Incidence, Risk Factors, and Survival of Bone Metastases and Skeletal-Related Events in Melanoma Patients: A Systematic Review and Quality Assessment of 29 Studies

Graphical abstract

• A considerable amount of melanoma patients suffers from bone metastases.• Bone metastases increase morbidity and mortality through skeletal-related events.
• Bone-directed agents lower skeletalrelated event risk in melanoma bone metastases.• There are limited studies on risk factors associated with skeletal-related events.• Bone metastasis survival is linked to clinical, tumor, and treatment factors.Background: Skeletal metastases make up 17% of all metastases from advanced-stage melanoma.Bone metastases are associated with increased morbidity and mortality and decreased quality of life due to their association with skeletal-related events (SREs), including pathological fracture, spinal cord compression, hypercalcemia, radiotherapy, and surgery.The study aimed to determine the incidence of bone metastases and SREs in melanoma, identify possible risk factors for the development of bone metastases and SREs, and investigate survival rates in this patient population.Methods: A computer-based literature search was conducted using Pubmed, Embase, and Cochrane Central Register of Controlled Trials up to July 2023.The Newcastle-Ottawa Quality Assessment Scale (NOS) was utilized for quality assessment.Study characteristics, patient information, risk factors for developing bone metastases and SREs, and characteristics for survival were recorded.

Introduction
Melanoma was the fifth most common cancer diagnosis in the United States in 2021, accounting for 5.6 % of all cancer diagnoses and an estimated 106,000 new cases annually [1].Melanoma's aggressive progression and development into bone metastasis makes the disease accountable for most skin cancer-related deaths [2].It is estimated that three to seven percent of patients with melanoma develop bone metastasis, representing up to 17 % of the metastases from metastatic melanoma [3][4][5].In recent years, advancements in melanoma diagnosis and treatment, such as multi-modal therapies including novel immunotherapies and targeted drugs, as well as enhanced imaging methods, have improved survival but have simultaneously led to a rising incidence of bone metastases [6].
Bone metastases are associated with increased morbidity and mortality and decreased quality of life due to its association with skeletalrelated events (SREs), which include pathological fracture, spinal cord compression, hypercalcemia, radiotherapy, and surgery [7].The high prevalence of melanoma makes the burden of bone metastasis and subsequent SREs considerable.A recent retrospective study reported that melanoma patients with bone metastasis suffered an SREs incidence rate of 66 %, of which radiotherapy was the most common SRE, with an incidence rate of 39 % [4].Despite the high prevalence of bone metastases and SREs in patients with melanoma, information on the overall incidence, risk factors, and prognosis is limited.A better understanding of the disease burden of bone metastases and SREs in patients with melanoma may guide healthcare management, improve the quality of care, and help clinicians and patients in management decisions to minimize the detrimental effects of skeletal metastases and SREs.
To our knowledge, only a handful of studies specifically report on bone metastases and SREs of patients with melanoma, and no systematic reviews have been conducted on this topic.Therefore, this systematic review aims to report the incidence, risk factors of developing bone metastases, and subsequent SREs in patients with melanoma.Our study offers the first overview of the scale of both bone metastases and SREs that can be used to guide the management of bone metastases in patients with metastatic melanoma to achieve the best possible outcome.

Protocol and registry
This systematic review was performed according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [8].We registered the protocol online in PROSPERO with CRD42023410881.

Search strategy
Three independent reviewers (M.R.S., O.N.L., D.T.) performed a systematic literature search using PubMed, EMBASE (OvidSP), and Cochrane Central Register of Controlled Trials for studies published from their inception to July 23, 2023.Boolean searches were performed using Medical Subject Headings (MeSH) for PubMed and Cochrane, Emtree for EMBASE, or keywords and search terms combining "melanoma" and "bone metastases" (Appendix A).

