Features of Spinal Metastases: A Retrospective View

INTRODUCTION

The skeleton is the most common region where primary cancers metastasize.¹ Bone metastasis is more common than primary bone cancers, especially in adults.² Although the overall incidence of bone metastases is unknown, publications show bone metastasis occurs in two-thirds of primary cancer patients.^3,4 Breast, prostate, and lung cancers are responsible for 80% of bone metastasis due to their high prevalence.^5,6

The spine is the most common site of bone metastasis.^7–9 About 70% of patients with cancer may develop spinal metastasis, and 20% of them often suffer from neurological deficits.^10,11 In these patients, 10% of spinal cord compression can develop, and surgical decompression may be needed with or without stabilization.^10,12 Bone metastases are the main cause of morbidity. They can cause severe pain, especially at night, decreased mobility, pathological fractures, spinal cord compression, bone marrow aplasia, and hypercalcemia.^13–15

Primary cancer-to-bone metastasis occurs through systemic blood circulation or vertebral venous plexus.^7–9 While overall bone metastases, especially extremity and visceral metastases, occur by bloodstream spread, diffusion from the venous system is the main process of spinal metastases.^16–18 The vertebral venous plexus is a valveless network system of veins extending from the cranium to the pelvis.¹⁶ Many studies exist on the effects of systemic blood circulation on bone metastasis, but few studies have examined the effects of the vertebral venous plexus on bone metastasis.^16,18,19

This study, in which visceral and extremity metastases occuring through blood circulation were not included, was carried out both to evaluate the features of spinal metastases and to reveal clues to shed light on the origin of spinal metastases with unknown primary.

MATERIAL AND METHODS

RESULTS

We included 156 patients (59 females and 97 males) in our study. Mean age was 65.21 ± 12.19 (range, 35–96). Males were significantly older than the females (P = .008; Table 1). The most common primary cancer was respiratory system cancers (47.44%) among all patients. The most common primary cancers were breast (54.24%), respiratory system (23.73%), and gastrointestinal tract (10.17%) cancers in females, and respiratory system (61.86%), prostate (18.56%), and gastrointestinal tract (10.31%) cancers in males (P < .001). Seventy-one (73.20%) male patients had multiple metastases, while 45 (76.27%) female patients had multiple metastases (P = .812; Table 1). The most common metastasis region was thoracic + lumbar spine for both genders (P = .292; Table 1).

Fifty-three (33.97%) patients had cervical spinal metastases. Metastasis was most common on C5 (15.38%) among all patients. Ten (6.41%) of them were on corpus, 4 (2.56%) on posterior, and 10 (6.41%) on both. Metastasis was most common on C5 (23.73%) in females, while most common on C7 (15.46%) in males. C5 spinal metastasis percentage was significantly higher in females than in males (P = .043; Table 1). No significant differences existed between genders regarding presence of other cervical spinal metastases (Table 1).

One hundred and twenty (76.92%) patients had thoracic spinal metastases. Metastasis was most common on T10 (28.21%) among all patients. Twenty-six (16.67%) of them were on corpus, 4 (2.56%) on posterior, and 14 (8.97%) on both. Metastasis was most common on T8 (35.59%) in females, while most common on T9 (26.80%) in males. T8 spinal metastasis percentage was significantly higher in females than in males (P = .028; Table 1). No significant differences existed between genders regarding presence of other thoracic spinal metastases (Table 1).

One hundred and nine (69.87%) patients had lumbar spinal metastases. Metastasis was most common on L3 (38.46%) among all patients. Twenty-nine (18.59%) of them were on corpus, 8 (5.13%) on posterior, and 23 (14.74%) on both. Metastasis was most common on L3 both in females (45.76%) and in males (34.02%). L1 spinal metastasis percentage was significantly higher in females (42.37%) than in males (21.65%; P = .010; Table 1). No significant differences existed between genders regarding presence of other lumbar spinal metastases (Table 1). Each spinal metastasis is shown in the Figure.

