Volume 112, Issue 1 p. 157-161
Original Article
Free Access

The first score predicting overall survival in patients with metastatic spinal cord compression

Dirk Rades MD

Corresponding Author

Dirk Rades MD

Department of Radiation Oncology, University Hospital Schleswig-Holstein, Luebeck, Germany

Fax: (011) 49-451-500-3324.

Department of Radiation Oncology, University Hospital Schleswig-Holstein, Campus Luebeck, Ratzeburger Allee 160, D-23538 Luebeck, Germany===Search for more papers by this author
Juergen Dunst MD

Juergen Dunst MD

Department of Radiation Oncology, University Hospital Schleswig-Holstein, Luebeck, Germany

Search for more papers by this author
Steven E. Schild MD

Steven E. Schild MD

Department of Radiation Oncology, Mayo Clinic Scottsdale, Scottsdale, Arizona

Search for more papers by this author
First published: 19 October 2007
Citations: 116

Abstract

BACKGROUND.

The current study was performed to create a scoring system to estimate the survival of patients with metastatic spinal cord compression (MSCC).

METHODS.

Based on a multivariate survival analysis of 1852 patients who were treated with radiotherapy (RT) for MSCC, a scoring system was developed. The system included the 6 prognostic factors found to be significant in that multivariate analysis: tumor type, interval between tumor diagnosis and MSCC, other bone or visceral metastases at the time of RT, ambulatory status, and duration of motor deficits. The score for each prognostic factor was determined by dividing the 6-month survival rate (given as the percentage) by 10. The total score represented the sum of the scores for each prognostic factor. Total scores ranged between 20 and 45 points and patients were divided into 5 groups. For each group, the survival was compared for short-course or long-course RT.

RESULTS.

The 6-month survival rates were 4% for patients with a score of 20 to 25 points,11% for patients with a score of 26 to 30 points, 48% for patients with a score of 31 to 35 points, 87% for patients with a score of 36 to 40 points, and 99% for patients with a score of 41 to 45 points (P < .001). Patients with scores ≥36 were found to have a significantly longer survival when comparing long-course versus short-course RT, and those with scores <36 were found to have similar survival regardless of whether short-course or long-course RT was used.

CONCLUSIONS.

Using the scoring system described herein, patients with MSCC can be grouped to estimate survival. Patients with scores ≥36 were found to survive longer with long-course RT, whereas patients with lower scores had a similar survival regardless of whether long-course or short-course RT was used. Cancer 2008. © 2007 American Cancer Society.

Radiotherapy (RT) is the most frequently applied treatment modality in patients with metastatic spinal cord compression (MSCC). The majority of MSCC patients have a life span of only a few months and will not live long enough to develop disease recurrence or progression of MSCC in the irradiated spinal region.1, 2 For these patients, short-course RT with 1 to 5 fractions given over 1 to 5 days (eg, 1 fraction of 8 grays [Gy], or 5 fractions of 4 Gy each) appears appropriate because it is associated with less patient discomfort and time spent in treatment. However, a considerable proportion of MSCC patients survive long enough (≥6 months) to develop disease recurrence (if a complete response was achieved) or disease progression (if a partial response was achieved) of MSCC in the irradiated spinal region.3 Because reirradiation to the spine may be associated with an increased risk of radiation myelopathy, many radiation oncologists are concerned about delivering a second course of RT to the same region of the spine. According to our previous report, long-course RT is believed to be associated with better local control of MSCC.3 The 1-year local control rates were found to be 90% with long-course RT and 74% with short-course RT (risk ratio [RR] of 0.54; 95% confidence interval [95% CI],0.45–0.65 [P < .001]). Thus, for those patients with a relatively good survival prognosis, long-course RT with 10 to 20 fractions given over 2 to 4 weeks (eg, 10 fractions of 3 Gy each or 20 fractions of 2 Gy each) appears to be the better option.

However, until now it was difficult to predict the survival of MSCC patients due to a lack of appropriate instruments. The current study presents a scoring system that allows for the adequate prediction of the probability of a patient surviving ≥6 months after RT. This score can be considered a simple tool that can help the clinician select the appropriate treatment for the individual patient with MSCC.

MATERIALS AND METHODS

The score is based on the retrospective analysis of 1852 patients treated with RT for MSCC between January 1992 and October 2005.3 In that analysis, 11 potential prognostic factors were investigated with respect to survival after RT on both univariate and multivariate analysis. In the multivariate analysis, which was performed using the Cox proportional hazards model, 6 prognostic factors were found to be significant with regard to overall survival. These 6 factors were considered for the survival score presented in the current study.

