Abstract
Background/Aim: According to our randomized trial, 5×4Gy was comparable to 10×3Gy for metastatic spinal cord compression. Since it remained unclear whether findings applied to poor and intermediate prognoses patients, subgroup analyses were performed. Patients and Methods: In patients with poor prognoses, 58 received 5×4Gy, 53 received 10×3Gy. In intermediate-prognoses patients, numbers were 43 and 49. Results: In patients with poor prognoses, 1-month overall response (OR) was 85% after 5×4Gy and 10×3Gy (p=0.99), improvement 38% vs. 42%, ambulatory status 60% vs. 64% (p=0.83), 6-month local progression-free survival (LPFS) 75% vs. 69% (p=0.74) and 6-month overall survival (OS) 26% vs. 19% (p=0.43). In patients with intermediate prognoses, 1-month OR was 89% after 5×4Gy and 93% after 10×3Gy (p=0.85), improvement 39% vs. 45%, ambulatory status 84% vs. 82% (p=0.90), 6-month LPFS 79% vs. 92% (p=0.17) and 6-months OS 65% vs. 58% (p=0.65). Conclusion: 5×4Gy was not significantly inferior to 10x3Gy in both subgroups.
- Metastatic spinal cord compression
- radiotherapy
- randomized trial
- survival prognosis
- subgroup analyses
- treatment outcomes
Metastatic spinal cord compression (MSCC) is an oncologic emergency that affects 5-10% of adult cancer patients during their disease course (1, 2). Radiotherapy alone is the most frequently used treatment, although upfront decompressive surgery became more popular after a randomized trial published in 2005 had shown that highly selected patients benefit from the addition of surgery in terms of a higher ambulation and survival rates (3). Patients irradiated for MSCC often have a median survival of only 2 to 6 months (1, 2). Radiotherapy sessions can be burdensome for the patients who are generally impaired by vertebral pain and neurologic deficits. Thus, the number of fractions should be kept as low as possible while producing positive outcomes. During the last three years, three randomized trials became available that compared two different radiation regimens for MSCC. At this year's annual ASCO meeting, the most recent one, the SCORAD III trial, was presented with 688 patients from the United Kingdom and Australia (4). This trial compared 1×8Gy to short-course multi-fraction radiotherapy with 5×4Gy over 1 week in patients with a poor survival prognosis, i.e. a median survival time of only 13 weeks. According to the SCORAD III trial, 1×8Gy was not inferior to 5×4Gy with respect to the ability to walk at 8 weeks following randomization. In another trial (ICORG 05-03) published in abstract form in 2014, 1×10Gy was not inferior to 5×4Gy for overall response (OR), defined as improvement or no further progression of motor deficits, and for ambulation at 5 weeks (5). The third randomized trial (SCORE-2) was published as a full paper in January 2016 (6). It compared the two most common radiotherapy regimens, 5×4Gy over 1 week and 10×3Gy over 2 weeks in patients with a poor or intermediate survival prognosis. According to the results of the SCORE-2 trial, 5×4Gy was not inferior to 10×3Gy regarding OR, effect on motor function, post-treatment ambulatory status, local progression-free survival (LPFS) and overall survival (OS). The patients' prognosis was estimated with a validated scoring tool (7). In the cohort used to create this tool, median survival times were 2.5 months in the poor-prognosis group and 5.5 months in the intermediate-prognosis group. Looking at this difference, it may be possible that the non-inferiority of 5×4Gy in the SCORE-2 trial is only limited to patients in either the intermediate- or the poor-prognosis group. This hypothesis is supported by a previous prospective study showing that short-course regimens including 5×4Gy are associated with more in-field recurrences of MSCC than longer-course regimens including 10×3Gy (8). The risk of developing such a recurrence increases with lifetime. Thus, it is reasonable to perform separate subgroup analyses in the cohort of the SCORE-2 trial for patients with poor estimated survival and those with intermediate prognoses.
Characteristics of the group of patients with a poor survival prognosis.
Characteristics of the group of patients with an intermediate survival prognosis.
Patients with poor survival prognoses: Comparison of 5×4Gy and 10×3Gy for the effect on motor function (improvement, no further progression, deterioration) directly (A), at 1 month (B), at 3 months (C), and at 6 months (D) after radiotherapy.
