doi: 10.1111/1346-8138.12659
Journal of Dermatology 2014; 41: 1082–1086
ORIGINAL ARTICLE
Motor function and survival following radiotherapy alone for metastatic epidural spinal cord compression in melanoma patients Stefan HUTTENLOCHER,1 Lena SEHMISCH,1 Volker RUDAT,2 Dirk RADES1 1
Department of Radiation Oncology, University of L€ubeck, L€ubeck, Germany, and 2Department of Radiation Oncology, Saad Specialist Hospital, Al-Khobar, Saudi Arabia
ABSTRACT The major goal of this study was the identification of predictors for motor function and survival after irradiation alone for metastatic epidural spinal cord compression (MESCC) from melanoma. Ten variables (age, gender, performance status, number of involved vertebrae, pre-radiotherapy ambulatory status, further bone metastases, visceral metastases, interval from melanoma diagnosis to MESCC, time developing motor deficits before radiotherapy, fractionation regimen) were investigated for post-radiotherapy motor function, ambulatory status and survival in 27 patients. On multivariate analysis, motor function was significantly associated with time developing motor deficits (P = 0.006). On univariate analysis, post-radiotherapy ambulatory rates were associated with pre-radiotherapy ambulatory status (P < 0.001) and performance status (P = 0.046). Variables having a significant impact on survival in the univariate analysis were performance status (P < 0.001), number of involved vertebrae (P = 0.007), pre-radiotherapy ambulatory status (P = 0.020), further bone metastases (P = 0.023), visceral metastases (P < 0.001), and time developing motor deficits (P = 0.038). On multivariate analysis of survival, the Eastern Cooperative Oncology Group (ECOG) performance status (risk ratio [RR] = 4.35; 95% confidence interval [CI] = 1.04–16.67; P = 0.044) and visceral metastases (RR = 3.70; 95% CI = 1.10–12.50; P = 0.034) remained significant and were included in a survival score. Scoring points were obtained from 6-month survival rates divided by 10. Total scores represented the sum scores of both variables and were 3, 9 or 15 points. Six-month survival rates were 7%, 29% and 100% (P = 0.004). Thus, three predictors for functional outcomes were identified. The newly developed survival score included three prognostic groups. Patients with 3 points may receive 1 3 8 Gy, patients with 9 points 5 3 4 Gy and patients achieving 15 points longer-course radiotherapy. In the latter two groups, upfront decompressive surgery may be considered.
Key words: survival.
ambulatory status, melanoma, metastatic epidural spinal cord compression, motor function,
INTRODUCTION Metastatic epidural spinal cord compression (MESCC) is a serious oncologic emergency situation. Since the treatment of malignant disease is constantly improving, the numbers of patients who live long enough to develop metastases will increase. This will likely also apply to melanoma patients, since new agents such as ipilimumab and vemorafenib have been introduced for this tumor entity.1,2 Currently, melanoma accounts for only 1–2% of patients presenting with MESCC.3 However, due to the new agents, this may change and the proportion of melanoma patients among those with MESCC may increase. Since melanoma is a less radiosensitive tumor, the response to radiotherapy for MESCC in this particular group of patients is worse than for other primary tumors.4,5 Therefore,
upfront decompressive surgery in addition to radiotherapy should be discussed for MESCC from melanoma. Decompressive surgery is generally not indicated for patients with a short expected survival time of very few months.6,7 Thus, it appears desirable to be able to identify patients with MESCC from melanoma before the start of treatment, who will not respond well to radiotherapy alone, and to be able to estimate a patient’s survival prognosis in this particular group of patients with MESCC. The expected survival time will also have an impact on the fractionation schedule chosen for the individual patient. Patients with a poor prognosis should receive a less burdensome short-course program (for example 5 9 4 Gy in 1 week).8 Those patients with a more favorable prognosis should receive a longer-course schedule (for example 10 9 3 Gy in 2 weeks) to achieve better local control of
Correspondence: Dirk Rades, M.D., Department of Radiation Oncology, University of L€ ubeck, Ratzeburger Allee 160, 23538 L€ ubeck, Germany. Email: [email protected] Received 14 August 2014; accepted 9 September 2014.