The 6 significant factors were as follows: type of primary tumor (breast cancer vs prostate cancer vs myeloma/lymphoma vs lung cancer vs other tumors: RR of 1.23; 95% CI, 1.18–1.29 [P < .001]), interval between tumor diagnosis and MSCC (≤15 months vs >15 months: RR of 0.80; 95% CI, 0.75–0.85 [P < .001]), presence of other bone metastases at the time of RT (no vs yes: RR of 1.29; 95% CI, 1.04–1.59 [P = .018]), visceral metastases at the time of RT (no vs yes: RR of 4.89; 95% CI, 4.26–5.60 [P < .001]), pretreatment ambulatory status (nonambulatory vs ambulatory: RR of 0.47; 95% CI 0.39–0.56 [P < .001]), and time to develop motor deficits before RT (faster [1–14 days] vs slower [>14 days]: RR of 0.72; 95% CI, 0.68–0.77 [P < .001]). The Eastern Cooperative Oncology Group (ECOG) performance status was not included in the scoring system because there is a strong correlation reported between ambulatory status and ECOG performance status (confounding variables). Ambulatory status is associated with an ECOG performance status of 1 to 2 and nonambulatory status is associated with an ECOG performance status of 3 to 4.

A separate score was calculated for each of the 6 significant prognostic factors (ie, the 6-month survival rate [given as the percentage] according to the Kaplan-Meier analysis of the previous report3 was divided by 10). The 6-month survival rates and the separate scores for each of the 6 prognostic factors are summarized in Table 1. For the scoring system presented herein, 3 groups with regard to the time to the development of motor deficits before RT were formed (1–7 days, 8–14 days, and >14 days) instead of 2 groups (1–14 days and >14 days) because previous reports suggested that a time to the development of motor deficits of 8 to 14 days was associated with significantly better survival than a time of 1 to 7 days.4, 5 The total score that was included in the scoring system represents the sum of the scores from all 6 prognostic factors.

Table 1. Actuarial Overall Survival Rates 6 Months After Radiotherapy and the Corresponding Score
Survival at 6 months, % Score
Type of primary tumor
 Breast cancer 78 8
 Prostate cancer 66 7
 Myeloma/lymphoma 85 9
 Lung cancer 25 3
 Other tumors 40 4
Other bone metastases at the time of RT
 Yes 48 5
 No 65 7
Visceral metastases at the time of RT
 Yes 17 2
 No 80 8
Interval from tumor diagnosis to MSCC
 ≤15 mo 41 4
 >15 mo 71 7
Ambulatory status before RT
 Ambulatory 71 7
 Nonambulatory 31 3
Time to develop motor deficits before RT
 1–7 d 26 3
 8–14 d 55 6
 >14 d 78 8
  • RT indicates radiotherapy; MSCC, metastatic spinal cord compression.

For example, a breast cancer patient (with a score of 8 points) with the absence of both visceral metastases (score of 8 points) or other bone metastases (score of 7 points) developed motor deficits within 5 days before RT (score of 3 points) and was not ambulatory before RT (score of 3 points). The interval between the first diagnosis of breast cancer and the onset of MSCC was 24 months (score of 7 points). Thus, the total score is 36 points (8 + 8 + 7 + 3 + 3 + 7 = 36 points), which is associated with a 6-month survival probability after RT of 78%.

The total scores ranged from 20 to 45 points (Fig. 1). Five groups were formed according to the total score based on the 6-month survival rates for each score as presented in Figure 1: 20 to 25 points (214 patients, Group A), 26 to 30 points (387 patients, Group B), 31 to 35 points (401 patients, Group C), 36 to 40 points (402 patients, Group D), and 41 to 45 points (448 patients, Group E). The 5 groups were compared for survival using the Kaplan-Meier method.6 The Kaplan-Meier curves were compared using the log-rank test. The difference was considered significant with P < .05.

Details are in the caption following the image

The total score, given as the percentage (which ranged between 20–45 points), in relation to the survival rate at 6 months after radiotherapy.

RESULTS

The 6-month survival rates were 4% for patients with a total score of 20 to 25 points, 11% for patients with a score of 26 to 30 points, 48% for patients with a score of 31 to 35 points, 87% for patients with a score of 36 to 40 points, and 99% for patients with a score of 41 to 45 points. The 1-year survival rates for the 5 groups were 0%, 6%, 23%, 70%, and 89%, respectively (Fig. 2). The difference between the 5 groups was statistically significant (P < .001). Subgroup analyses revealed that the differences between Groups A and B (P < .001), Groups B and C (P < .001), Groups C and D (P < .001), and Groups D and E (P < .001) were statistically significant. This finding supported the appropriate nature of the selection criteria used for the 5 groups. A comparison of patient survival for those receiving short-course RT compared with patients receiving long-course RT was performed for each of the 5 groups. This analysis revealed no statistically significant difference in the survival of the patients in Groups A (P = .90), B (P = .47), and C (P = .73), whereas survival was significantly better after long-course RT compared with short-course RT in Groups D (P = .011) and E (P = .002). These results are shown in Figure 3.