• Download figure
• Open in new tab
• Download powerpoint

Figure

Percentages of each spinal metastasis.

Patients with prostate cancer were significantly older than patients with respiratory system, breast, gastrointestinal tract, and gynecologic cancers (P = .001; Table 2). No significant differences existed between other cancer types regarding age. Fifty-four (72.97%) patients with respiratory system cancers had multiple metastases, while 29 (87.88%) patients with breast cancer, 8 (50.00%) patients with gastrointestinal tract cancers, 6 (60.00%) patients with urinary tract cancers, 15 (83.33%) patients with prostate cancer, and 4 (80.00%) patients with gynecologic cancers had multiple metastases (P = .071; Table 2). The most common metastasis region was the thoracic spine in the respiratory system cancers (28.38%), the thoracic + lumbar spine in breast (42.42%), prostate (50.00%), and gynecologic (40.00%) cancers, and the lumbar spine in gastrointestinal (37.50%) and urinary (30.00%) tract cancers (P = .313; Table 2).

Twenty-six (35.14%) patients with respiratory system cancers had metastasis on the cervical spine, while 13 (39.39%) patients with breast cancer, 6 (37.50%) patients with gastrointestinal tract cancers, 4 (40.00%) patients with urinary tract cancers, 3 (16.67%) patients with prostate cancer, and 1 (20.00%) patient with gynecologic cancer had metastases on the cervical spine (P = .613; Table 3). Respiratory system cancer metastasis was most common on C7 (14.86%). Four (5.41%) of them were on corpus, 4 (5.41%) on posterior, and 3 (4.05%) on both. Breast cancer metastasis was most common on C5 (30.30%). Two (6.06%) of them were on corpus, 3 (9.09%) on posterior, and 5 (15.15%) on both. Gastrointestinal tract cancer metastasis was most common on C5 (25.00%). One (6.25%) of them was on corpus, 1 (6.25%) on posterior, and 2 (12.5%) on both. Urinary tract cancer metastasis was most common on C5 (20.00%). One (10.00%) of them was on corpus and 1 (10.00%) on both corpus and posterior. Prostate cancer metastasis was most common on C4 (16.67%). Two (11.11%) of them were on corpus and 1 (5.56%) on both corpus and posterior. C5 spinal metastasis percentages were significantly higher in breast and gastrointestinal tract cancers than in respiratory system cancers (P = .042; Table 3). No significant differences existed between cancer types regarding presence of other cervical spinal metastases (Table 3).

Sixty (81.08%) patients with respiratory system cancers had metastases on the thoracic spine, while 27 (81.82%) patients with breast cancers, 9 (56.25%) patients with gastrointestinal tract cancers, 6 (60.00%) patients with urinary tract cancers, 14 (77.78%) patients with prostate cancer, and 4 (80.00%) patients with gynecologic cancers had metastases on the thoracic spine (P = .247; Table 4). Respiratory system cancer metastasis was most common on T9 (27.03%). Fifteen (20.27%) of them were on corpus, 2 (2.70%) on posterior, and 3 (4.05%) on both. Breast cancer metastasis was most common on T8 (39.39%) and T10 (39.39%). Nine (27.27%) of the T8s were on corpus, 1 (3.03%) on posterior, and 3 (9.09%) on both, while 9 (27.27%) of T10s were on corpus and 4 (12.12%) on both corpus and posterior. Gastrointestinal tract cancer metastasis was most common on T3 (31.25%). Four (25.00%) of them were on corpus and 1 (6.25%) on both corpus and posterior. Urinary tract cancer metastasis was most common on T7 (30.00%). Two (20.00%) of them were on corpus and 1 (10.00%) on posterior. Prostate cancer metastasis was most common on T10 (50.00%). Four (22.22%) of them were on corpus, 1 (5.56%) on posterior, and 4 (22.22%) on both. Gynecologic cancer metastasis was most common on T5 (80.00%). Three (60.00%) of them were on corpus and 1 (20.00%) on posterior. T5 spinal metastasis percentage was significantly higher in the gynecologic cancers than in the other cancers (P = .002), while T10 spinal metastasis percentages were significantly higher in the prostate and gynecologic cancers than in the respiratory system, gastrointestinal, and urinary tract cancers (P = .016). No significant differences existed between cancer types regarding presence of other thoracic spinal metastases (Table 4).