1082
© 2014 Japanese Dermatological Association
Radiotherapy for MESCC from melanoma
MESCC.9 This study aimed to provide both predictors for postradiotherapy functional outcome (motor function and ambulatory status) and a tool for predicting a patient’s survival time in a cohort of patients treated with radiotherapy alone for MESCC from melanoma.
METHODS Twenty-seven patients who received irradiation alone for MESCC were included in this retrospective analysis. Inclusion criteria were presence of motor deficits of the legs, no prior surgery or irradiation to the involved parts of the spinal cord, and confirmation of MESCC by spinal computed tomography or magnetic resonance imaging. The option of upfront decompressive surgery had been discussed before radiotherapy was started. Patient characteristics are shown in Table 1. Irradiation was performed with a linear accelerator and 6–10 MV photon beams. Treatment volumes encompassed one normal vertebra above and below the metastatic lesions. Motor function and ambulatory status were graded before and up to 6 months following irradiation with a 5-point scale10: Grade 0, normal strength; Grade 1, ambulatory without aid; Grade 2, ambulatory with aid; Grade 3, not ambulatory; Grade 4, paraplegia. Improvement or deterioration was rated as a change of at least one grade. A total of 10 variables were analyzed for association with post-radiotherapy motor function. These variables included age (15 months), time developing motor deficits before the start of radiotherapy (≤7 vs >7 days), and the fractionation regimen (5 9 4 Gy vs longer-course radiotherapy with 10 9 3 Gy, 15 9 2.5 Gy or 20 9 2 Gy). The cut-off with respect to the time developing motor deficits before the start of radiotherapy (≤7 vs >7 days) was chosen based on the data shown in Table 2. All 10 variables were investigated with respect to post-radiotherapy motor function and ambulatory status, as well as with respect to survival. The effect on motor function was evaluated in a multivariate analysis performed with the ordered logit model. Post-radiotherapy ambulatory rates were analyzed in a univariate manner with the Chi-square test. For the univariate analysis of survival, the Kaplan–Meier method and the log-rank test were used. The variables achieving significance (P < 0.05) were additionally evaluated in multivariate manner using the Cox regression model.
RESULTS Improvement of motor function was noted in five patients (19%), no further progression of MESCC in 18 patients (67%), and deterioration in four patients (15%). The time developing motor deficits before radiotherapy was the only variable
© 2014 Japanese Dermatological Association
Table 1. Patient characteristics No. patients Age 15 months 15 Time developing motor deficits 0–3 days 6 4–7 days 5 8–11 days 5 12–15 days 3 >15 days 8 Fractionation schedule 5 9 4 Gy 8 Longer-course radiotherapy 19
% 48 52 33 67 30 70 44 56 41 59 48 52 41 59 44 56 22 19 19 11 30 30 70
significantly associated with the effect on motor function (P = 0.006). The results of the analysis of motor function are presented in Table 3. On univariate analysis, post-radiotherapy ambulatory status showed a significant association with ambulatory status before radiotherapy (P < 0.001) and ECOG performance status (P = 0.046). Time developing motor deficits before radiotherapy showed a strong trend (P = 0.051). In those patients who had all three favorable prognostic factors, the post-radiotherapy ambulatory rate was 100%. The post-radiotherapy ambulatory rates are shown in Table 4. The survival rates at 6 and 12 months were 33% and 22%, respectively. Variables that had a significant impact on survival were ECOG performance status (P < 0.001), number of involved vertebrae (P = 0.007), pre-radiotherapy ambulatory status (P = 0.020), further bone metastases (P = 0.023), visceral metastases (P < 0.001), and time developing motor deficits before the start of radiotherapy (P = 0.038) (see Table 5). In the additional multivariate analysis, ECOG performance status (risk ratio [RR] = 4.35; 95% confidence interval [CI] = 1.04– 16.67; P = 0.044) and visceral metastases (RR = 3.70; 95% CI = 1.10–12.50; P = 0.034) remained significant. Number of