Eligibility criteria
All titles and abstracts from the literature search were exported into Mendeley and uploaded to Rayyan [9].A series of consecutive stages were performed for study selection, including duplicate identification, screening of titles and abstracts, and full article review in accordance with the eligibility criteria by the same three independent reviewers.All human studies in English related to melanoma, bone metastases (including spinal metastases), and SREs were included.Studies were excluded if they (1) were case series, letters, comments, editorials, conferences, guidelines, abstracts, supplements, technique papers, systematic reviews, or regular review articles; (2) included fewer than ten patients with bone metastases from melanoma; (3) focused on metastatic uveal melanoma or sinonasal mucosal melanoma; (4) did not prioritize melanoma as the primary cancer, focused on visceral metastases or oligometastatic disease, or prioritized recurrent disease; or (5) were clinical trials focusing on metastatic melanoma treatment unless median survival was explicitly mentioned.Metastatic uveal and sinonasal mucosal melanoma patients were excluded given their distinct biological and clinical features compared to cutaneous melanoma that contribute to significantly different prognosis and overall survival as demonstrated in previous studies [10,11].All included studies were manually referenced and reviewed for any missed studies.Disagreements or uncertainties were resolved through discussion with an experienced fourth reviewer (O.Q.G.; Fig. 1).

Quality assessment
The quality of the included studies was independently assessed by two reviewers (M.R.S., O.N.L.) using the Newcastle-Ottawa Quality Assessment Scale (NOS) [12].The NOS was developed to determine the quality of cohort-and case-control studies using a "star system" in which a study is evaluated on selection, comparability, and exposure/outcome of interest of the study group.Study ratings ranged from zero to nine: studies with more than six stars were deemed to have a "low risk of bias", those with three to five stars an "unclear risk of bias", and those with less than two stars a "high risk of bias" [12].Any disparities were discussed with a third reviewer (O.Q.G.) present to reach a final consensus.

Data extraction
The following data were extracted from each study independently by two investigators (M.R.S. and O.N.L.) -(1) study characteristics: first author, year of publication, study country, study type, and study period; (2) patient information: demographics, tumor characteristics, and bone metastases; (3) statistical analysis of risk factors for developing melanoma bone metastases and SREs; (4) statistical analysis of prognostic characteristics influencing survival in patients with melanoma bone metastases; and (5) data on bone metastases diagnosis, SREs development, therapy, and prognostic outcomes, when provided.Data not mentioned in the text of the studies but available on graphs or subgroup analysis were extrapolated and included in this systematic review.

Study inclusion and quality assessment
In total, 4,641 studies were screened after removing all duplicates.After screening the title and abstract, 4,605 studies were excluded, leaving 36 full-text studies.After assessing eligibility based on the selection criteria, the final set comprised 29 studies (Appendix B) [4,5,.Study quality assessment using NOS demonstrated that 66 % (19/29) of studies had a "low risk of bias" with scores above six, and 34 % (10/29) of studies had an "unclear risk of bias" with a NOS score of five.An "unclear risk of bias" was mainly due to the low comparability scores, as most studies involved only one cohort.All 29 studies were included for data extraction as none had a score lower than five, and thus, none were considered to have a "high risk of bias".(Appendix C, D).

Study characteristics
Except for one prospective single-center study, all 29 investigations were retrospective.Of the 28 retrospective cohorts, 79 % (22/28) were either single-center or two-center studies.Study periods ranged from three to 40 years between 1955 and 2019.Median follow-up time ranged from three to 76 months, with one study recording a minimum follow-up of 60 months [37].97 % (28/29) of studies focused solely on patients with melanoma as the primary tumor, and 59 % (17/29) of studies included only patients with bone metastatic melanoma.The number of patients enrolled in the studies ranged from 11 to 7,010.Of the 23,998 patients included in the 29 studies, 13 % (3,130/23,998) had a recorded bone metastatic melanoma.Male patients comprised 56 % of the study cohorts while female patients comprised 44 %.The median age of melanoma diagnosis ranged from 42 to 67 (Table 1).

