As retrospective study reported, a total of 539 patients diagnosed with BM in the University of Tokyo Hospital were categorized into four groups according to bone metastatic sites [29]. Patients with metastasis to humerus or femur were defined as ‘rhizometastasis’. The study concluded that the patients with rhizometastasis accounted for 22.5% with the median survival of 16.0 months[29]. Another study included 164 patients with limb metastasis and the top five common primary tumor lesions were, in descending order, breast, lung, renal, prostate and myeloma [30]. Being different from the above-mentioned studies with single center data, we performed an international investigation and reported 316 patients derived from five cohorts in China and Russia. The baseline characteristics of patients were summarized, and survival outcome and limb function were evaluated. Breast cancer, renal cancer and lung cancer were proved to be the most common primary tumors while different survival outcomes were confirmed between subgroups.
As for patients with spinal metastasis, primary tumor was identified as the fundamental prognostic predictor in several well-established score models including Tomita score, Tokuhashi score and Linden score[28]. Similarly, multiple studies confirmed the impact of primary tumor on prognosis in patients with limbs metastasis[16, 19]. A study, including a total of 301 limbs metastatic patients, concluded that breast or prostate was associated with a better survival outcome compared with other primary lesions[16]. The postoperative survival was up to 16.0 months and 17.0 months in breast cancer and prostate cancer, respectively. The survival ranged from 4.0 to 9.0 months in more aggressive tumors such as lung cancer, bladder cancer and renal cell carcinoma[16]. Based on the analysis of 102 patients with upper extremity metastasis, another study demonstrated that there was a 4.4-fold increased risk of death in patients with rapid growth primary tumor [31]. It was reported that primary tumor was a risk factor for 30-day postoperative complications in patients with limb metastasis, and the complication rate was associated with 1-year mortality after surgery[32].
Compared to bone metastasis, visceral metastasis was generally accepted to be associated with overall cancer mortality in patients with advanced-stage cancer. The current medical therapies for visceral metastasis were inconsistent and limited despite of the recent therapeutic advances, and this metastatic disease was thought to be irreversible and incurable[33, 34]. Lung, liver and brain were regarded as the major target sites of visceral metastasis and patients with these metastatic organs always suffered from malignant pleural effusion, dysfunction of liver, headache and focal neurological deficits[35]. Based on SEER database, a total of 12,794 prostate cancer patients with BM were included and prognostic factors were identified[36]. The presence of lung, liver and brain metastasis were three predictive factors for worse survival outcome in bone metastatic prostate cancer [36]. There were significant differences of median survival between patients with or without visceral metastasis[36]. In the current study, visceral metastasis was retrospectively analyzed and there was a 1.48-fold increased risk of death for patients with visceral metastasis, which was consistent with previous studies [35–38]. Besides, the total cohort benefited on survival from the performance of chemotherapy with a 0.7-fold increased risk of death in our study. Except for anti-tumor role of chemotherapy, the potential cause could be better performance status, more family supports and higher economic possess in patients received systematic chemotherapy.
Musculoskeletal Tumor Society (MSTS) scoring system was widely accepted to measure limb function after surgery in oncological surgeon. The scoring system was originally developed in 1985 and subsequently adopted in 1993[25]. MSTS scoring system comprised six categories (0–5 points for each category) and higher total points indicated better limb function[25]. Three categories were fitted for both upper and lower extremities: pain, function, emotional acceptance. While supports, walking ability and gait were for the lower extremity; and hand positioning, dexterity and lifting ability were used for the upper extremity. Our study concluded that patients with multiple bone metastatic sites and pathological fracture presented lower total points, which could be on account of impaired mobility and emotional distress. In the current study, patients who underwent surgery of intramedullary nailing presented lower MSTS score than patients with prosthesis surgery. Usually, these two surgical procedures were selected according to the metastatic tumor site. Prosthesis surgery was used on patients with tumor adjacent to the joint while intramedullary nailing on patients with fracture located in the backbone[9]. Though previous studies suggested a lower complication rate and shorter hospitalization in patients received intramedullary nailing[27, 39]. Compared with intramedullary nailing, our study revealed the further functional benefit from prosthesis surgery. Thus, prosthesis surgery should be encouraged on the patients with the clear indication.
The study was performed retrospectively, thus inherent selection bias was hardly avoided. Besides, only MSTS scoring system was used to evaluate limb function. Other scoring systems such as Toronto Extremity Salvage Score (TESS) could be used in future study.