Patients and Methods
The SCORE-2 trial compared 5×4Gy over 1 week to 10×3Gy over 2 weeks in patients with MSCC and a poor or intermediate survival prognosis (6) with respect to OR defined as improvement or at least no further progression of motor deficits, effect of radiotherapy on motor deficits, post-treatment ambulatory status, LPFS and OS. The SCORE-2 trial was approved by local ethic committees (leading committee: University of Lübeck), the expert committee of the German Society for Radiation Oncology and the German Cancer Society. It was performed in accord with an assurance filed with and in accordance with the precepts established by the Helsinki Declaration. Criteria for inclusion and exclusion, diagnostic procedures, radiation techniques and radiation parameters have been previously described (6). After written informed consent, 203 patients irradiated for MSCC between 07/2010 and 05/2015, were stratified for pre-treatment ambulatory status, time developing motor deficits prior to radiotherapy, and primary tumor type, and randomly assigned to 5×4Gy or 10×3Gy.
Motor function was graded with an 8-point scale (9): 0, complete paraplegia; 1, palpable/visible muscle contractions; 2, active movement of the leg without gravity; 3, against gravity; 4, against mild resistance; 5, against intermediate resistance; 6, against strong resistance; and 7, normal strength. Motor function was recorded for each leg resulting in 0 to 14 points. Improvement or deterioration of motor function was defined as a change of two or more points.
In the present study, additional separate analyses were performed for patients with a poor survival prognosis (n=111) and those patients with an intermediate survival prognosis (n=92). The survival prognosis of each patient was estimated with a validated instrument developed specifically for patients irradiated for MSCC (7). Patient characteristics of both cohorts are summarized in Table I (poor prognosis group) and Table II (intermediate prognosis group).
Statistical considerations. The primary endpoint was OR with respect to motor deficits at 1 month following radiotherapy. In addition, the effect of radiotherapy on motor function (improvement, no further progression, deterioration) was evaluated with the ordered logit model, and post-radiotherapy ambulatory rates were compared with the Chi-square test. OR, effect of radiotherapy on motor function and post-treatment ambulatory rates were evaluated directly and at 1, 3 and 6 months following radiotherapy. p-Values of less than 0.05 were considered significant. LPFS and OS were referenced from the last day of radiotherapy. LPFS was defined as no deterioration of motor deficits during or directly after radiotherapy and no in-field recurrence of MSCC during follow up. The analyses of LPFS and OS were performed with the Kaplan-Meier-method and the log-rank test (10). Again, p-values of less than 0.05 were regarded significant.
Results
Patients with poor survival prognoses. The OR rates after 5×4Gy and 10×3Gy were 82% (45/55 patients) vs. 79% (37/47 patients) directly after radiotherapy (p=0.86), 85% (34/40 patients) vs. 85% (28/33 patients) at 1 month following radiotherapy (p=0.99), 100% (18/18 patients) vs. 87% (14/16 patients) at 3 months following radiotherapy (p=0.70), and 100% (12/12 patients) vs. 89% (8/9 patients) at 6 months following radiotherapy (p=0.79). The rates of improvement, no further progression and deterioration of motor deficits assessed directly and 1 month, 3 months and 6 months after radiotherapy are given in Figure 1. The post-treatment ambulatory rates after 5×4Gy and 10×3Gy were 53% (29/55 patients) vs. 51% (24/47 patients) directly after radiotherapy (p=0.91), 60% (24/40 patients) vs. 64% (21/33 patients) at 1 month following radiotherapy (p=0.83), 72% (13/18 patients) vs. 56% (9/16 patients) at 3 months following radiotherapy (p=0.55), and 75% (9/12 patients) vs. 78% (7/9 patients) at 6 months following radiotherapy (p=0.96).
The LPFS rates at 3 and 6 months were 75% vs. 75%, respectively, after 5×4Gy and 77% vs. 69%, respectively, after 10×3Gy (p=0.74). Median OS times were 2 months and 2 months, respectively. The OS rates at 3 and 6 months were 31% vs. 26%, respectively, after 5×4Gy and 30% vs. 19% after 10×3Gy (p=0.43). The corresponding Kaplan-Meier curves for both LPFS and OS are given in Figure 2.
Patients with intermediate survival prognoses. OR rates after 5×4Gy and 10×3Gy were 93% (38/41 patients) vs. 96% (47/49 patients) directly after radiotherapy (p=0.88), 89% (34/38 patients) vs. 93% (41/44 patients) at 1 month following radiotherapy (p=0.85), 97% (28/29 patients) vs. 97% (28/29 patients) at 3 months following radiotherapy (p=1.0), and 95% (20/21 patients) vs. 100% (21/21 patients) at 6 months following radiotherapy (p=0.88). The rates of improvement, no further progression and deterioration of motor deficits are shown in Figure 3. Post-treatment ambulatory rates after 5×4Gy and 10×3Gy were 78% (32/41 patients) vs. 78% (38/49 patients) directly after radiotherapy (p=0.98), 84% (32/38 patients) vs. 82% (36/44 patients) at 1 month (p=0.90), 86% (25/29 patients) vs. 83% (24/29 patients) at 3 months (p=0.89), and 86% (18/21 patients) vs. 86% (18/21 patients) at 6 months following radiotherapy (p=1.0).
LPFS rates at 3 and 6 months were 82% vs. 79%, respectively, after 5×4Gy and 92% vs. 92%, respectively, after 10×3Gy (p=0.17). Median OS times were >6 months and >6 months, respectively. OS rates at 3 and 6 months were 74% vs. 65%, respectively, after 5×4Gy and 77% vs. 58%, respectively, after 10×3Gy (p=0.65). The corresponding Kaplan-Meier curves for both LPFS and OS are given in Figure 4.
Discussion
Many patients with MSCC present in a reduced performance status and are not fit enough to withstand decompressive surgery or have a very radiosensitive primary tumor not requiring upfront surgery (3, 11-15). Therefore, the majority of patients with MSCC receive radiotherapy plus/minus dexamethasone. If patients are assigned to radiotherapy, several fractionation regimens are available including single-fraction regimens, short-course multi-fraction regimens lasting about 1 week and longer-course regimens lasting 2-4 weeks (1, 2). The optimal individual regimen should take into account the patient's survival prognosis. According to a scoring system developed to predict the probability to survive at least 6 months following radiotherapy for MSCC, three major prognostic groups (20-30 points, 31-35 points and 36-45 points) were identified with 6-month survival rates of 9%, 48% and 93%, respectively (p<0.0001) (7, 16). LPFS is another important endpoint, which also includes freedom from an in-field recurrence of MSCC in the irradiated part of the spine. In-field recurrences are often difficult to treat, because spinal surgery may not be possible and a second course of radiotherapy may exceed the tolerance dose of the spinal cord and lead to radiation myelopathy associated with severe neurologic deficits (17-19). Therefore, in-field recurrences are best avoided. In 2005, a large retrospective study suggested that in-field recurrences occur significantly more often after single-fraction (1×8Gy) or short-course multi-fraction (5×4Gy over 1 week) radiotherapy than after longer-course radiotherapy including 10×3Gy over 2 weeks (p<0.001) (20). These results were confirmed in the prospective SCORE-1 study, which compared two unselected series of patients irradiated for MSCC (8). Patients treated in the Netherlands received 1×8Gy or 5×4Gy (n=131) and patients treated in Germany longer-course radiotherapy with 10×3Gy, 15×2.5Gy or 20×2Gy (n=134). One-year local control rates were 81% after longer-course irradiation and 61% after 1×8Gy/5×4Gy, respectively (p=0.005). This difference was also significant on multivariate analysis (p=0.018). Since the risk of in-field recurrences increases with lifetime, patients with a more favorable survival prognosis should receive longer-course radiotherapy. Moreover, in a matched-pair study of 382 patients with MSCC, 15×2.5Gy and 20×2Gy resulted in significantly better local control (92% vs. 71%, p=0.012) and survival (68% vs. 53%, p=0.032) at 2 years than 10x3Gy in patients with a very favorable survival prognosis (score of 36-45 points) (21). The results regarding local control and survival were also significant on multivariate analyses. When interpreting the data of this matched-pair study, one has to be aware that the data were retrospectively collected. However, patients of both groups were matched 1:1 for ten important potential prognostic factors. The results of these studies suggest that patients with very favorable survival prognoses should not be irradiated with 1×8Gy or 5×4Gy.
Patients with poor survival prognoses: Comparison of 5×4Gy and 10×3Gy for local progression-free survival (top) and survival (bottom).
Patients with intermediate survival prognoses: Comparison of 5×4Gy and 10×3Gy for the effect on motor function (improvement, no further progression, deterioration) directly (A), at 1 month (B), at 3 months (C), and at 6 months (D) after radiotherapy.
However, the situation would be different for patients with a poor or intermediate survival prognosis. For these patients, local control is less important than in those with very favorable prognoses, since many of these patients do not live long enough to experience an in-field recurrence. A radiotherapy regimen with less fractions and a shorter overall treatment time would result in the patients spending less time in treatment and more time with relatives and friends. Furthermore, for patients suffering from pain and neurologic deficits, a radiation session may cause discomfort. Short-course regimens could only be recommended if they were similarly effective as longer-course regimens. A few randomized trials compared fractionation regimens to evaluate if a reduction of number of fractions and treatment time is feasible and efficacious.
In 2005, a phase III trial from Italy compared 2×8Gy with a gap of 6 days between the two fractions to a split-course regimen of 3×5Gy followed by 4 days without treatment and 5×3Gy (overall treatment time of 2 weeks) in 276 patients with MSCC and poor survival prognoses (22). Following radiotherapy, 68% and 71% of the patients, respectively, were able to walk. Median duration of the improvement of motor function was 3.5 months in both groups. In another Italian phase III trial, 2 fractions of 8Gy given on days 1 and 7 were compared to a single fraction of 8Gy in 303 patients with MSCC and a short expected survival time (23). Response rates were similar in both groups, and median duration times of response were 5 months after 2×8Gy and 4.5 months after 1×8Gy, respectively (p=0.4). Since the split-course regimen and 2×8Gy with 4 days rest are very rarely used outside Italy, the results of both trials cannot be generalized to other countries. The third randomized trial (ICORG 05-03), a non-inferiority trial from the Republic of Ireland compared 1×10Gy to 5×4Gy over 1 week 115 eligible patients with MSCC and a poor expected survival (5). Improvement/stability rates of mobility at 5 weeks were 78.9% after 1×10Gy and 68.4% after 5×4Gy, respectively (difference not significant). Mobility deterioration-free survival times were 1.4 months in both groups. The most recent phase III trial (SCORAD III) compared 1x8Gy to 5×4Gy over 1 week in 688 Patients with MSCC (4). Median survival in the entire cohort was only about 3 months. Primary endpoint was ambulatory status at 8 weeks following randomization; the non-inferiority margin (1×8Gy vs. 5×4Gy) was 11%. Of those patients evaluable at 8 weeks, 69.5% (114 of 164 patients) after 1×8Gy and 73.3% (129 of 176 patients) were ambulatory with or without aid.
Our own phase III trial (SCORE-2) compared the most common regimens for MSCC, 5×4Gy over 1 week and 10×3Gy over 2 weeks, in 203 patients (6). In contrast to the other four trials (4, 5, 22, 23), the SCORE-2 trial included both patients with poor and intermediate survival prognoses. According to its results, 5x4Gy was not significantly inferior to 10x3Gy with respect to OR, effect of radiotherapy on motor function and post-treatment ambulatory rates directly and at 1, 3 and 6 months after radiotherapy. Six-month LPFS (75.2% vs. 81.8%, p=0.51) and 6-month survival (42.3% vs. 37.8%, p=0.68) were also not significantly different. Two previous studies suggested that longer-course regimens such as 10×3Gy lead to a better local control of MSCC or LPFS than 5×4Gy (8, 20). This benefit appears to be of most value to patients with a better survival prognosis, since the risk of an in-field recurrence of MSCC increases with time after radiotherapy and a patient would have to live long enough to be at risk. In a previous study, median survival times for patients with poor and intermediate survival prognoses were 2.5 and 5.5 months, respectively (7). It may well be that the non-inferiority of 5×4Gy to 10×3Gy in the SCORE-2 trial regarding LPFS was caused by the patients with poor survival prognoses and may not apply to those with intermediate prognoses. Therefore, the present study was initiated, and these subgroup analyses of the SCORE-2 cohort were performed. The data revealed the non-inferiority of 5×4Gy to 10×3Gy regarding all investigated endpoints that was irrespective of the patients' survival prognoses (poor or intermediate). Thus, patients with intermediate prognoses (31-35 points on the survival score) may be treated with 5×4Gy instead of 10×3Gy, which results in fewer hospital visits (7). For patients with poor estimated survival (20-30 points), 1×8Gy or 1×10Gy is a reasonable option, as two randomized trials demonstrated that single-fraction radiotherapy was not inferior to 5x4Gy in these patients (4, 5). In contrast, patients with a very favorable survival prognosis (36-45 points) appear better treated with longer-course radiotherapy including total doses greater than 30Gy (21).
Patients with intermediate survival prognoses: Comparison of 5×4Gy and 10×3Gy for local progression-free survival (top) and survival (bottom).
In summary, according to subgroup analyses, 5×4Gy was not inferior to 10×3Gy for MSCC in both poor- and intermediate-risk patients. Thus, 5×4Gy appears appropriate for patients with intermediate prognoses. Patients with poor expected survival may be treated with 1×8Gy or 1×10Gy, since randomized trials demonstrated these regimens are not inferior to 5×4Gy in poor-risk patients.
Footnotes
Conflicts of Interest
On behalf of all Authors, the corresponding Author states that there is no conflict of interest related to this study.
- Received November 13, 2017.
- Revision received November 27, 2017.
- Accepted November 28, 2017.
- Copyright© 2018, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