Details are in the caption following the image

Kaplan-Meier survival curves for the 5 groups Group A: 21 to 25 points; Group B: 26 to 30 points; Group C: 31 to 35 points; Group D: 36 to 40 points; and Group E: 41 to 45 points.

Details are in the caption following the image

Comparison of short-course radiotherapy (SC-RT) and long-course radiotherapy (LC-RT) with respect to survival in Groups A, B, C, D, and E.

DISCUSSION

Short-course RT with 1 to 5 fractions given over 1 to 5 days is considered appropriate for MSCC patients with a survival of <6 months. Short-course RT is associated with less patient discomfort than long-course RT (10–20 fractions given over 2–4 weeks) because of fewer trips to the radiation oncology department and fewer, mostly painful, positionings on the treatment couch. Furthermore, a shorter RT regimen uses far less of a patients' limited lifespan for daily treatments and reduces the cost of therapy.7

However, a considerable proportion of patients with MSCC survive >6 months and it is these patients who are at greater risk of developing a recurrence or progression of MSCC in the previously irradiated spinal region. Long-course RT has been reported to be associated with fewer instances of disease recurrence or progression than short-course RT.3 Disease recurrence or progression occurs most frequently ≥6 months after RT. Therefore, for MSCC patients with an expected survival of ≥6 months, long-course RT may be the better option. To select the appropriate RT regimen (short-course vs long-course) for the individual MSCC patient, an instrument that allowed the prediction of survival was needed.

The current analysis was performed to create a scoring system to help physicians estimate the survival of patients with MSCC. The score is based on the prognostic factors of 1852 patients at the time of MSCC and their subsequent survival. This system can be used to determine a score for each patient that can range from 20 to 45 points and allows one to classify each patient into 1 of 5 groups with distinctly different survival rates. Patients with a score of 20 and 30 points (Groups A and B) were found to have a 6-month and a 1-year survival probability of <20% and <10%, respectively, and therefore these patients may be treated with short-course RT. Furthermore, in Groups A and B, long-course RT did not appear to improve survival compared with short-course RT (Fig. 3). In contrast, patients who achieved a score of 36 and 45 points (Groups D and E) were found to have 6-month and 1-year survival probabilities of 78% to 100% and 70% to 89%, respectively, and therefore should be treated with long-course RT. Furthermore, long-course RT was associated with better survival than short-course RT in both Group D and Group E patients. A score of 31 to 35 points (Group C) may be considered the “gray zone.” However, the 1-year survival probability of these patients is <25%. Thus, they may be expected to be adequately treated with short-course RT. This recommendation is supported by the finding that there was no significant survival advantage observed among Group C patients with long-course RT (Fig. 3). In patients with a comparably good survival prognosis (a score of 36 to 45 points), local control of MSCC may be further improved with the administration of bisphosphonates such as zoledronic acid, which have been demonstrated to reduce complications due to bone metastases.8, 9

This scoring system is based on what to our knowledge is the largest series of MSCC patients published to date, which makes it quite reliable.8 Conversely, the score is based on data from a retrospective study.3 Therefore, hidden biases may have been introduced in that analysis. However, one has to be aware that an appropriately large prospective series of MSCC patients with which to develop a survival score is not currently available.

In the previously reported study, better survival was found to be significantly associated with the absence of visceral metastases or other bone metastases at the time of RT, ambulatory status before RT, a longer interval (>15 months) between tumor diagnosis and MSCC, a slower development of motor deficits before RT (>14 days), and favorable primary tumor type (myeloma/lymphoma, breast cancer, prostate cancer). The impact of visceral metastases or other bone metastases, pre-RT ambulatory status, and the type of primary tumor on survival has been previously described.1, 10-12 The negative prognostic impact of both a shorter interval between tumor diagnosis and the development of MSCC and the faster development of motor deficits before RT most likely reflects the faster growth of more aggressive tumors.

In conclusion, patients who achieve a score of 20 to 30 points appear to be well-treated with short-course RT because their 6-month and 1-year survival rates are very low. A score of 31 to 35 points may be considered the “gray zone,” and short-course RT may be considered in these patients. If the score is between 36 and 45 points, long-course RT appears to be more appropriate because the majority of these patients may live long enough to develop a recurrence or progression of MSCC.