1083
S. Huttenlocher et al.
Table 2. Time developing motor deficits prior to radiotherapy: associations with the effect of radiotherapy on motor function, with the ambulatory status following radiotherapy and with survival Time developing motor deficits prior to radiotherapy, %
Effect of radiotherapy on motor function Improvement No change Deterioration Ambulatory patients after radiotherapy Survival rates at 6 months
0–3 days (n = 6)
4–7 days (n = 5)
8–11 days (n = 5)
12–15 days (n = 3)
>15 days (n = 8)
0 67 33 33 17
0 60 40 20 20
20 80 0 80 40
33 67 0 67 33
38 63 0 88 50
Table 3. Effect of radiotherapy on motor function. P-values were obtained from the multivariate analysis (ordered logit model) Improvement n (%) Age 15 months 4 Time developing motor deficits ≤7 days 0 >7 days 5 Fractionation schedule 5 9 4 Gy 2 Longer-course radiotherapy 3
No change n (%)
Deterioration n (%)
P
(23) (14)
8 (62) 10 (71)
2 (15) 2 (14)
0.66
(11) (22)
7 (78) 11 (61)
1 (11) 3 (17)
0.44
(38) (11)
5 (63) 13 (68)
0 (0) 4 (21)
0.20
(33) (7)
7 (58) 11 (73)
1 (8) 3 (20)
0.07
(9) (25)
7 (64) 11 (69)
3 (27) 1 (6)
0.11
(38) (0)
7 (54) 11 (79)
1 (8) 3 (21)
0.08
(36) (6)
5 (45) 13 (81)
2 (18) 2 (13)
0.07
(8) (27)
9 (75) 9 (60)
2 (17) 1 (13)
0.13
(0) (31)
7 (64) 11 (69)
4 (36) 0 (0)
0.006
(25) (16)
5 (63) 13 (68)
1 (13) 3 (16)
0.09
Significant P values are given in bold.
involved vertebrae (RR = 1.82; 95% CI = 0.53–6.25; P = 0.34), pre-radiotherapy ambulatory status (RR = 1.66; 95% CI = 0.59–4.69; P = 0.34), further bone metastases (RR = 1.45; 95% CI = 0.51–4.12; P = 0.49), and time developing motor deficits before radiotherapy (RR = 1.21; 95% CI = 0.46–3.15; P = 0.70) were no longer significant in the multivariate analysis. The two significant variables, ECOG performance status and visceral metastases, were included in the predictive survival tool. Scoring points were obtained from the 6-month survival rates (given in %) divided by 10. The scores of the two
1084
variables were: 8 points for ECOG 2; 2 points for ECOG 3–4; 7 points for absence of visceral metastases; and 1 point for presence of visceral metastases, respectively. The total scores represented the sum of the scores of the two variables and were 3 points, 9 points or 15 points, respectively. The corresponding 6-month survival rates were 7%, 29% and 100%, respectively (P = 0.004), and the 12-month survival rates 0%, 14% and 83%, respectively (P = 0.001). In the group of patients with 3 points, 71% died within 2 months after radiotherapy.
© 2014 Japanese Dermatological Association
Radiotherapy for MESCC from melanoma
Table 4. Post-radiotherapy ambulatory status. P-values were obtained from the univariate analysis (ordinal regression) Ambulatory patients after radiotherapy (%) Age 15 months 53 Time developing motor deficits ≤7 days 27 >7 days 81 Fractionation schedule 5 9 4 Gy 75 Longer-course radiotherapy 53
Table 5. Survival rates at 6 months and at 12 months. P-values were obtained from the univariate analysis (log rank test) At 6 months (%)
P 0.79
0.76
0.046
0.52
7 days 44 Fractionation schedule 5 9 4 Gy 25 Longer-course radiotherapy 37
At 12 months (%)
P
23 21
0.78
22 22
0.72
63 5
Journal of Dermatology 2014; 41: 1082–1086
ORIGINAL ARTICLE
Motor function and survival following radiotherapy alone for metastatic epidural spinal cord compression in melanoma patients Stefan HUTTENLOCHER,1 Lena SEHMISCH,1 Volker RUDAT,2 Dirk RADES1 1
Department of Radiation Oncology, University of L€ubeck, L€ubeck, Germany, and 2Department of Radiation Oncology, Saad Specialist Hospital, Al-Khobar, Saudi Arabia
ABSTRACT The major goal of this study was the identification of predictors for motor function and survival after irradiation alone for metastatic epidural spinal cord compression (MESCC) from melanoma. Ten variables (age, gender, performance status, number of involved vertebrae, pre-radiotherapy ambulatory status, further bone metastases, visceral metastases, interval from melanoma diagnosis to MESCC, time developing motor deficits before radiotherapy, fractionation regimen) were investigated for post-radiotherapy motor function, ambulatory status and survival in 27 patients. On multivariate analysis, motor function was significantly associated with time developing motor deficits (P = 0.006). On univariate analysis, post-radiotherapy ambulatory rates were associated with pre-radiotherapy ambulatory status (P < 0.001) and performance status (P = 0.046). Variables having a significant impact on survival in the univariate analysis were performance status (P < 0.001), number of involved vertebrae (P = 0.007), pre-radiotherapy ambulatory status (P = 0.020), further bone metastases (P = 0.023), visceral metastases (P < 0.001), and time developing motor deficits (P = 0.038). On multivariate analysis of survival, the Eastern Cooperative Oncology Group (ECOG) performance status (risk ratio [RR] = 4.35; 95% confidence interval [CI] = 1.04–16.67; P = 0.044) and visceral metastases (RR = 3.70; 95% CI = 1.10–12.50; P = 0.034) remained significant and were included in a survival score. Scoring points were obtained from 6-month survival rates divided by 10. Total scores represented the sum scores of both variables and were 3, 9 or 15 points. Six-month survival rates were 7%, 29% and 100% (P = 0.004). Thus, three predictors for functional outcomes were identified. The newly developed survival score included three prognostic groups. Patients with 3 points may receive 1 3 8 Gy, patients with 9 points 5 3 4 Gy and patients achieving 15 points longer-course radiotherapy. In the latter two groups, upfront decompressive surgery may be considered.
Key words: survival.
ambulatory status, melanoma, metastatic epidural spinal cord compression, motor function,
INTRODUCTION Metastatic epidural spinal cord compression (MESCC) is a serious oncologic emergency situation. Since the treatment of malignant disease is constantly improving, the numbers of patients who live long enough to develop metastases will increase. This will likely also apply to melanoma patients, since new agents such as ipilimumab and vemorafenib have been introduced for this tumor entity.1,2 Currently, melanoma accounts for only 1–2% of patients presenting with MESCC.3 However, due to the new agents, this may change and the proportion of melanoma patients among those with MESCC may increase. Since melanoma is a less radiosensitive tumor, the response to radiotherapy for MESCC in this particular group of patients is worse than for other primary tumors.4,5 Therefore,
upfront decompressive surgery in addition to radiotherapy should be discussed for MESCC from melanoma. Decompressive surgery is generally not indicated for patients with a short expected survival time of very few months.6,7 Thus, it appears desirable to be able to identify patients with MESCC from melanoma before the start of treatment, who will not respond well to radiotherapy alone, and to be able to estimate a patient’s survival prognosis in this particular group of patients with MESCC. The expected survival time will also have an impact on the fractionation schedule chosen for the individual patient. Patients with a poor prognosis should receive a less burdensome short-course program (for example 5 9 4 Gy in 1 week).8 Those patients with a more favorable prognosis should receive a longer-course schedule (for example 10 9 3 Gy in 2 weeks) to achieve better local control of
Correspondence: Dirk Rades, M.D., Department of Radiation Oncology, University of L€ ubeck, Ratzeburger Allee 160, 23538 L€ ubeck, Germany. Email: [email protected] Received 14 August 2014; accepted 9 September 2014.
1082
© 2014 Japanese Dermatological Association
Radiotherapy for MESCC from melanoma
MESCC.9 This study aimed to provide both predictors for postradiotherapy functional outcome (motor function and ambulatory status) and a tool for predicting a patient’s survival time in a cohort of patients treated with radiotherapy alone for MESCC from melanoma.
METHODS Twenty-seven patients who received irradiation alone for MESCC were included in this retrospective analysis. Inclusion criteria were presence of motor deficits of the legs, no prior surgery or irradiation to the involved parts of the spinal cord, and confirmation of MESCC by spinal computed tomography or magnetic resonance imaging. The option of upfront decompressive surgery had been discussed before radiotherapy was started. Patient characteristics are shown in Table 1. Irradiation was performed with a linear accelerator and 6–10 MV photon beams. Treatment volumes encompassed one normal vertebra above and below the metastatic lesions. Motor function and ambulatory status were graded before and up to 6 months following irradiation with a 5-point scale10: Grade 0, normal strength; Grade 1, ambulatory without aid; Grade 2, ambulatory with aid; Grade 3, not ambulatory; Grade 4, paraplegia. Improvement or deterioration was rated as a change of at least one grade. A total of 10 variables were analyzed for association with post-radiotherapy motor function. These variables included age (15 months), time developing motor deficits before the start of radiotherapy (≤7 vs >7 days), and the fractionation regimen (5 9 4 Gy vs longer-course radiotherapy with 10 9 3 Gy, 15 9 2.5 Gy or 20 9 2 Gy). The cut-off with respect to the time developing motor deficits before the start of radiotherapy (≤7 vs >7 days) was chosen based on the data shown in Table 2. All 10 variables were investigated with respect to post-radiotherapy motor function and ambulatory status, as well as with respect to survival. The effect on motor function was evaluated in a multivariate analysis performed with the ordered logit model. Post-radiotherapy ambulatory rates were analyzed in a univariate manner with the Chi-square test. For the univariate analysis of survival, the Kaplan–Meier method and the log-rank test were used. The variables achieving significance (P < 0.05) were additionally evaluated in multivariate manner using the Cox regression model.
RESULTS Improvement of motor function was noted in five patients (19%), no further progression of MESCC in 18 patients (67%), and deterioration in four patients (15%). The time developing motor deficits before radiotherapy was the only variable
© 2014 Japanese Dermatological Association
Table 1. Patient characteristics No. patients Age 15 months 15 Time developing motor deficits 0–3 days 6 4–7 days 5 8–11 days 5 12–15 days 3 >15 days 8 Fractionation schedule 5 9 4 Gy 8 Longer-course radiotherapy 19
% 48 52 33 67 30 70 44 56 41 59 48 52 41 59 44 56 22 19 19 11 30 30 70
significantly associated with the effect on motor function (P = 0.006). The results of the analysis of motor function are presented in Table 3. On univariate analysis, post-radiotherapy ambulatory status showed a significant association with ambulatory status before radiotherapy (P < 0.001) and ECOG performance status (P = 0.046). Time developing motor deficits before radiotherapy showed a strong trend (P = 0.051). In those patients who had all three favorable prognostic factors, the post-radiotherapy ambulatory rate was 100%. The post-radiotherapy ambulatory rates are shown in Table 4. The survival rates at 6 and 12 months were 33% and 22%, respectively. Variables that had a significant impact on survival were ECOG performance status (P < 0.001), number of involved vertebrae (P = 0.007), pre-radiotherapy ambulatory status (P = 0.020), further bone metastases (P = 0.023), visceral metastases (P < 0.001), and time developing motor deficits before the start of radiotherapy (P = 0.038) (see Table 5). In the additional multivariate analysis, ECOG performance status (risk ratio [RR] = 4.35; 95% confidence interval [CI] = 1.04– 16.67; P = 0.044) and visceral metastases (RR = 3.70; 95% CI = 1.10–12.50; P = 0.034) remained significant. Number of
1083
S. Huttenlocher et al.
Table 2. Time developing motor deficits prior to radiotherapy: associations with the effect of radiotherapy on motor function, with the ambulatory status following radiotherapy and with survival Time developing motor deficits prior to radiotherapy, %
Effect of radiotherapy on motor function Improvement No change Deterioration Ambulatory patients after radiotherapy Survival rates at 6 months
0–3 days (n = 6)
4–7 days (n = 5)
8–11 days (n = 5)
12–15 days (n = 3)
>15 days (n = 8)
0 67 33 33 17
0 60 40 20 20
20 80 0 80 40
33 67 0 67 33
38 63 0 88 50
Table 3. Effect of radiotherapy on motor function. P-values were obtained from the multivariate analysis (ordered logit model) Improvement n (%) Age 15 months 4 Time developing motor deficits ≤7 days 0 >7 days 5 Fractionation schedule 5 9 4 Gy 2 Longer-course radiotherapy 3
No change n (%)
Deterioration n (%)
P
(23) (14)
8 (62) 10 (71)
2 (15) 2 (14)
0.66
(11) (22)
7 (78) 11 (61)
1 (11) 3 (17)
0.44
(38) (11)
5 (63) 13 (68)
0 (0) 4 (21)
0.20
(33) (7)
7 (58) 11 (73)
1 (8) 3 (20)
0.07
(9) (25)
7 (64) 11 (69)
3 (27) 1 (6)
0.11
(38) (0)
7 (54) 11 (79)
1 (8) 3 (21)
0.08
(36) (6)
5 (45) 13 (81)
2 (18) 2 (13)
0.07
(8) (27)
9 (75) 9 (60)
2 (17) 1 (13)
0.13
(0) (31)
7 (64) 11 (69)
4 (36) 0 (0)
0.006
(25) (16)
5 (63) 13 (68)
1 (13) 3 (16)
0.09
Significant P values are given in bold.
involved vertebrae (RR = 1.82; 95% CI = 0.53–6.25; P = 0.34), pre-radiotherapy ambulatory status (RR = 1.66; 95% CI = 0.59–4.69; P = 0.34), further bone metastases (RR = 1.45; 95% CI = 0.51–4.12; P = 0.49), and time developing motor deficits before radiotherapy (RR = 1.21; 95% CI = 0.46–3.15; P = 0.70) were no longer significant in the multivariate analysis. The two significant variables, ECOG performance status and visceral metastases, were included in the predictive survival tool. Scoring points were obtained from the 6-month survival rates (given in %) divided by 10. The scores of the two
1084
variables were: 8 points for ECOG 2; 2 points for ECOG 3–4; 7 points for absence of visceral metastases; and 1 point for presence of visceral metastases, respectively. The total scores represented the sum of the scores of the two variables and were 3 points, 9 points or 15 points, respectively. The corresponding 6-month survival rates were 7%, 29% and 100%, respectively (P = 0.004), and the 12-month survival rates 0%, 14% and 83%, respectively (P = 0.001). In the group of patients with 3 points, 71% died within 2 months after radiotherapy.
© 2014 Japanese Dermatological Association
Radiotherapy for MESCC from melanoma
Table 4. Post-radiotherapy ambulatory status. P-values were obtained from the univariate analysis (ordinal regression) Ambulatory patients after radiotherapy (%) Age 15 months 53 Time developing motor deficits ≤7 days 27 >7 days 81 Fractionation schedule 5 9 4 Gy 75 Longer-course radiotherapy 53
Table 5. Survival rates at 6 months and at 12 months. P-values were obtained from the univariate analysis (log rank test) At 6 months (%)
P 0.79
0.76
0.046
0.52
7 days 44 Fractionation schedule 5 9 4 Gy 25 Longer-course radiotherapy 37
At 12 months (%)
P
23 21
0.78
22 22
0.72
63 5
