Low Bone Mineral Density and Bone Metastases – a Two-Way Street

1. Introduction

Bone metastases (BM) have a detrimental effect on survival for patients with solid cancers. Characteristics such as number of bone metastases, time from diagnostic to bone metastasis development [1], alkaline phosphatase levels [2], and others can be used as prognostic factors in survival estimation for patients with bone metastatic disease. The estimated time to develop bone metastasis is around 400 days [3] with a cumulative incidence of 5.6% at two years, 6.9% at five years, and 8,4% at ten years [3]. The tendency to develop bone metastasis varies widely between the cancer subtype and stage at diagnosis. Prostate cancer patients have the highest risk of developing bone disease followed by lung, renal, and breast [3]. While some cancers have a high tendency to develop bone metastasis in time, a combination of high survival with a high risk of developing bone metastasis leads to high prevalence rates, some cancers have a very high risk of bone disease at diagnosis. At diagnosis lung – 18.05%, liver-6.63%, nasopharyngeal carcinoma – 6.33%, and renal cancer -5.45% had the highest rates of presenting themselves with bone metastasis [4]. The presence of BM increased the risk of mortality and decreased survival compared to previous stages, especially for patients with prostate cancer (aHR: 18,24) [4].

While survival rates for cancer patients have been improving, due to advancements in therapeutic options, the prevalence of metastatic bone disease has been on the rise. Due to the increased prevalence of BM, the complications of this bone disease are experienced by an increasing number of patients. Reports from 2004 from NIH estimated that in the US 1.5 million people suffer a fracture due to metastatic bone disease [5]. Other skeletal complications that impede the quality of life of patients with BM are hypercalcemia, bone pain, spinal cord compression due to BM, and cachexia [6]. All these complications require complex interventions such as surgical interventions, radiotherapy, systemic treatments, pain management, nutritional management, etc. Management of skeletal-related events (SREs) also imposes a significant economic burden on medical systems. The estimated lifetime SRE-related cost per lung cancer patient with bone metastatic disease was 11,979 USD (95% CI 10,193-13,766 USD) in a retrospective analysis performed on a large US health insurance claims database published in 2005 [7]. Radiotherapy accounted for the highest proportion of the cost. High costs of SRE management are also reported in Europe, with regards to costs involved in treating a prostate cancer patient with bone metastasis, a study from The Netherlands published in 2003 [8], reported an average cost of treatment per patient of €13,051 over 24 months with an average cost of €6973 per patient to treat SREs [8]. A systematic review of relevant economic analyses regarding the management of bone metastasis concluded that bisphosphonate treatments are cost-effective options, and denosumab has a less advantageous value proposition. Evidence concerning EBR and Sr89 is limited and less definitive [9].

Osteoporosis (OP) is a pathological state of the skeletal system in which bone mineral density and bone mass decrease, many studies identify cancer patients as being at major risk for developing this bone disease [10]. Having certain kinds of cancer, such as cervical cancer [11], and cancer-specific treatments can predispose patients to osteoporosis [12]. Many modern treatments such as androgen deprivation therapy (ADT), aromatase inhibitors (AIs), selective estrogen receptor modulators (SERMs), chemotherapy regimens, high doses of corticosteroids usage over long periods, and other treatments used in cancer patient treatment can lead to early development of osteoporosis [12].

While an undeniable relationship exists between cancer patients during treatment and an increased risk of osteopenia and osteoporosis development, some argue that an inverse relationship also exists between patients with osteoporosis and an increased risk of developing bone metastases. The idea that metastatic cancer cells don’t develop in random patterns dates to 1889 with the “seed and soil” hypothesis. Recent studies refined our understanding of specific microenvironment traits that can lead to the development of cancer cell colonies from a micro-metastasis state to macro-metastasis [13]. Some studies suggest the existence of a so-called “pre-metastatic niche” characterized by the presence of abnormal immune cells and a specific configuration of the extracellular matrix proteins [13]. This niche seems to be created by a combination of soluble factors produced by the primary tumor and other non-tumor-dependent factors such as endocrine or metabolic factors.

