasized that skeletal complications significantly compromise QoL by contributing to persistent pain, impaired physical functioning, sleep disturbance, and emotional distress. The authors advocate for early and individualized management—including analgesia, radiotherapy, and bone-targeted agents—to mitigate the cumulative burden of BM on patient well-being (63). In addition, recent reviews in neuroendocrine tumors have highlighted that validated QoL instruments, such as the EORTC QLQ-C30, QLQ-GI.NET21, and Norfolk QoL-NET, include items specifically addressing bone or musculoskeletal pain, which represent a key determinant of reduced QoL in this population (64). While these data are derived from broader NET cohorts, they reinforce the relevance of systematically assessing bone pain and related symptoms in patients with mPPGL and BM.

Multidisciplinary management and the use of targeted therapies, such as bone resorption inhibitors and therapeutic radionuclides, can enhance symptom control and, in some cases, improve QoL in patients with BM. However, current evidence suggests that symptomatic relief does not always translate into prolonged survival, highlighting the need for individualized therapeutic goals based on clinical context and patient preferences (65). In this context, supportive care focused on pain control and the prevention of SREs is essential to preserving QoL in these patients (66).

In conclusion, the available evidence highlights the significant impact of BM on QoL in patients with PPGL. These findings underscore the importance of early supportive care, multidisciplinary management, and timely use of systemic and bone-targeted therapies to mitigate morbidity in this population.

The prognosis of patients with PPGL and BM is influenced by a range of clinical, histopathological, biochemical, and molecular features. However, most available data derive from studies on mPPGL in general, with limited evidence specifically addressing bone involvement. This section summarizes current knowledge regarding prognostic factors, survival differences related SREs, and the impact of selected treatments.

The evaluation of BM in PPGL requires a multimodal strategy that integrates anatomical and functional imaging, both for diagnosis and for assessing treatment response, which is commonly evaluated using RECIST criteria (4, 32, 70). However, no validated criteria currently exist for assessing skeletal response specifically in PPGL. The following section focuses on the role of imaging in the diagnosis and evaluation of BM in patients with mPPGL. In this section, we use the following abbreviations: computed tomography (CT), magnetic resonance imaging (MRI), and positron emission tomography/computed tomography (PET/CT).

Currently, there are no randomized controlled trials or open-label studies to establish evidence-based guidelines for the management of metastatic bone disease in mPPGL. Treatment strategies should be individualized for each patient, and whenever possible, decisions should be guided by the recommendations of a multidisciplinary endocrine tumor board (6).

Therapeutic approaches can be broadly categorized into three main groups: (1) general measures and supportive care, (2) management of localized bone disease, and (3) systemic treatment strategies. An overview of these treatment domains is illustrated in Figure 3 .

In summary, BM are common in patients with mPPGL and are associated with a high incidence of SREs, underscoring the importance of long-term monitoring and proactive management. Early identification and appropriate treatment of bone involvement are essential to improve both QoL and OS. The potential for late-onset metastatic spread, even years after the initial diagnosis, should always be considered. An integrated, multidisciplinary approach (encompassing endocrinology, oncology, nuclear medicine, palliative care, radiotherapy, orthopedic surgery, and neurosurgery) is fundamental to providing optimal care for these complex patients. To aid clinical decision-making, a proposed algorithm outlining challenges and possible solutions is presented in Figure 6 .