Personalization of cancer treatment, which has become more important in oncology during recent years, must take into account the patient’s life expectancy. This accounts in particular for a palliative situation such as MSCC. Survival scores help estimate the survival prognosis of each patient. Several scores created to estimate the survival of patients with bone metastases already exist. A few prognostic scores have been developed in particular for patients with bone metastases in the vertebral column.
The majority of these scores were designed by to help surgeons decide whether spinal surgery may be indicated or not. In 1990, Tokuhashi et al. presented a score based on the data of 64 patients with a metastatic spine tumor who underwent spinal surgery [5
]. Their score has been revised 15
years later in a series of 246 patients [6
]. Bauer et al. reported a scoring system including scoring for pathological fracture based on the data of 88 patients with spinal metastasis plus 153 patients with bone metastasis of the extremities in 1995 [7
]. Leithner et al., who compared different scoring systems in their series of 69 patients in 2008, suggested a modified Bauer score without scoring for pathological fracture [8
]. The Tomita score presented in 2001 included the data of 67 patients [9
]. Except the revised Tokuhashi score [6
], these scoring systems may have a limited validity due to the relatively small number of patients included. Furthermore, these scores were designed for patients with spinal metastasis in general, and not particularly for patients with motor deficits due to MSCC.
In 2005, Van der Linden et al. presented a score in a larger series of patients (n
342) with painful spinal metastases who had received radiotherapy alone and no surgery [10
]. In that study, patients with neurologic impairment were not included. All these previous scoring systems included patients with spinal metastasis from many different primary tumors. However, because various primary tumor types behave differently, it is important to have separate scores for the different tumor entities, in particular for the most common ones such as breast cancer, prostate cancer, and NSCLC [1
In the present study, four independent prognostic factors were found to be significantly associated with survival in patients with MSCC from NSCLC in a comparably large series of patients. These significant factors included the ECOG-PS, pre-radiotherapy ambulatory status, visceral metastases, and the time developing motor deficits. In our previous report on prognostic factors for different outcomes in the entire cohort of 356 patients, gender, other bone metastases, and the interval from the first diagnosis of NSCLC to radiotherapy of MSCC were also significantly associated with survival [11
]. However, we included only those prognostic factors found to be independent in the multivariate analysis of the test group in the present score, because we felt that this would make the score more robust.
When compared to MSCC from other solid tumors, patients with MSCC from NSCLC have a less favorable estimated survival [1
]. This is reflected by the fact that the worst prognostic group, group A, was the largest group in the present study. Based on the 6-months survival times related to the four independent prognostic factors, three prognostic groups were formed. Group A patients had the worst prognosis, only 6% of patients in the test group and 4% in the validation group survived at least 6
months following irradiation. These patients may be considered candidates for single-fraction radiotherapy or even best supportive care alone. Group B patients had 6-months survival rates of 29% and 24%, respectively and may be treated with short-course multi-fraction radiotherapy such as 20
Gy in 5 fractions over one week. Short-course radiotherapy is as effective as longer programs with respect to post-radiotherapy motor function [12
]. In contrast, local control of MSCC is better with longer-course than with short-course radiotherapy [3
]. However, local control of MSCC appears of minor importance in group B patients, because most of these patients will not live long enough to experience a local recurrence of MSCC. In contrast, group C patients who achieved 6-months survival rates of 78% and 76%, respectively, are at a higher risk of developing a local recurrence of MSCC and, therefore, are likely to benefit from longer-course radiotherapy such as 10x3 Gy in 2
weeks or 20x2 Gy in 4
weeks. In group B and group C patients, upfront decompressive surgery in addition to radiotherapy may be reasonable for selected patients with a good performance status and involvement of only one spinal segment. This accounts in particular for patients who are unlikely to be able to walk after radiotherapy alone. In a randomized trial of 101 patients, decompressive surgery followed by radiotherapy led to better pre-radiotherapy ambulatory function and survival than radiotherapy alone in such patients [2
The present score focused on a single tumor entity. In contrast, previous prognostic indices for patients with vertebral metastases or other palliative situations included many different tumor types [5
]. Therefore, the present scoring system takes more into account the patient’s individual situation. In order to validate our score, the risk groups A, B and C of the test group were compared to the corresponding groups A, B and C of the validation group. The 6-months survival rates of the three groups in the validation group proved to be similar to the corresponding 6-months survival rates in the test group. Thus, this new score for MSCC from NSCLC appears valid and reproducible. However, the score is based on retrospective data. Furthermore, data on systemic treatment following treatment was not available in most patients. These two aspects may have led to a hidden selection bias. Therefore, the results of the present study need to be confirmed in a prospective series of patients.