The endocrine system plays a key role in the development of osteoporosis and, some authors argue, in developing the premetastatic niches needed to develop bone metastases in certain subtypes of cancers [14]. Bone homeostasis is maintained by a fine-tuned balance between bone formation promoted by osteoblast activity, a cellular subset derived from mesenchymal stem cells, and bone removal performed by the osteoclasts, a cellular subset derived from hematopoietic stem cells [15]. The major endocrine promoters of high bone turnover found in malignant pathologies are PTH and PTH-like substances, which also have a key role in calcium-level regulation, high glucocorticoid levels, low progesterone, androgen, and estrogen levels. The osteoblasts and osteoclasts present on their surface estrogen receptors thus making estrogen levels especially important in osteoporosis pathogenesis and bone metastasis formation [16]. Antiresorptive agents such as bisphosphonates and anti-RANKL monoclonal antibodies are widely used in patients with metastatic bone disease treatment regimens. These agents have a beneficial effect on bone density and reduce the fracture risk of these patients. While bone density increases predict a risk reduction for subsequent fractures, BMD increases and reductions do not fully explain fracture risk [17]. Bone architecture plays a major role alongside bone mass density (BMD) in the physical characteristics of the bone, for example, lumbar vertebrae murine models showed that two of the strongest correlations that predict the compressive strength of the bone are BMD and trabecular bone volume [18]. Some studies report that antiresorptive agents, such as risendronate and others, have a beneficial effect on bone architecture as well as BMD [19]. Due to the intertwined relationship between osteoporosis and bone metastasis development, we designed a systematic literature review with two main goals. Firstly, we investigated osteoporosis as a risk factor for bone metastasis development, and secondly, we investigated the risk factors for osteoporosis development in oncological patients.

2. Materials and Methods

2.1. Search Strategy and Study Selection

The systematic review respected the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines [20]. The identification of relevant studies was done by searching multiple online bibliographic databases such as Google Scholar, Cochrane Library, Embase, International Clinical Trials Registry Platform (ICTRP), and Cumulative Index to Nursing and Allied Health Literature (CINAHL) database. ClinicalTrials.gov was also included in the research to identify any ongoing trials concerning osteoporosis and bone metastases. The search was performed from inception until the date of the search, August 2024. Citations of the retrieved articles were also analyzed using a forward-backward search method. The publications selected this way were chosen using the same selection criteria as the rest of the articles. The search query included combinations of the terms: “osteoporosis”, “cancer patients”, “bone metastases/metastasis”, “risk factors”, and “solid cancers”. The selection process involved screening the title and abstract of the retrieved articles and extended reading of selected articles, any conflicts of selection between the first two authors were resolved by the third author. Two selection processes were performed, one in which the focus was on identifying OP as a risk factor for bone metastases development, and, for the secondary objective of the study a second selection was performed in which all risk factors for osteoporosis development in the context of cancer patients were identified.

2.2. Data Extraction

Data was extracted from the articles by the authors and two databases were created in Microsoft Excel spreadsheet application. In the first database studies that investigated OP as a risk factor for BM were selected. An ID was attributed to each study with an appropriate identification tied to the study reference, year of publication, country of publication, study type, number of patients included, controls, groups that were compared, and statistical analysis measures were extracted from the main text and summarized in the database. In the second database, data from studies that investigated risk factors for OP development in cancer patients was selected. An ID was attributed to each study with an appropriate identification tied to the study reference, year of publication, country of publication, number of patients investigated, cancer subtype investigated, specified risk or protective factor, type of effect of the said factor, and statistical analysis measures of the effect.