Data analysis
A quantitative meta-analysis for pooling incidences and hazard ratios was not performed due to the heterogeneity in patient populations, stages of disease, differences in treatment approaches and different diagnosis modalities for both bone metastases and SREs, inconsistent follow-up times, and variability in reporting the outcomes.Providing pooled incidences or hazard ratios with the data for different outcomes would result in hazard ratios with minimal clinical validity.Because all included studies were of comparable quality, scoring five or greater on the NOS scale, a quantitative summarization was provided by utilizing

Factors associated with developing bone metastases and/or SREs
Four studies investigated risk factors associated with the development of bone metastases [13,19,28,31].Risk factors found to be independently associated with developing bone metastases included: age, younger than 40 years; primary lesions that were ulcerated, deeper than Clark level II, thicker than 0.76 mm, or located on trunk or mucosal surfaces; presence of tumor-infiltrating lymphocytes; and elevated LDH levels (all p < 0.05).Other factors considered but not found to be associated with bone metastases were race, gender, laterality, T stage and N stage, mitotic index, and BRAF or NRAS mutations.Studies on the factors associated with SREs development were limited; one study by Mannavola et al. investigated this using univariate and multivariate analysis.The authors found that only the use of bone modifying agents such as bisphosphonates or denosumab before developing SREs was an independent prognostic factor that decreased the risk of developing SREs (OR 0.38, 95 % CI 0.2-0.72,p = 0.003) [23].No clinical or tumorrelated variables in this study correlated with SREs development (Table 4).

Survival analysis
Of the studies included in this review, 66 % (19/29) investigated survival outcomes in patients with bone metastatic melanoma (Table 5).Survival after the detection of bone metastases ranged between three and 13 months.Sellin et al. found the median survival after 64 consecutive patients undergoing surgery for spinal metastases to be six months (95 % CI, 2.7-28.7)[26].Bostel et al. reported that only 37 % (15/41) of patients were alive six months after radiotherapy [35].Patients who underwent immunotherapy and palliative radiotherapy showed an overall survival with a median of 17 months (95 % CI, 10.0-23.3) in 108 patients [23].Factors found to be associated with survival included clinical (neurologic involvement, LDH and hemoglobin levels, Karnofsky performance status, ECOG), tumor-related (number and location of bone metastases, melanoma stage, presence of visceral/skin metastases, presence of pathological compression fracture of the spine), and treatment (systemic treatment) features.While some studies found no difference between axial and appendicular metastases on overall survival, DeBoer et al. found a survival advantage at 10 years in appendicular bone metastases (27 %) compared with axial metastases (0 %; p = 0.001) [39].None of the studies provided an easy-to-use nomogram or algorithm to calculate survival percentages.

Discussion
Bone metastases and SREs pose a common problem in patients with melanoma, with incidence ranging up to 49 % and 66 %, respectively.

M.R. Shimizu et al.
While research has improved our understanding of the incidence of bone metastatic melanoma, there remains a substantial lack of knowledge in the literature, especially on the incidence of SREs and predictive factors for the development of bone metastases and SREs.A more comprehensive understanding through future studies investigating and reporting on predictive factors may help with patient and physician decision-making, improving outcomes in this patient population.Particularly, the establishment of easy-to-use nomograms or prediction models for developing bone metastases and subsequent SREs may guide both clinicians and patients in the clinical decision-making process.In addition, it can help identify patients at high risk for developing bone metastases and/or SREs, enabling the implementation of preventive measures and screening policies.

Incidence and characteristics of bone metastases
The cumulative incidence of bone metastases in melanoma patients was 13% in this systematic review, with the majority of diagnoses confirmed by imaging.The studies included in this review demonstrated more axial skeleton metastases sites, comparable to other studies examining skeletal sites in other primary solid tumors such as prostate and colorectal [40,41].Given the similar pattern of distribution across multiple primary tumors, this could be attributed to differences in local bone microenvironments between the appendicular and axial skeleton [42].DeBoer et al. were unique in investigating patients with isolated bone metastasis [39].Brountzos et al. meanwhile reported 3.7 % of patients with bone metastases as the initial and sole site of recurrence in a study of 28 patients [36].While bone is a common site of metastasis, it is rarely an isolated event as it results from hematogenous spread, inherently making it likely that metastases also spreads to other organs.The proportion of patients with a BRAF and/or NRAS mutation was available in five studies.Apart from Brown et al., the incidence of BRAF mutations ranged between 49 % to 59 %, which is in concordance with that of all patients with melanoma [43].While the frequency of NRAS mutation in the included studies was seemingly higher than in the general melanoma population, as reported in previous literature [44], this was not statistically significant due to the limited data available.