Forty-five (60.81%) patients with respiratory system cancers had metastases on the lumbar spine, while 27 (81.82%) patients with breast cancer, 12 (75.00%) patients with gastrointestinal tract cancers, 5 (50.00%) patients with urinary tract cancers, 16 (88.89%) patients with prostate cancer, and 4 (80.00%) patients with gynecologic cancers had metastases on the lumbar spine (P = .061; Table 5). Respiratory system cancer metastasis was most common on L3 (33.78%). Sixteen (21.62%) of them were on corpus, 3 (4.05%) on posterior, and 6 (8.11%) on both. Breast cancer metastasis was most common on L1 (54.55%) and L3 (54.55%). Nine (27.27%) of L1s were on corpus, 4 (12.12%) on posterior, and 5 (15.15%) on both, while 7 (21.21%) of L3s were on corpus, 3 (9.09%) on posterior, and 8 (24.24%) on both. Gastrointestinal tract cancer metastasis was most common on L4 (43.75%). Four (25.00%) of them were on corpus and 3 (18.75%) on both corpus and posterior. Urinary tract cancer metastasis was most common on L3 (30.00%). Two (20.00%) of them were on corpus and 1 (10.00%) on posterior. Prostate cancer metastasis was most common on L5 (66.67%). Six (33.33%) of them were on corpus, 1 (5.56%) on posterior, and 5 (27.78%) on both. Gynecologic cancer metastasis was most common on L2 (60.00%). One (20.00%) of them was on corpus and 2 (40.00%) on both corpus and posterior. L1 spinal metastasis percentage was significantly higher in breast cancer than in respiratory system, gastrointestinal tract, urinary tract, and gynecologic cancers (P = .009). L2 spinal metastasis percentages were significantly higher in breast, prostate, and gynecologic cancers than in respiratory system, gastrointestinal tract, and urinary tract cancers (P = .011). L4 spinal metastasis percentages were significantly higher in breast and prostate cancers than in respiratory system, urinary tract, and gynecological cancers (P = .041). L5 spinal metastasis percentage was significantly higher in prostate cancer than in respiratory system, breast, gastrointestinal tract, and urinary tract cancers (P = .029). No significant differences existed between cancer types regarding presence of L3 spinal metastases (Table 5).

DISCUSSION

In this study, clinical data of 156 patients with isolated spinal metastasis without visceral organ metastasis were analyzed retrospectively, including demographic characteristics (age, gender), pathological diagnostic reports, and PET-CT results. Spinal involvement characteristics of primary cancers were evaluated by analyzing both the main spine structures and specific locations in each. Although not statistically significant, respiratory system cancers metastasize to the thoracic spine (23.38%), breast cancer metastasizes to the thoracic + lumbal spine (42.42%), gastrointestinal (37.5%) and urinary (30%) tract cancers metastasize to the lumbal spine, and prostate (50.0%) and gynecological (40%) cancers metastasize to the thoracic + lumbal spine commonly. These statistics suggest that a primary cancer tends to metastasize to a given spine section and specific location within. The C5 vertebrawas the most frequently involved spine in the cervical spine. Respiratory system cancers most common ly metastasized to the corpus or posterior of the C4 spine, and prostate cancers metastasized to the corpus of the C4 spine. Breast and urinary tract cancers were statistically significantly more metastasized to the C5 spine than other cancers (P < .05; Table 3). While breast cancer spread most commonly to the corpus + posterior of the C5 spine, urinary tract cancers have metastasized equally to the corpus and corpus + posterior regions of the C5 spine. As a result of the present study, while respiratory system cancers metastasized to the corpus of the T9 spine in the thoracic spine commonly, breast cancer metastasized to the corpus of the T8 and T10 spine, gastrointestinal tract cancers metastasized to the corpus of the T3 spine, urinary tract cancers metastasized to the corpus of the T3 spine, prostate cancer spread to the corpus and corpus + posterior of the T10 spine, and gynecological cancers metastasized to the corpus of the T5 spine most commonly. It has been observed that gynecological cancers statistically significantly metastasized to the T5 spine, and prostate cancer metastasized to the T10 spine more than other cancers (P < .05; Table 4). In the lumbar spine, it has been observed that breast cancer statistically significantly metastasized to the L1 spine, breast, prostate, and gynecological cancers to the L2 spine, breast and prostate cancer to the L4 spine, and prostate cancer to the L5 spine more than other cancers (P < .05; Table 5).