2.3. Inclusion and Exclusion Criteria

Studies that were peer-reviewed with available texts in English were included. Types of studies evaluated for inclusion have consisted of randomized controlled trials, cross-sectional, cohort studies, case-control, systematic analyses, and meta-analyses. When analyzing studies exploring OP as a risk factor for BM the following inclusion criteria were used: (1) Randomised control trials, cross-sectional, cohort studies, and case-controls; (2) Studies that explored OP as a risk factor for BM; (3) Studies that included adult patients, defined as above 18 years of age. Exclusion criteria (1) Language of publication – not English; (2) Studies that included pediatric patients; (3) Articles that explored only animal models/cell lines. When exploring the risk and protective factors involved in OP development in oncological contexts, the same inclusion and exclusion criteria were applied with the mention that the included study types were extended to include systematic analyses and meta-analyses.

2.4. Quality Assessment

Quality assessment for the studies that analyzed the main objective of the study was performed using the NIH Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies [21]. This tool includes a 14-item questionnaire in which a total score of 14 points can be obtained. Each question can be answered in one of 3 ways: Yes (1 point); No (0 points); or CD, cannot be determined (0 points)/NA, not applicable (0 points)/NR, not reported (0 points). The questions evaluate how the research was done; study population; groups recruited from the same population and uniform eligibility criteria; sample size justification; exposure assessed before outcome measurement; sufficient timeframe to see an effect; different levels of the exposure of interest; exposure measures and assessment; repeated exposure assessment; outcome measures; blinding of outcome assessors; follow up rate and, lastly statistical analyses. A quality rating of good, fair, or poor can be attributed to each evaluated study as a conclusion to the assessment.

3. Results

3.1. Search Results

The online database searches, with multiple search combinations of the keywords, as mentioned earlier led to the identification of 382 articles of interest and 4 articles identified through citation research. The databases were searched from the date of inception until June 2024. For the main objective of this review only studies that reported a statistical analysis of the risk OP represents for BM development were selected, the selection process is detailed in Figure 1. Only 3 studies respected the inclusion criteria and were included in the final analysis.

For the second objective of the study, investigating risk factors reported in studies as being risk factors for OP development in the context of cancer patients, to the 382 articles found through databases research 10 more articles were added through citation research. A total of 61 articles were assessed in full text, which led to the identification of 32 articles that met the inclusion criteria as detailed in Figure 2. For the secondary objective, the inclusion criteria have been expanded to include not only studies that reported a statistical analysis of the identified risk factor for OP development but also systematic reviews and meta-analyses that concluded concerning said risk factor. The reasoning behind this decision was supported by the intent of the secondary objective, which is the identification of said risk factors. A meta-analysis of the data reported in this systematic analysis was excluded as the lack of uniformity of data reporting and the lack of data with regards to specific subgroup analyses such as cancer types, cancer stages, pre-postmenopausal female patients, and others, led to a high heterogeneity of data which in turn made a meta-analysis of this statistical data undue at the time of analysis.

3.2. Characteristics of the Studies and Participants – OP a Risk Factor for BM Development?

Only limited data regarding this topic is available. All 3 studies identified were large retrospective studies, the first two studies [22,23], one from Taiwan and the other from South Korea, investigated breast cancer patients. In the first study, data was extracted from a nationwide database, the intrinsic subtype of breast cancer was not specified. This study concluded that even if precancer osteoporosis was not associated with a difference in BM incidence, in the context of breast cancer patients, those who had precancer osteoporosis had a shorter time for BM development. This predisposition was mitigated by treating OP as seen in Table 1. To sum up, OP does not increase the incidence of de novo BM, but it leads to accelerated BM progression when it occurs if left untreated.

In the second study [23], 348 patients with Luminal A breast cancer subtype were investigated. In this retrospective study, the patients were grouped into 2 categories, normal BMD vs low BMD (the authors included osteoporosis and osteopenia in this category). No statistically significant link between BMD and BM was found. Based on the trend of BMD after surgery, the subgroup analysis showed that upward trends, downward trends, and stable BMD did not show any significant differences either.