Risk factors of bone metastases
Elevated LDH, a younger age at diagnosis, and a melanoma site in the trunk/axial or mucosal areas were associated with a higher risk of developing bone metastases.While the prognostic influence of the primary melanoma site is less significant than that of histological factors and age, trunk melanomas have been significantly associated with advanced clinical stages and unfavorable survival rates and outcomes [45].Its more aggressive characteristic could contribute to the metastatic spread to the bone.Potential reasons for a lower risk of bone metastases with increasing age could be angiogenic dormancy, the inability of micrometastases to grow due to a lack of vascularization, and age-related changes in the host tissue microenvironment [17].Interestingly, other potential risk factors of bone metastases showed conflicting associations, including histology (e.g., ulceration) and cancer staging.This discrepancy could be attributed to the fact that while these features predict a poorer prognosis for patients with metastatic melanoma, they do not specifically predict metastasis to bone.

Characteristics of SREs
Zekri et al. reported an estimated average of 2.5 SREs per patient for every year of follow-up, which is consistent with previous reports worldwide [4], with an average range of 1.9 to 2.9 SREs per patient-year across all advanced cancers in various parts of Europe [46].Radiation therapy and surgery were frequently used to manage SREs, serving as a common denominator in multiple studies investigating bone metastatic melanoma, where bone pain, pathological fracture, and spinal cord compression were often the indications for these interventions.While spinal cord compression and pathological fractures were recorded in approximately half of the studies related to SREs in this patient population, records on bone pain and hypercalcemia were sparse.Bone pain was a leading indication for intervention in all three studies that referenced SREs in patients with spinal metastatic melanoma [18,28,35].This may suggest that the stability of vertebral bodies affected by bone metastases greatly influences patients' quality of life and necessitates intervention for pain relief.Of the four studies that recorded hypercalcemia as an SRE, Zekri et al. reported the highest proportion of bone metastatic melanoma patients with hypercalcemia at approximately 23 %, demonstrating the imbalance between bone formation and resorption commonly seen in bone metastases from various primary tumors [4].

Risk factors of SREs
Mannavola et al. considered various melanoma and patientassociated factors for predicting SREs development in their patient cohort and identified the use of bone-directed agents as a protective factor against the development of SREs, resulting in a 62 % reduced risk of experiencing SREs and delayed time to their occurrence [23].The decreased risk and delayed occurrence are consistent with findings in other primary tumor types, including breast, lung, and colorectal cancer [47][48][49].The use of bone-directed agents such as bisphosphonates inhibits bone resorption, thereby decreasing serum calcium levels and delaying the pre-terminal SREs of hypercalcemia.Initiating earlier detection and treatment of bone metastases with bone-modifying agents could further amplify their potential benefits in reducing SREs occurrence.No other literature reviewed in this systematic study investigated risk factors of SREs.Further research focusing on risk factors for SREs is warranted to improve the quality of care.