The type and location of the primary cancer is unknown in 3%–15% of all patients with metastatic carcinoma, and approximately 5%–20% of these patients constitute the first detectable lesion bone metastases.^20–23 Spinal metastases have been reported to account for approximately 80% of cancers of unknown primary.^20,24 Therefore, it is very important to determine the distribution properties of bone metastases. I believe that the data in this study will be beneficial in evaluating cancers of unknown primary that have metastasized to the spine.

Although authors have investigated spinal metastases in the literature, no authors have investigated isolated spinal metastases without visceral and additional bone metastases. Authors have shown that the thoracic spine is the most common metastasis localization as well as that the lumbar spine is the most common area of metastasis.^25–27 The cervical spine is the area with the least metastasis.²⁸ In the present study, the thoracic + lumbal spine was shown to be the most common area of metastasis. In accordance with the literature, the cervical spine is the place where metastasis was seen least. This study reveals the metastasis spread rates of each spine in terms of primary cancer, together with the main spinal sections, unlike existing studies in the literature.

Bone metastases of cancers can occur through systemic blood circulation or vertebral venous plexus.^7–9 Metastases that spread using the blood circulatory system can spread to any bone, especially to extremities, and this is considered to occur randomly.^16–18 Diffusion from the venous system is the main process of spinal metastasis, and it rarely reaches the extremities by using the venous system.^16–18 Therefore, patients with extremity, visceral, or metastases to both were excluded from the present study to examine spinal metastases that occurred specifically through the vertebral venous system. In this way, I thought the spinal metastasis characteristics of the vertebral venous system could be better understood. Few studies are available in which authors investigate metastases through the vertebral venous plexus.^18,19 In the study of Zhu et al,¹⁸ bone metastases in the whole body without visceral metastasis were investigated, and they stated that the primary cancer tends to metastasize to the bone close to it.¹⁸ Although similar results are available in this study, the main difference between this study and the study of Zhu et al¹⁸ is the detailed evaluation of isolated spinal metastases. Wang et al¹⁹ investigated bone metastases of lung and prostate cancers in their study.¹⁹ Authors of publications in the literature have examined bone metastases for a specific type of cancer, such as the study of Wang et al.^19,29–32 This study evaluates spinal metastasis features by vertebral venous plexus in detail by examining cancers in all body regions.

This study has some limitations. First, the most important limitation is that the majority of the patients in the study had respiratory system and breast cancers. It should be taken into consideration that both cancers are the most common cancers in the society in this case. However, studies with a higher number of patients are needed for more objective results.

CONCLUSIONS

Although the vertebral venous system is an important pathway for spinal metastasis formation, its characteristic features have not been revealed clearly yet. As a result of this study, primary cancers without visceral metastasis tend to metastasize to the near spine. Comprehensive studies involving more patients are needed to reveal the mechanism of metastasis formation with the vertebral venous system more clearly. The results obtained by detailed examination of spinal metastases may provide effective and useful clues in finding cancers of unknown primary.