The last articles explored the relationship between OP and BM in the context of urologic cancers. Published in 2024 in the USA [24], it investigated a large, matched control cohort from a pool of 685066 urological cancer patients, as seen in Table 1. The article concluded that there is a strong correlation between precancer deficit of bone mineral density and BM apparition, especially in the first week after diagnosis for all investigated cancer subtypes. Bisphosphonate administration reduces the risk of BM for kidney, bladder, and prostate cancers.

To sum up the findings, for breast cancer a better investigation of OP as a risk factor for BM is required as many articles do not differentiate between intrinsic breast cancer subtypes [22]. Up to this point, BMD does not seem to contribute in a significant manner when investigating Luminal A breast cancer [23] - defined as estrogenic receptor (ER) and progesterone receptor (PR) positive, HER 2 negative, ki67 low [25]. When investigating breast cancer regardless of intrinsic subtype, untreated OP seems to increase the risk of bone metastasis after diagnosis when compared to no OP group or treated OP group. In contrast to breast cancer, which appears to be a heterogeneous group due to the different behavior that it has due to the intrinsic subtypes of cancer that it encompasses, in all urological cancer cases, OP seems to predispose the patients to BM, especially early after diagnosis.

3.3. Characteristics of the Studies and Participants – Risk Factors for OP Development in Cancer Patients

Out of the 32 articles identified, as seen in Figure 2, most articles identified risk factors for osteoporosis development in the context of breast cancer patients. 19 articles contained data regarding breast cancer patients [26,27,28,29,30,31,32,33,34,35,36,37,38,44,47,49,51,52,54,57], most common deleterious factors for bone mass density identified were treatments that led to ovarian failure and a low estrogenic bone environment. The most notable examples are chemotherapy that resulted in ovarian insufficiency/ovarian failure, an analog of luteinizing hormone-releasing hormone usage, aromatase inhibitors usage, steroidal inhibitor- exemestane, SERM usage when used in the pre-menopausal setting, oophorectomy, high dose glucocorticoids and lastly but not least breast cancer in of itself can serve as a risk factor for OP, articles findings summarized in Figure 2. Protective factors identified in this setting include weight/high BMI (body mass index), high levels of gonadal inhibin of the transforming (TGF)-B superfamily levels, SERM usage in the postmenopausal/low estrogen context, statin usage, melatonin supplementation, and mixed-mode exercises.

Prostate cancer patients are also explored in large meta-analyses [41], large propensity score-matched retrospective studies [43], and other studies, Table 2. The most frequent detrimental effects identified are the presence of bone metastasis, the presence of prostate cancer, low BMI, advanced age, high dose glucocorticoids, inactivity, bilateral orchiectomy, and most notably ADT (androgen deprivation therapy) usage, which is the backbone of prostate cancer treatment in both metastatic and neo/adjuvant treatment [58]. One of the most cited protective factors is physical activity with many regimens being studied now [47] to determine the most effective regimen for this scenario. Other urological malignancies that imply a raised risk of OP development are renal cancers. For most non-metastatic cases of renal cancer, the routine treatment involves either radical nephrectomy or nephron-sparing surgery. Compared to the more radical surgical approaches nephron-sparing surgery has a protective effect on BMD [53].

Another cluster of studied cancers is gynecological malignancies. While most authors agree that cervical and ovarian cancers have a detrimental effect on BMD [39,45,55,56] even in the pretreatment setting, the same thing cannot be concluded for endometrial cancer [39; 45]. Treating all 3 types of malignancies has a deleterious effect on BMD. Changes in BMD are lowest after surgical treatment. The highest detrimental effect on BMD is observed when multimodal treatment involving chemoradiotherapy after surgery is used [45].

Other, more scarcely researched risk factors, like melatonin supplementation [54], statins administration [49], or post-operative thyroid-stimulating hormone suppression after surgery for papillary or follicular thyroid carcinoma surgery [46] have also been included in this review, summaries available in Table 2. Melatonin supplementation seems to have a protective effect on bone density, but notable changes in BMD have been observed only in prolonged administration regimens [54]. Statins appear to have a favorable impact on BMD, with multiple articles observing an inverse correlation between LDL cholesterol levels and BMD at multiple sites such as the spine and hip while HDL cholesterol levels have conflicting correlations with BMD, mostly measured at the femoral neck level in this scenario [49]. Postoperative thyroid–stimulating hormone suppression has also been explored, but no statistically significant difference has been observed [46].

3.4. Quality Assessment

As mentioned before, Quality assessment for the studies that analyzed the study's main objective, as seen in Table 1, was performed using the NIH Quality Assessment Tool for Observational Cohort and Cross-Sectional Studies [21]. Scoring between 0-5 points was interpreted as being Poor, 6-10 as being Fair, and 11-14 as being Good. The first two authors evaluated the studies, and the third author mediated any conflicts. All three studies were graded as being Fair.

4. Discussion

While investigating OP as a risk factor for BM, from the studies included in Table 1, it becomes clear that there is a high grade of heterogeneity in the behavior of each type of cancer in its dynamic concerning bone metastasis, and thus, OP might have different behaviors as a risk factor depending on the situation. When investigating breast cancer, regardless of intrinsic subtype, untreated OP seems to increase the incidence of BM when compared to the normal BMD group or the treated subgroup after it has been diagnosed [22]. The same things cannot be said about the risk of BM in the OP group vs. the normal BMD group at diagnosis [22]. Also, breast cancer needs to be investigated in this regard considering the intrinsic subtype investigated, because, for example, lower BMD does not seem to impact Luminal A breast cancer patients in the same way [23]. Recent ESMO breast cancer guidelines have included adjuvant bisphosphonates as part of the standard therapy in early breast cancer, with the mention that bisphosphonate therapy is approved for treating bone metastases and osteoporosis and not for the prevention of relapse, with a level of evidence of [I, A] [59].

By contrast, urological cancers seem to exhibit a more homogenous behavior. The strongest association between OP and BM is observed 1 week after diagnosis, with adjusted odds ratios for bladder: 2.37; kidney: 2.37; and prostate: 2.84 being similar and statistically significant (all P<.001) [24]. The only outlier is testicular cancer in which the adjusted odds ratio was reported to be 4.45 with a P<.001 [24]. Still, this result only supports the trend presented by the other urological malignancies. Data regarding low bone mineral density and its effect on the risk of bone metastasis in other cancers is lacking. More studies are needed to understand and adjust treatment intensity with antiresorptive agents. Data thus far points towards a necessity in screening oncological patients for osteoporosis and osteopenia and treating it accordingly as it may also impact the oncological outcomes.

The need for implementing a coherent screening of bone mineral density in all oncological patients is also backed by the plethora of risk factors that oncological patients who receive treatment are subjected to. Multiple solid cancers present an increased risk of osteoporosis development, such as prostate cancer [41], cervical cancer [45,55,56], ovarian cancer [45], and BM presence [43]. Also, most treatments used that induce ovarian failure, testicular failure, complex multimodal treatments, and high doses of corticosteroid usage also seem to create a context favorable for BMD degradation, Table 2.

5. Conclusions

In conclusion, it has been found that oncological patients with low bone mineral density have an increased tendency to develop bone metastases. The importance of osteoporosis as a risk factor for developing bone metastases varies depending on the type of cancer that it interacts with. A good example in this regard is a heterogeneity of behavior in this regard can be observed when investigating different types of breast cancer. Also, there are a myriad of cancer-related risk factors and cancer treatment-related risk factors that predispose oncological patients to bone loss. There is a need for data regarding cancer subtypes that have not been explored concerning their interaction with the skeletal system in a low bone mineral density scenario. These results can lead to the development of a coherent strategy and an intensive osteoporosis treatment can lead to improved quality of life due to a decrease in bone metastasis incidence.