Survivorship
The prognoses for patients with bone metastatic melanoma varied.Abdel et al. compared the survival between different metastatic sites within melanoma patients and observed worse survival outcomes for those who had single-site melanoma metastases at other sites, including skin, lymph nodes, and lungs [13].The inferior survival is likely associated with the sequelae of the disease, with melanoma metastasis to bone occurring at a later stage in disease progression compared to other sites [50].In studies prior to 2010, the average survival time ranged between four and six months [4,15,16,19,20].Mannavola et al. included the most recent cohort of patients up until 2019 with a median overall survival from diagnosis of bone metastases of 10.7 months.Among these patients, those who underwent both immunotherapy and palliative therapy exhibited the highest survival rate, reaching 16.5 months [23].These improvements in survival reflect the significant advancements in melanoma therapies that have been made in the past decade, with immunotherapies such as ipilimumab and targeted therapies such as vemurafenib offering promising options in treatment for patients with metastatic melanoma to the bone.
Survival rates of patients who underwent radiotherapy or surgical interventions were also reported.Bhanot et al. found that mortality following surgery for spinal metastatic melanoma was among the highest compared with other primary tumors, with a postoperative 90day mortality of 51 % (46/91) [34].It is important to emphasize that this does not reflect the survivability from the diagnosis of the primary tumor or bone metastases, as interventions might have occurred late in the disease process.In carefully selected patients, interventions can prolong survival.For example, Colman et al. reported a median survival of 11.8 months in those who underwent metastasectomy versus 4.8 months after nonoperative intervention, after controlling for patientand disease-related variables [38].
Twelve studies investigated prognostic factors, where multiple bone metastases or the presence of visceral metastases were associated with decreased survival [21][22][23][26][27][28][29][30][31]35,38,39].Lower Karnofsky score and higher ECOG were associated with decreased survival [21,27,31,35].Surprisingly, Bostel et al. discovered that the presence of skin metastases was associated with increased survival in patients with spinal metastases after palliative radiotherapy [35].This counterintuitive finding could be due to the cohort with skin metastases presented with better Karnofsky scores, lower rates of metastases to visceral organs, and more than one bone metastases.A multivariate analysis of this subgroup might have clarified this anomaly but was not conducted.In one retrospective review of 283 patients, patients with appendicular skeleton involvement survived three times as long as those with axial metastases, suggesting a survival advantage for this subset of patients [39].Systemic treatment, such as targeted therapy or immune checkpoint inhibitor, demonstrated favorable survival compared to chemotherapy alone [23] or no immunotherapy at all [27], suggesting the utility and effectiveness of these innovative therapies.Other clinical variables independently associated with poor survival included elevated LDH levels [23,30,31], larger tumor size [29], and lower hemoglobin levels [30].While several variables were explored across the studies, only a handful were consistently investigated, making it difficult to interpret the results.

Limitations and future recommendations
This systematic review has several limitations.First, the quality and comparability of the data were insufficient for a meta-analysis.However, we have systematically reviewed data according to the NOS guidelines to provide relevant information about current findings on bone metastatic melanoma and associated SREs.Second, we were limited to published material from the included studies, and other factors of interest, such as granular information regarding therapy, imaging methods used for diagnosis of bone metastases, and mutation type, were not readily available for review.Third, it was unclear whether patients were consecutively included in many of the studies, which introduces a risk of publication bias.Fourth, the study design and cohort differed among studies, making comparison difficult.Fifth, other relevant outcomes that could be considered more critical than survival, such as pain relief, preservation of physical function, quality of life, or adverse events following treatment for bone metastases and/or SREs, including surgical complications, hospitalizations, and reoperations, were not reported by the included studies.The findings of this review warrant well-designed retrospective and prospective studies that focus on the diagnosis, incidence and treatment outcomes of bone metastases and SREs in melanoma patients to better understand the current advancements and areas of improvement in treatments for this patient population.Providing user-friendly prediction models predicting not only survival but other important outcomes, such as quality of life, may guide and enhance the shared decision-making process of determining the optimal treatment.In addition, it can help start preventive measures and policies by identifying patients at high risk for developing bone metastases and subsequential detrimental SREs, such as early-on imaging for patients at high risk of developing bone metastases.

Conclusion
Advances in the treatment of melanoma have prolonged survival but have also resulted in increasing frequency of bone metastases ranging up to 49 %.The elevated incidence of bone metastases and the subsequent risk of SREs continue to pose challenges in patient outcomes and healthcare management.For all outcomes, present, for each study: (a) summary statistics for each group (where appropriate) and (b) an effect estimate and its precision (e.g.confidence/credible interval), ideally using structured tables or plots.

Results of syntheses 20a
For each synthesis, briefly summarise the characteristics and risk of bias among contributing studies.

NA 20b
Present results of all statistical syntheses conducted.If meta-analysis was done, present for each the summary estimate and its precision (e.g.confidence/credible interval) and measures of statistical heterogeneity.If comparing groups, describe the direction of the effect.

NA 20c
Present results of all investigations of possible causes of heterogeneity among study results.

NA 20d
Present results of all sensitivity analyses conducted to assess the robustness of the synthesized results.

Table 1
Cohort characteristics of all included studies with melanoma patients (n = 29).range of incidence of bone metastases and SREs and listing risk factors for SREs and survival.We used Microsoft Excel Version 16.76 (Microsoft Inc, Redmond, WA, USA) to extract data using standardized forms and to create all figures and tables, Mendeley Desktop Version 1.19.8 (Mendeley Ltd, London, UK) as reference software, and Rayyan.ai(Rayyan Systems Inc, Boston, MA, USA) for article screening.
Abbreviations: N = number of patients; yr = year; BMM = bone metastatic melanoma; M/F = male/female; BRAF = b-raf murine sarcoma viral oncogene homolog B1; NRAS = neuroblastoma ras viral oncogene homolog; T1 = time point measured until death; mo = months; USA = United States of America; UK = United Kingdom; BM = bone metastasis a Overall survival is measured as the time from T1 to death.bMedian reported c Mean reported M.R. Shimizu et al.the

Table 4
Analysis of risk factors for developing bone metastasis and skeletal-related events in melanoma patients (n = 6).

Table 5
Analysis of survival of melanoma patients (n = 19).
29e current systematic review provides a comprehensive overview of the incidence, risk factors, and outcomes associated with bone metastases and SREs in patients with melanoma encompassing29studies.Yet, existing literature with high-quality data Abbreviations: UA = univariate analysis; MA = multivariate analysis; KM = Kaplan-Meier; RT = radiotherapy; ECOG = Eastern Cooperative Oncology Group; LDH = lactate dehydrogenase; KPS = Karnofsky Performance Status; BDA = bone directed agents; SINS = Spinal Instability Neoplastic Score; ESCC = epidural spinal cord compression; OS = overall survival; BS = bone survival; PFS = progression-free survival; BM = bone metastasis (continued ) Conway JW, Rawson RV, Lo S, Ahmed T, Vergara IA, Gide TN, et al.Unveiling the tumor immune microenvironment of organ-specific melanoma metastatic sites.J Immunother Cancer [Internet].2022;10(9).Available from: http://dx.https://doi.org/10.1136/jitc-2022-004884Nofocus on melanoma bone metastases Appendix C. Quality assessment according to the Newcastle-Ottawa Scale (NOS) Specify the methods used to collect data from reports, including how many reviewers collected data from each report, whether they worked independently, any processes for obtaining or confirming data from study investigators, and if applicable, details of automation tools used in the process.alloutcomesforwhichdataweresought.Specify whether all results that were compatible with each outcome domain in each study were sought (e.g. for all measures, time points, analyses), and if not, the methods used to decide which results to collect.required to prepare the data for presentation or synthesis, such as handling of missing summary statistics, or data conversions.NA 13cDescribe any methods used to tabulate or visually display results of individual studies and syntheses.NA 13dDescribe any methods used to synthesize results and provide a rationale for the choice(s).If meta-analysis was performed, describe the model(s), method(s) to identify the presence and extent of statistical heterogeneity, and software package(s) used.P. 8 13eDescribe any methods used to explore possible causes of heterogeneity among study results (e.g.subgroup analysis, meta-regression).NA 13fDescribe any sensitivity analyses conducted to assess robustness of the synthesized results.NA Reporting bias assessment 14 Describe any methods used to assess risk of bias due to missing results in a synthesis (arising from reporting biases).Describe any methods used to assess certainty (or confidence) in the body of evidence for an outcome.NARESULTSStudy selection 16aDescribe the results of the search and selection process, from the number of records identified in the search to the number of studies included in the review, ideally using a flow diagram.Fig.1; P. 7 16bCite studies that might appear to meet the inclusion criteria, but which were excluded, and explain why they were excluded.
Indicate where the review protocol can be accessed, or state that a protocol was not prepared.P. 5 24cDescribe and explain any amendments to information provided at registration or in the protocol.NA Support 25 Describe sources of financial or non-financial support for the review, and the role of the funders or sponsors in the review.Report which of the following are publicly available and where they can be found: template data collection forms; data extracted from included studies; data used for all analyses; analytic code; any other materials used in the review.PageMJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, et al.The PRISMA 2020 statement: an updated guideline for reporting systematic reviews.BMJ 2021;372:n71.https://doi.org/10.1136/bmj.n71.Studies after duplicates removed(n = 4,641). NAFrom: