Evaluation of Five Radiation Schedules and Prognostic Factors for Metastatic Spinal Cord Compression
http://www.100md.com
《临床肿瘤学》
the Department of Radiation Oncology, University Hospital, Hamburg
Department of Radiation Oncology, Medical School, Hannover
Department of Radiation Oncology, University Hospital, Luebeck, Germany
Department of Radiotherapy, Academic Medical Center, Amsterdam
Department of Radiation Oncology, Dr Bernard Verbeeten Institute, Tilburg, the Netherlands
Mount Vernon Centre for Cancer Treatment, Northwood, Middlesex, United Kingdom
Department of Radiation Oncology, University Hospital, Sarajevo, Bosnia-Herzegovina
Department of Radiation Oncology, Mayo Clinic, Scottsdale, AZ
ABSTRACT
PURPOSE: To study five radiotherapy (RT) schedules and potential prognostic factors for functional outcome in metastatic spinal cord compression (MSCC).
PATIENTS AND METHODS: One thousand three hundred four patients who were irradiated from January 1992 to December 2003 were included in this retrospective review. The schedules of 1 x 8 Gy in 1 day (n = 261), 5 x 4 Gy in 1 week (n = 279), 10 x 3 Gy in 2 weeks (n = 274), 15 x 2.5 Gy in 3 weeks (n = 233), and 20 x 2 Gy in 4 weeks (n = 257) were compared for motor function, ambulatory status, and in-field recurrences. The following potential prognostic factors were investigated: age, sex, performance status, histology, number of involved vertebra, interval from cancer diagnosis to MSCC, pretreatment ambulatory status, and time of developing motor deficits before RT. A multivariate analysis was performed with the ordered logit model.
RESULTS: Motor function improved in 26% (1 x 8 Gy), 28% (5 x 4 Gy), 27% (10 x 3 Gy), 31% (15 x 2.5 Gy), and 28% (20 x 2 Gy); and posttreatment ambulatory rates were 69%, 68%, 63%, 66%, and 74% (P = .578), respectively. On multivariate analysis, age, performance status, primary tumor, involved vertebra, interval from cancer diagnosis to MSCC, pretreatment ambulatory status, and time of developing motor deficits were significantly associated with functional outcome, whereas the RT schedule was not. Acute toxicity was mild, and late toxicity was not observed. In-field recurrence rates at 2 years were 24% (1 x 8 Gy), 26% (5 x 4 Gy), 14% (10 x 3 Gy), 9% (15 x 2.5 Gy), and 7% (20 x 2 Gy) (P < .001). Neither the difference between 1 x 8 Gy and 5 x 4 Gy (P = .44) nor between 10 x 3 Gy, 15 x 2.5 Gy, and 20 x 2 Gy (P = .71) was significant.
CONCLUSION: The five RT schedules provided similar functional outcome. The three more protracted schedules seemed to result in fewer in-field recurrences. To minimize treatment time, the following two schedules are recommended: 1 x 8 Gy for patients with poor predicted survival and 10 x 3 Gy for other patients. Results should be confirmed in a prospective randomized trial.
INTRODUCTION
Metastatic spinal cord compression (MSCC) occurs in 5% to 10% of all cancer patients during the course of their disease. MSCC is associated with neurologic signs including diminished motor function. If pain is the only clinical symptom or if the diagnosis is based on radiologic studies alone, it should be described as impending MSCC.
Urgent treatment is required for MSCC to avoid progression of motor deficits resulting in paraplegia.1 Radiotherapy (RT) and decompressive surgery are the most important treatment modalities.2
The most appropriate RT schedule is unknown because of the lack of studies comparing various regimens. Because many RT schedules are used worldwide, such as 1 x 8 Gy, 1 x 10 Gy, 2 x 6 Gy, 2 x 8 Gy, 5 x 4 Gy, 6 x 4 Gy, 5 x 5 Gy, 10 x 3 Gy, 14 x 2.5 Gy, 15 x 2.5 Gy, 20 x 2 Gy, and various split-course regimens, there is no standardized RT for MSCC.
A short overall treatment time would be desirable for patient convenience and comfort, especially because the life expectancy of patients with MSCC is generally short and many MSCC patients are debilitated and unable to walk.3 The daily visits to the RT department and the positioning on the treatment couch are associated with significant discomfort. The cost of providing longer RT schedules is greater and is only justified if longer regimens produce better results. A shorter treatment course, although desirable, can only be recommended if it provides similar functional outcome as more protracted RT schedules.
Another important end point is the rate of in-field recurrences, which can require reirradiation of a previously treated region. This end point has not yet been sufficiently investigated in studies of MSCC treatment.
The selection of the treatment schedule is often influenced by prognostic factors. In the literature, the following four relevant prognostic factors have been recognized for functional outcome of RT for MSCC: the length of the time of developing motor deficits before RT, the histology of the primary tumor, the interval from first diagnosis of cancer to MSCC, and the pretreatment ambulatory status.4-12 In addition to these factors, this study evaluated four other potential prognostic factors of age, sex, performance status, and the number of involved vertebra. The potential prognostic factors to be investigated were defined a priori.
This international multicenter study included the largest MSCC patient cohort ever presented and compared five different RT schedules used for MSCC with respect to resulting functional outcome. The primary goal was to investigate the optimal therapy for this common clinical problem.
PATIENTS AND METHODS
A total of 1,304 MSCC patients were included in this retrospective analysis. The data were obtained from the patient files and from the patients' general practitioners. Inclusion criteria were as follows: motor dysfunction of the lower extremities, no previous surgery or RT of the spinal region concerned, no concurrent chemotherapy, and survival of at least 1 month after RT. The latter criterion allows the evaluation of motor function 1 month after RT. This seems more appropriate than the evaluation directly after RT because no treatment effect can be expected directly after single-fraction RT such as 1 x 8 Gy. Before RT was started, the patients were usually presented to a neurosurgeon to discuss the option of decompressive surgery if indicated.
Patients with a history of a brain tumor, brain metastasis, or other major neurologic diseases, which may result in motor dysfunction, were excluded. The diagnosis of MSCC was confirmed by magnetic resonance imaging (MRI) or computed tomography (CT). Patients received dexamethasone at a moderate to intermediate dose level (16 to 32 mg/d) during the whole RT. If 1 x 8 Gy or 5 x 4 Gy were applied, dexamethasone was administered for a week.
The five RT schedules that were compared were as follows: 1 x 8 Gy in 1 day (n = 261), 5 x 4 Gy in 1 week (n = 279), 10 x 3 Gy in 2 weeks (n = 274), 15 x 2.5 Gy in 3 weeks (n = 233), and 20 x 2 Gy in 4 weeks (n = 257). The RT schedules were compared for the following three posttreatment end points: motor function, ambulatory status, and in-field recurrences. The patients were treated between January 1992 and December 2003 and observed for at least 6 months or until death. Each series of the contributing centers represented an unselected group of patients treated in a specific time period. Of the five investigated RT schedules, each center used two to three different schedules, usually related to the treatment capacity on the linear accelerators and not to the patients' performance status or to the type of primary tumor. Parts of the data from Hamburg and Hannover, Germany, and Northwood, United Kingdom, have been included in previously published studies.4,5,13
Irradiation was performed with a linear accelerator (6 to 10 MV) or with a cobalt 60 treatment machine. RT doses were prescribed to the spinal cord using CT scans and MRI. The treatment volume encompassed one to two normal vertebra above and below the metastatic lesions. Irradiation was delivered through a single posterior field or parallel opposed fields depending on the depth of spinal cord.
The median age was 63 years (range, 23 to 89 years). Of the patients, 543 (42%) were female, and 761 (59%) were male. The distribution of the primary tumors was as follows: breast cancer, 335 patients (26%); prostate cancer, 276 patients (21%); lung cancer, 184 patients (14%); lymphoma or myeloma, 138 patients (11%); cancer of unknown primary, 102 patients (8%); gastrointestinal cancer, 95 patients (7%); renal cancer, 71 patients (5%); and other cancers, 241 patients (18%). The median interval from first diagnosis of cancer to MSCC was 24 months (range, 1 to 276 months). Patient characteristics related to the five treatment groups are listed in Table 1. The treatment groups were well balanced for the investigated variables.
Motor function and ambulatory status were evaluated before RT and at 1 month, 3 months, and 6 months after RT. Motor function was graded with a 5-point scale according to Tomita et al14 as follows: grade 0, normal strength; grade 1, ambulatory without aid; grade 2, ambulatory with aid; grade 3, not ambulatory; and grade 4, paraplegia. Improvement or deterioration of motor function was defined as a change of at least 1 point.
The following potential prognostic factors were evaluated with respect to functional outcome: age ( 63 years v 64 years), Eastern Cooperative Oncology Group (ECOG) performance status (1 to 2 v 3 to 4), number of involved vertebra (one to two v three to four v five vertebra), interval from cancer diagnosis to MSCC ( 24 months v > 24 months), ambulatory status before RT (ambulatory/Tomita grade 0 to 2 v nonambulatory/Tomita grade 3 to 414), type of primary tumor (favorable: myeloma, lymphoma, testicular seminoma; v unfavorable: cancer of unknown primary, lung cancer, melanoma; v intermediate: other tumors), and the length of time of developing motor deficits before RT (1 to 7 days v 8 to 14 days v > 14 days; Table 2).
The potential prognostic factors were included in a multivariate analysis that was performed with the ordered logit model, which is also known as the proportional odds model, because the data regarding functional outcome are ordinal (–1 = deterioration, 0 = no change, and 1 = improvement). The comparison of the five RT schedules with respect to functional outcome was also performed with the ordered logit model.
In-field recurrence was defined as clinically significant recurrence of MSCC associated with motor deficits in the preirradiated spinal region. Diagnosis of recurrence was confirmed by spinal CT or MRI. Time to recurrence was measured from the end of RT. The comparison of the five treatment groups with respect to in-field recurrences was performed with the Kaplan-Meier method and the log-rank test.15
Furthermore, potential prognostic factors for survival after RT of MSCC were evaluated including age, performance status, number of involved vertebra, interval from cancer diagnosis to MSCC, ambulatory status before RT, type of primary tumor, time of developing motor deficits before RT, and response to RT. Univariate analysis was performed with the Kaplan-Meier method and the log-rank test,15 and multivariate analysis was performed with the Cox proportional hazards model.
RESULTS
The evaluation of potential prognostic factors for resulting motor function is summarized in Table 2. The comparison of the five RT schedules for improvement, no change, and deterioration of motor function is shown in Figures 1, 2, and 3, respectively, and in Table 3. No significant differences were found between the five groups at each time of follow-up. The results of the multivariate analysis are listed in Table 4.
Ambulatory rates before and after RT were not significantly different for the five treatment groups (Fig 4). Twenty-six percent of the patients (126 of 477 patients) who were not ambulatory before RT regained the ability to walk, 25% (23 of 91 patients) after 1 x 8 Gy, 26% (27 of 104 patients) after 5 x 4 Gy, 26% (31 of 118 patients) after 10 x 3 Gy, 24% (22 of 90 patients) after 15 x 2.5 Gy, and 30% (23 of 76 patients) after 20 x 2 Gy (P = .96).
The proportions of patients with a follow-up of 3 months were 87% (1 x 8 Gy), 85% (5 x 4 Gy), 84% (10 x 3 Gy), 89% (15 x 2.5 Gy), and 89% (20 x 2 Gy); and the proportions of patients with a follow-up of 6 months were 59%, 61%, 58%, 60%, and 68%, respectively. The proportions of patients with a follow-up of 12 months were 33%, 35%, 34%, 37%, and 39%, respectively. Median follow-up times for survivors were as follows: 14 months (range, 3 to 58 months) after 1 x 8 Gy, 15 months (range, 2 to 61 months) after 5 x 4 Gy, 15 months (range, 4 to 38 months) after 10 x 3 Gy, 16 months (range, 4 to 89 months) after 15 x 2.5 Gy, and 14 months (range, 4 to 76 months) after 20 x 2 Gy. Thus, a relevant bias as a result of differences in follow-up by treatment group seemed unlikely.
The results of the Kaplan-Meier analysis for in-field recurrences are demonstrated in Figure 5. Significantly more recurrences occurred after 1 x 8 Gy and 5 x 4 Gy than after 10 x 3 Gy, 15 x 2.5 Gy, and 20 x 2 Gy. Between 1 x 8 Gy and 5 x 4 Gy, no significant difference was observed. There was also no significant difference observed between 10 x 3 Gy, 15 x 2.5 Gy, and 20 x 2 Gy.
The treatment modalities applied at the time of a recurrence and the outcomes are listed in Table 5. Surgery, reirradiation, and chemotherapy resulted in improvement of motor function in seven (85%) of 12 patients, 26 (35%) of 74 patients, and none (0%) of five patients, respectively. Ten patients, who were irradiated with one of the three more protracted regimens, received no further treatment at the time of recurrence and deteriorated. Median follow-up times after recurrence were 8 months (range, 2 to 27 months) in the 1 x 8 Gy group, 7 months (range, 1 to 29 months) in the 5 x 4 Gy group, 4.5 months (range, 2 to 14 months) in the 10 x 3 Gy group, 5 months (range, 1 to 22 months) in the 15 x 2.5 Gy group, and 7 months (range, 1 to 28 months) in the 20 x 2 Gy group.
In the five treatment groups, no relevant acute or late RT-related toxicity was observed. Acute toxicity did not exceed grade 1 according to the National Cancer Institute Common Toxicity Criteria.16 Late toxicity, such as RT myelopathy, did not occur. In the patients surviving 6 months, the median follow-up period was 13 months (range, 6 to 92 months).
The results of the univariate and the multivariate analyses investigating potential prognostic factors with respect to survival after RT of MSCC are listed in Table 6. ECOG performance status, number of involved vertebra, interval from cancer diagnosis to MSCC, ambulatory status before RT, type of primary tumor, time of developing motor deficits before RT, and response to RT were significantly associated with survival, whereas age had no significant impact.
DISCUSSION
MSCC causes neurologic deficits, which are almost always preceded by pain. According to the literature, no particular RT schedule is superior to others with respect to pain relief.17-20 Since two large randomized trials were published in 1999, single-fraction RT with 1 x 8 Gy has been frequently used for painful bone metastasis.18,20 In contrast, the optimum RT schedule for the treatment of motor dysfunction is unclear because of a lack of comparative studies examining functional outcome in MSCC.
This multicenter study is the largest one ever assembled and the first comparing five different RT schedules with respect to subsequent functional outcome in MSCC. The primary goal of this study was to investigate whether RT schedules with a short overall treatment time, such as 1 x 8 Gy in 1 day and 5 x 4 Gy in 1 week, are as effective as longer treatment programs with respect to functional outcome.
The data demonstrated that the five investigated RT schedules were comparably effective for functional outcome and posttreatment ambulatory rates. A similar proportion of patients who were not ambulatory before RT regained the ability to walk in each of the five treatment groups.
However, it has to be taken into account that this is a retrospective analysis, which may be associated with potential biases. The retrospective nature of this analysis did not allow the application of a more differentiated grading system for motor function than the Tomita scale because more detailed information about the functional status was not available in most patients. Appropriate information about the effect of RT on potential sensory deficits or dysfunction of bowel and bladder was also not available. Thus, our analysis focuses on motor function. Another potential bias may have been introduced because of the fact that patients with a survival time of less than 1 month after RT were excluded from the analysis.
Application of 1 x 8 Gy instead of 20 x 2 Gy means a reduction of the overall treatment time from 4 weeks to only 1 day and a reduction of the number of treatment sessions from 20 to only one. Because life expectancy in most MSCC patients is limited to a few months and the majority of the patients are quite debilitated, a shorter treatment program is highly desirable. Each treatment can cause discomfort and inconvenience for the patient during transportation to the radiation department and the positioning on the treatment couch. Furthermore, a longer program substantially increases the cost of therapy.
A prospective study from Germany suggested that a regimen of 10 x 3 Gy is more effective with respect to bone recalcification than a regimen of 1 x 8 Gy.21 Bone recalcification can be expected only several months after RT. Thus, this might be important for patients with a relatively long life expectancy only. Extended survival can be expected most likely for patients with breast cancer, prostate cancer, myeloma, or lymphoma.10,12,22 For these patients, longer fractionated RT schedules could be more appropriate.
According to our results, 1 x 8 Gy is comparable to 5 x 4 Gy with respect to functional outcome. Thus, patients with a poor survival prognosis (eg, < 4 to 6 months), in whom bone recalcification is not important, could be treated with 1 x 8 Gy to keep the overall treatment time as short as possible, resulting in less discomfort for the patient. For patients with a better prognosis (expected survival of at least 4 to 6 months), a more protracted RT schedule should be applied to achieve better bone recalcification and fewer recurrences. Because our results demonstrated a comparable effect on functional outcome for the three more protracted regimens, 10 x 3 Gy should be used for such patients because it is associated with the shortest overall treatment time and the least discomfort for the patient. According to our data, survival after RT of MSCC depends on various prognostic factors such as performance status, number of involved vertebra, interval from cancer diagnosis to MSCC, ambulatory status before RT, type of primary tumor, time of developing motor deficits before RT, and response to RT.
The five groups compared in this study were balanced with respect to the seven prognostic factors examined (Table 1). Therefore, these factors did not have a relevant impact on the dose response. The optimum dose of corticosteroids could not be investigated because patients received comparable doses of dexamethasone during RT. In a randomized trial comparing high-dose dexamethasone (96 mg daily) to no corticosteroids during RT, gait function after RT was obtained in 81% of the patients who had received high-dose dexamethasone compared with 63% of the patients who did not receive corticosteroids.23 A historical case-control series that compared high-dose with moderate-dose (16 mg daily) dexamethasone demonstrated a significantly higher rate of serious side effects for the high-dose regimen (14% v 0%, respectively).24 The patients in the present series received dexamethasone at a moderate to intermediate dose level. There is a need for further investigation to determine the optimum dose of dexamethasone or other corticosteroids.25
Because each series contributed by the participating centers represented an unselected group of patients treated in a specific time period, the potential selection bias was excluded. The administration of the five schedules (two or three schedules per center) was related to the waiting list of patients and not to other factors, such as type of tumor histology or performance status. This makes the possibility of bias unlikely. Between the centers contributing patients, there was no relevant difference regarding RT techniques and the prescription of RT doses. Again, a selection bias seemed unlikely.
In RT oncology, the effect of a RT schedule on tumor control and on late toxicity depends on both the total dose and the dose per fraction. The five schedules can be compared with the equivalent dose in 2 Gy fractions (EQD2), which is calculated by the following equation: EQD2 = D x [(d + /)/(2 Gy + /)], which is derived from the linear-quadratic model; D represents total dose, d represents dose per fraction, represents linear (first-order, dose-dependent) component of cell killing, represents quadratic (second-order, dose-dependent) component of cell killing (more reparable), and / ratio represents the dose at which both components of cell killing are equal.26,27
The / ratio suggested for tumor effect is 10 Gy, resulting in an EQD2 of 12 Gy for the schedule of 1 x 8 Gy, 23.3 Gy for the schedule of 5 x 4 Gy, 32.5 Gy for the schedule of 10 x 3 Gy, 39 Gy for the schedule of 15 x 2.5 Gy, and 40 Gy for the schedule of 20 x 2 Gy. Thus, 15 x 2.5 Gy and 20 x 2 Gy would be expected to be the most effective schedules for tumor effect and functional outcome. However, our results did not reveal a significant difference between the five RT schedules in their effect on functional outcome or ambulatory status.
In addition to the five fractionation schedules, this study investigated seven potential prognostic factors for functional outcome in MSCC. A significant impact was found for the length of time of developing motor deficits before RT, the histology of the primary tumor, the ECOG performance status, the interval from cancer diagnosis to MSCC, the pretreatment ambulatory status, and the number of involved vertebra. The histology of the primary tumor and the pretreatment ambulatory status have been recognized earlier as prognostic factors.6-12 The ECOG performance status has not yet been established as a relevant prognostic factor, but it is correlated with the ambulatory status. Patients who are not able to walk, even with aid, have an ECOG performance status of either 3 or 4. A potential prognostic impact of the number of involved vertebra has been mentioned by Huddart et al22 in a series of 67 prostate cancer patients. The time of developing motor deficits before RT was already suggested to be a relevant prognostic factor in MSCC treatment.4,5 The poor functional outcome after a rapid development of motor dysfunction can be explained by disruption of the arterial blood flow by a rapidly constricted spinal cord, which leads to spinal cord infarction. A slower growing lesion is suggested to first cause venous congestion, which is more reversible.28 Another possible explanation for longer lasting symptoms predicting better functional outcome may be a reflection of tumor biology. More aggressive tumors would be expected to grow faster and cause a faster course of motor dysfunction. The aggressiveness of a tumor may also be reflected by the interval from cancer diagnosis to MSCC, which has been previously described as a potential prognostic factor.7 A longer interval is suggested to predict better outcome. This is in accordance with our results that demonstrated a significant association between the interval from first diagnosis of cancer to MSCC.
Relevant acute or late RT-related toxicity was not observed in our series. The tolerance dose (5% late toxicity within 5 years) for RT myelopathy is considered to be 45 to 50 Gy for conventional fractionation (dose per fraction of 1.8 to 2 Gy).29 If the end point is myelopathy, the EQD2 has to be calculated with an / ratio of 2 Gy, resulting in an EQD2 of 20 Gy for the schedule of 1 x 8 Gy, 30 Gy for the schedule of 5 x 4 Gy, 37.5 Gy for the schedule of 10 x 3 Gy, 42 Gy for the schedule of 15 x 2.5 Gy, and 40 Gy for the schedule of 20 x 2 Gy.26,27 These doses are below the tolerance dose, making late toxicity unlikely. The limited survival associated with MSCC also makes late toxicity unlikely. However, toxicity might be enhanced if concurrent chemotherapy is administered during or immediately after RT. Wallington et al12 reported six (18%) of 34 deaths in a series of lymphoma and myeloma patients to be related to toxicity from chemotherapy. In the present study, patients with concurrent chemotherapy were excluded from the analysis.
The five RT schedules were also compared with respect to in-field recurrences. Recurrences occurred significantly more frequent after 1 x 8 Gy and 5 x 4 Gy than after 10 x 3 Gy, 15 x 2.5 Gy, and 20 x 2 Gy. However, a bias may have been introduced as a result of the fact that more patients may have had in-field recurrences but were not referred for reirradiation because of poor prognosis or the wrong impression that reirradiation is not effective. The reirradiation rate may have also been influenced by the concern about myelopathy, which becomes more of an issue after initial RT with higher doses such as 30 to 40 Gy. This question can only be properly answered in a prospective manner, where all patients who are treated with different RT schedules are assessed the same way at the same time points.
In our series, approximately every fifth patient treated with 1 x 8 Gy, and approximately every 10th patient treated with 10 x 3 Gy, 15 x 3 Gy, or 20 x 2 Gy, needed re-treatment within 1 year after initial RT. Taking into account the EQD2 for late toxicity and the tolerance dose for myelopathy, it seems that reirradiation can be performed more safely after 1 x 8 Gy and 5 x 4 Gy than after a more protracted schedule with a higher total dose (EQD2). This has been considered by the contributing centers. Reirradiation was performed in almost every patient with a recurrence after 1 x 8 Gy and 5 x 4 Gy, whereas only 12 (35%) of the 34 patients initially treated with 10 x 3 Gy, 15 x 2.5 Gy, or 20 x 2 Gy received a second RT treatment. However, in these 12 patients, RT-related myelopathy was not observed after a median follow-up of 8 months (range, two to 29 months). This finding was supported by a recently presented analysis that demonstrated that reirradiation of the spinal cord seemed possible with an acceptable risk of myelopathy, also after initial treatment with 10 x 3 Gy, 15 x 2.5 Gy, or 20 x 2 Gy.30 The most effective treatment for in-field recurrence was decompressive surgery. However, it was only performed in selected patients with only one involved vertebra. The value of surgery in case of involvement of only one vertebra has recently been shown by Patchell et al.2
Our data demonstrated that the five investigated RT schedules were comparably effective with respect to functional outcome in MSCC. The schedule with the shortest overall treatment time (1 x 8 Gy) should be administered to patients with a poor estimated survival (eg, < 4 to 6 months). These short programs are associated with good initial results in terms of improved functionality.
More protracted schedules (10 x 3 Gy, 15 x 2.5 Gy, and 20 x 2 Gy) seem to be associated with fewer relapses, although this finding may have been confounded by a potential bias because of concerns about reirradiation if the primary RT was performed with comparably high total doses. This question needs to be clarified in a prospective manner.
Furthermore, more protracted RT schedules are suggested to result in better recalcification than 1 x 8 Gy and 5 x 4 Gy. Because significant bone recalcification occurs only several months after RT, more protracted schedules can be recommended for patients with a longer expected survival (eg, longer than 4 to 6 months). Because no significant difference was observed between the three more protracted schedules investigated in this study, 10 x 3 Gy would be the best option because it offers the shortest overall treatment time and the lowest costs. This study supports the application of a more standardized RT approach for MSCC by using only two schedules, 1 x 8 Gy or 10 x 3 Gy, depending on the estimated patient survival. Because of the retrospective nature of our study, the results need to be confirmed in a prospective, randomized, multicenter trial. Such a comparative trial is underway in our institutions.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
NOTES
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Poortmans P, Vulto A, Raaijmakers E: Always on a Friday Time pattern of referral for spinal cord compression. Acta Oncol 40:88-91, 2001
Patchell R, Tibbs PA, Regine WF, et al: A randomized trial of direct decompressive surgical resection in the treatment of spinal cord compression caused by metastasis. Proc Am Soc Clin Oncol 22:1, 2003 (abstr 2)
Ampil F: Epidural compression from metastatic tumor with resultant paralysis. J Neurooncol 7:129-136, 1989
Rades D, Karstens JH, Alberti W: Role of radiotherapy in the treatment of motor dysfunction due to metastatic spinal cord compression: Comparison of three different fractionation schedules. Int J Radiat Oncol Biol Phys 54:1160-1164, 2002
Rades D, Heidenreich F, Karstens JH: Final results of a prospective study for the prognostic value of the time of developing motor deficits before irradiation in metastatic spinal cord compression. Int J Radiat Oncol Biol Phys 53:975-979, 2002
Helweg-Larsen S: Clinical outcome in metastatic spinal cord compression. Acta Neurol Scand 94:269-275, 1996
Helweg-Larsen S, Srensen PS, Kreiner S: Prognostic factors in metastatic spinal cord compression: A prospective study using multivariate analysis of variables influencing survival and gait function in 153 patients. Int J Radiat Oncol Biol Phys 46:1163-1169, 2000
Kim RY, Smith JW, Spencer SA, et al: Malignant epidural spinal cord compression associated with a paravertebral mass: Its radiotherapeutic outcome on radiosensitivity. Int J Radiat Oncol Biol Phys 27:1079-1083, 1993
Leviov M, Dale J, Stein M, et al: The management of metastatic spinal cord compression: A radiotherapeutic success ceiling. Int J Radiat Oncol Biol Phys 27:231-234, 1993
Maranzano E, Latini P: Effectiveness of radiation therapy without surgery in metastatic spinal cord compression: Final results from a prospective trial. Int J Radiat Oncol Biol Phys 32:959-967, 1995
Srensen PS, Brgesen SE, Rohde K, et al: Metastatic epidural spinal cord compression. Cancer 65:1502-1509, 1990
Wallington M, Mendis S, Premawardhana U, et al: Local control and survival in spinal cord compression from lymphoma and myeloma. Radiother Oncol 42:43-47, 1997
Hoskin PJ, Grover A, Bhana R: Metastatic spinal cord compression: Radiotherapy outcome and dose fractionation. Radiother Oncol 68:175-180, 2003
Tomita T, Galicich JH, Sundaresan N: Radiation therapy for spinal epidural metastases with complete block. Acta Radiol Oncol 22:135-143, 1983
Kaplan EL, Meier P: Non parametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958
Trotti A, Byhardt R, Stetz J, et al: Common toxicity criteria: Version 2.0—An improved reference for grading the acute effects of cancer treatment: Impact on radiotherapy. Int J Radiat Oncol Biol Phys 47:13-47, 2000
Niewald M, Tkocz HJ, Abel U, et al: Rapid course radiation therapy vs. more standard treatment: A randomised trial for bone metastases. Int J Radiat Oncol Biol Phys 36:1085-1089, 1996
Bone Pain Trial Working Party: 8 Gy single fraction radiotherapy for the treatment of metastatic skeletal pain: Randomised comparison with a multifraction schedule over 12 months of patient follow-up. Radiother Oncol 52:111-121, 1999
Price P, Hoskin PJ, Easton D, et al: Prospective randomised trial of single and multifraction radiotherapy schedules in the treatment of painful bony metastases. Radiother Oncol 6:247-255, 1986
Steenland E, Leer J, van Houwelingen H, et al: The effect of a single fraction compared to multiple fraction on painful bone metastases: A global analysis of the Dutch Bone Metastasis Study. Radiother Oncol 52:101-109, 1999
Koswig S, Budach V: Re-calcification and pain relief following radiotherapy for bone metastases: A randomised trial of 2 different fractionation schedules (10 x 3 Gy vs 1 x 8 Gy). Strahlenther Onkol 175:500-508, 1999
Huddart RA, Rajan B, Law M, et al: Spinal cord compression in prostate cancer: Treatment outcome and prognostic factors. Radiother Oncol 44:229-236, 1997
Srensen PS, Helweg-Larsen S, Mouridsen H, et al: Effect of high-dose dexamethasone in carcinomatous metastatic spinal cord compression treated with radiotherapy: A randomized trial. Eur J Cancer 30A:22-27, 1994
Heimdal K, Hirschberg H, Slettebo H, et al: High incidence of serious side effects of high-dose dexamethasone treatment in patients with epidural spinal cord compression. J Neurooncol 12:141-144, 1992
Loblaw DA, Lapierre NJ: Emergency treatment of malignant extradural spinal cord compression: An evidence-based guideline. J Clin Oncol 16:1613-1624, 1998
Barendsen GW: Dose fractionation, dose rate and iso-effect relationships for normal tissue responses. Int J Radiat Oncol Biol Phys 8:1981-1997, 1982
Joiner MC, Van der Kogel AJ: The linear-quadratic approach to fractionation and calculation of isoeffect relationships, in Steel GG (ed): Basic Clinical Radiobiology. New York, NY, Oxford University Press, 1997, pp 106-112
Manabe S, Tanaka H, Higo Y, et al: Experimental analysis of the spinal cord compressed by spinal metastasis. Spine 14:1308-1315, 1989
Emami B, Lyman J, Brown A, et al: Tolerance of normal tissue to therapeutic irradiation. Int J Radiat Oncol Biol Phys 21:109-122, 1991
Andratschke N, Nieder C, Grosu AL, et al: A risk score for human spinal cord re-irradiation based on clinical data from 40 patients. Radiother Oncol 73:S207, 2004 (abstr 466)(Dirk Rades, Lukas J.A. St)
Department of Radiation Oncology, Medical School, Hannover
Department of Radiation Oncology, University Hospital, Luebeck, Germany
Department of Radiotherapy, Academic Medical Center, Amsterdam
Department of Radiation Oncology, Dr Bernard Verbeeten Institute, Tilburg, the Netherlands
Mount Vernon Centre for Cancer Treatment, Northwood, Middlesex, United Kingdom
Department of Radiation Oncology, University Hospital, Sarajevo, Bosnia-Herzegovina
Department of Radiation Oncology, Mayo Clinic, Scottsdale, AZ
ABSTRACT
PURPOSE: To study five radiotherapy (RT) schedules and potential prognostic factors for functional outcome in metastatic spinal cord compression (MSCC).
PATIENTS AND METHODS: One thousand three hundred four patients who were irradiated from January 1992 to December 2003 were included in this retrospective review. The schedules of 1 x 8 Gy in 1 day (n = 261), 5 x 4 Gy in 1 week (n = 279), 10 x 3 Gy in 2 weeks (n = 274), 15 x 2.5 Gy in 3 weeks (n = 233), and 20 x 2 Gy in 4 weeks (n = 257) were compared for motor function, ambulatory status, and in-field recurrences. The following potential prognostic factors were investigated: age, sex, performance status, histology, number of involved vertebra, interval from cancer diagnosis to MSCC, pretreatment ambulatory status, and time of developing motor deficits before RT. A multivariate analysis was performed with the ordered logit model.
RESULTS: Motor function improved in 26% (1 x 8 Gy), 28% (5 x 4 Gy), 27% (10 x 3 Gy), 31% (15 x 2.5 Gy), and 28% (20 x 2 Gy); and posttreatment ambulatory rates were 69%, 68%, 63%, 66%, and 74% (P = .578), respectively. On multivariate analysis, age, performance status, primary tumor, involved vertebra, interval from cancer diagnosis to MSCC, pretreatment ambulatory status, and time of developing motor deficits were significantly associated with functional outcome, whereas the RT schedule was not. Acute toxicity was mild, and late toxicity was not observed. In-field recurrence rates at 2 years were 24% (1 x 8 Gy), 26% (5 x 4 Gy), 14% (10 x 3 Gy), 9% (15 x 2.5 Gy), and 7% (20 x 2 Gy) (P < .001). Neither the difference between 1 x 8 Gy and 5 x 4 Gy (P = .44) nor between 10 x 3 Gy, 15 x 2.5 Gy, and 20 x 2 Gy (P = .71) was significant.
CONCLUSION: The five RT schedules provided similar functional outcome. The three more protracted schedules seemed to result in fewer in-field recurrences. To minimize treatment time, the following two schedules are recommended: 1 x 8 Gy for patients with poor predicted survival and 10 x 3 Gy for other patients. Results should be confirmed in a prospective randomized trial.
INTRODUCTION
Metastatic spinal cord compression (MSCC) occurs in 5% to 10% of all cancer patients during the course of their disease. MSCC is associated with neurologic signs including diminished motor function. If pain is the only clinical symptom or if the diagnosis is based on radiologic studies alone, it should be described as impending MSCC.
Urgent treatment is required for MSCC to avoid progression of motor deficits resulting in paraplegia.1 Radiotherapy (RT) and decompressive surgery are the most important treatment modalities.2
The most appropriate RT schedule is unknown because of the lack of studies comparing various regimens. Because many RT schedules are used worldwide, such as 1 x 8 Gy, 1 x 10 Gy, 2 x 6 Gy, 2 x 8 Gy, 5 x 4 Gy, 6 x 4 Gy, 5 x 5 Gy, 10 x 3 Gy, 14 x 2.5 Gy, 15 x 2.5 Gy, 20 x 2 Gy, and various split-course regimens, there is no standardized RT for MSCC.
A short overall treatment time would be desirable for patient convenience and comfort, especially because the life expectancy of patients with MSCC is generally short and many MSCC patients are debilitated and unable to walk.3 The daily visits to the RT department and the positioning on the treatment couch are associated with significant discomfort. The cost of providing longer RT schedules is greater and is only justified if longer regimens produce better results. A shorter treatment course, although desirable, can only be recommended if it provides similar functional outcome as more protracted RT schedules.
Another important end point is the rate of in-field recurrences, which can require reirradiation of a previously treated region. This end point has not yet been sufficiently investigated in studies of MSCC treatment.
The selection of the treatment schedule is often influenced by prognostic factors. In the literature, the following four relevant prognostic factors have been recognized for functional outcome of RT for MSCC: the length of the time of developing motor deficits before RT, the histology of the primary tumor, the interval from first diagnosis of cancer to MSCC, and the pretreatment ambulatory status.4-12 In addition to these factors, this study evaluated four other potential prognostic factors of age, sex, performance status, and the number of involved vertebra. The potential prognostic factors to be investigated were defined a priori.
This international multicenter study included the largest MSCC patient cohort ever presented and compared five different RT schedules used for MSCC with respect to resulting functional outcome. The primary goal was to investigate the optimal therapy for this common clinical problem.
PATIENTS AND METHODS
A total of 1,304 MSCC patients were included in this retrospective analysis. The data were obtained from the patient files and from the patients' general practitioners. Inclusion criteria were as follows: motor dysfunction of the lower extremities, no previous surgery or RT of the spinal region concerned, no concurrent chemotherapy, and survival of at least 1 month after RT. The latter criterion allows the evaluation of motor function 1 month after RT. This seems more appropriate than the evaluation directly after RT because no treatment effect can be expected directly after single-fraction RT such as 1 x 8 Gy. Before RT was started, the patients were usually presented to a neurosurgeon to discuss the option of decompressive surgery if indicated.
Patients with a history of a brain tumor, brain metastasis, or other major neurologic diseases, which may result in motor dysfunction, were excluded. The diagnosis of MSCC was confirmed by magnetic resonance imaging (MRI) or computed tomography (CT). Patients received dexamethasone at a moderate to intermediate dose level (16 to 32 mg/d) during the whole RT. If 1 x 8 Gy or 5 x 4 Gy were applied, dexamethasone was administered for a week.
The five RT schedules that were compared were as follows: 1 x 8 Gy in 1 day (n = 261), 5 x 4 Gy in 1 week (n = 279), 10 x 3 Gy in 2 weeks (n = 274), 15 x 2.5 Gy in 3 weeks (n = 233), and 20 x 2 Gy in 4 weeks (n = 257). The RT schedules were compared for the following three posttreatment end points: motor function, ambulatory status, and in-field recurrences. The patients were treated between January 1992 and December 2003 and observed for at least 6 months or until death. Each series of the contributing centers represented an unselected group of patients treated in a specific time period. Of the five investigated RT schedules, each center used two to three different schedules, usually related to the treatment capacity on the linear accelerators and not to the patients' performance status or to the type of primary tumor. Parts of the data from Hamburg and Hannover, Germany, and Northwood, United Kingdom, have been included in previously published studies.4,5,13
Irradiation was performed with a linear accelerator (6 to 10 MV) or with a cobalt 60 treatment machine. RT doses were prescribed to the spinal cord using CT scans and MRI. The treatment volume encompassed one to two normal vertebra above and below the metastatic lesions. Irradiation was delivered through a single posterior field or parallel opposed fields depending on the depth of spinal cord.
The median age was 63 years (range, 23 to 89 years). Of the patients, 543 (42%) were female, and 761 (59%) were male. The distribution of the primary tumors was as follows: breast cancer, 335 patients (26%); prostate cancer, 276 patients (21%); lung cancer, 184 patients (14%); lymphoma or myeloma, 138 patients (11%); cancer of unknown primary, 102 patients (8%); gastrointestinal cancer, 95 patients (7%); renal cancer, 71 patients (5%); and other cancers, 241 patients (18%). The median interval from first diagnosis of cancer to MSCC was 24 months (range, 1 to 276 months). Patient characteristics related to the five treatment groups are listed in Table 1. The treatment groups were well balanced for the investigated variables.
Motor function and ambulatory status were evaluated before RT and at 1 month, 3 months, and 6 months after RT. Motor function was graded with a 5-point scale according to Tomita et al14 as follows: grade 0, normal strength; grade 1, ambulatory without aid; grade 2, ambulatory with aid; grade 3, not ambulatory; and grade 4, paraplegia. Improvement or deterioration of motor function was defined as a change of at least 1 point.
The following potential prognostic factors were evaluated with respect to functional outcome: age ( 63 years v 64 years), Eastern Cooperative Oncology Group (ECOG) performance status (1 to 2 v 3 to 4), number of involved vertebra (one to two v three to four v five vertebra), interval from cancer diagnosis to MSCC ( 24 months v > 24 months), ambulatory status before RT (ambulatory/Tomita grade 0 to 2 v nonambulatory/Tomita grade 3 to 414), type of primary tumor (favorable: myeloma, lymphoma, testicular seminoma; v unfavorable: cancer of unknown primary, lung cancer, melanoma; v intermediate: other tumors), and the length of time of developing motor deficits before RT (1 to 7 days v 8 to 14 days v > 14 days; Table 2).
The potential prognostic factors were included in a multivariate analysis that was performed with the ordered logit model, which is also known as the proportional odds model, because the data regarding functional outcome are ordinal (–1 = deterioration, 0 = no change, and 1 = improvement). The comparison of the five RT schedules with respect to functional outcome was also performed with the ordered logit model.
In-field recurrence was defined as clinically significant recurrence of MSCC associated with motor deficits in the preirradiated spinal region. Diagnosis of recurrence was confirmed by spinal CT or MRI. Time to recurrence was measured from the end of RT. The comparison of the five treatment groups with respect to in-field recurrences was performed with the Kaplan-Meier method and the log-rank test.15
Furthermore, potential prognostic factors for survival after RT of MSCC were evaluated including age, performance status, number of involved vertebra, interval from cancer diagnosis to MSCC, ambulatory status before RT, type of primary tumor, time of developing motor deficits before RT, and response to RT. Univariate analysis was performed with the Kaplan-Meier method and the log-rank test,15 and multivariate analysis was performed with the Cox proportional hazards model.
RESULTS
The evaluation of potential prognostic factors for resulting motor function is summarized in Table 2. The comparison of the five RT schedules for improvement, no change, and deterioration of motor function is shown in Figures 1, 2, and 3, respectively, and in Table 3. No significant differences were found between the five groups at each time of follow-up. The results of the multivariate analysis are listed in Table 4.
Ambulatory rates before and after RT were not significantly different for the five treatment groups (Fig 4). Twenty-six percent of the patients (126 of 477 patients) who were not ambulatory before RT regained the ability to walk, 25% (23 of 91 patients) after 1 x 8 Gy, 26% (27 of 104 patients) after 5 x 4 Gy, 26% (31 of 118 patients) after 10 x 3 Gy, 24% (22 of 90 patients) after 15 x 2.5 Gy, and 30% (23 of 76 patients) after 20 x 2 Gy (P = .96).
The proportions of patients with a follow-up of 3 months were 87% (1 x 8 Gy), 85% (5 x 4 Gy), 84% (10 x 3 Gy), 89% (15 x 2.5 Gy), and 89% (20 x 2 Gy); and the proportions of patients with a follow-up of 6 months were 59%, 61%, 58%, 60%, and 68%, respectively. The proportions of patients with a follow-up of 12 months were 33%, 35%, 34%, 37%, and 39%, respectively. Median follow-up times for survivors were as follows: 14 months (range, 3 to 58 months) after 1 x 8 Gy, 15 months (range, 2 to 61 months) after 5 x 4 Gy, 15 months (range, 4 to 38 months) after 10 x 3 Gy, 16 months (range, 4 to 89 months) after 15 x 2.5 Gy, and 14 months (range, 4 to 76 months) after 20 x 2 Gy. Thus, a relevant bias as a result of differences in follow-up by treatment group seemed unlikely.
The results of the Kaplan-Meier analysis for in-field recurrences are demonstrated in Figure 5. Significantly more recurrences occurred after 1 x 8 Gy and 5 x 4 Gy than after 10 x 3 Gy, 15 x 2.5 Gy, and 20 x 2 Gy. Between 1 x 8 Gy and 5 x 4 Gy, no significant difference was observed. There was also no significant difference observed between 10 x 3 Gy, 15 x 2.5 Gy, and 20 x 2 Gy.
The treatment modalities applied at the time of a recurrence and the outcomes are listed in Table 5. Surgery, reirradiation, and chemotherapy resulted in improvement of motor function in seven (85%) of 12 patients, 26 (35%) of 74 patients, and none (0%) of five patients, respectively. Ten patients, who were irradiated with one of the three more protracted regimens, received no further treatment at the time of recurrence and deteriorated. Median follow-up times after recurrence were 8 months (range, 2 to 27 months) in the 1 x 8 Gy group, 7 months (range, 1 to 29 months) in the 5 x 4 Gy group, 4.5 months (range, 2 to 14 months) in the 10 x 3 Gy group, 5 months (range, 1 to 22 months) in the 15 x 2.5 Gy group, and 7 months (range, 1 to 28 months) in the 20 x 2 Gy group.
In the five treatment groups, no relevant acute or late RT-related toxicity was observed. Acute toxicity did not exceed grade 1 according to the National Cancer Institute Common Toxicity Criteria.16 Late toxicity, such as RT myelopathy, did not occur. In the patients surviving 6 months, the median follow-up period was 13 months (range, 6 to 92 months).
The results of the univariate and the multivariate analyses investigating potential prognostic factors with respect to survival after RT of MSCC are listed in Table 6. ECOG performance status, number of involved vertebra, interval from cancer diagnosis to MSCC, ambulatory status before RT, type of primary tumor, time of developing motor deficits before RT, and response to RT were significantly associated with survival, whereas age had no significant impact.
DISCUSSION
MSCC causes neurologic deficits, which are almost always preceded by pain. According to the literature, no particular RT schedule is superior to others with respect to pain relief.17-20 Since two large randomized trials were published in 1999, single-fraction RT with 1 x 8 Gy has been frequently used for painful bone metastasis.18,20 In contrast, the optimum RT schedule for the treatment of motor dysfunction is unclear because of a lack of comparative studies examining functional outcome in MSCC.
This multicenter study is the largest one ever assembled and the first comparing five different RT schedules with respect to subsequent functional outcome in MSCC. The primary goal of this study was to investigate whether RT schedules with a short overall treatment time, such as 1 x 8 Gy in 1 day and 5 x 4 Gy in 1 week, are as effective as longer treatment programs with respect to functional outcome.
The data demonstrated that the five investigated RT schedules were comparably effective for functional outcome and posttreatment ambulatory rates. A similar proportion of patients who were not ambulatory before RT regained the ability to walk in each of the five treatment groups.
However, it has to be taken into account that this is a retrospective analysis, which may be associated with potential biases. The retrospective nature of this analysis did not allow the application of a more differentiated grading system for motor function than the Tomita scale because more detailed information about the functional status was not available in most patients. Appropriate information about the effect of RT on potential sensory deficits or dysfunction of bowel and bladder was also not available. Thus, our analysis focuses on motor function. Another potential bias may have been introduced because of the fact that patients with a survival time of less than 1 month after RT were excluded from the analysis.
Application of 1 x 8 Gy instead of 20 x 2 Gy means a reduction of the overall treatment time from 4 weeks to only 1 day and a reduction of the number of treatment sessions from 20 to only one. Because life expectancy in most MSCC patients is limited to a few months and the majority of the patients are quite debilitated, a shorter treatment program is highly desirable. Each treatment can cause discomfort and inconvenience for the patient during transportation to the radiation department and the positioning on the treatment couch. Furthermore, a longer program substantially increases the cost of therapy.
A prospective study from Germany suggested that a regimen of 10 x 3 Gy is more effective with respect to bone recalcification than a regimen of 1 x 8 Gy.21 Bone recalcification can be expected only several months after RT. Thus, this might be important for patients with a relatively long life expectancy only. Extended survival can be expected most likely for patients with breast cancer, prostate cancer, myeloma, or lymphoma.10,12,22 For these patients, longer fractionated RT schedules could be more appropriate.
According to our results, 1 x 8 Gy is comparable to 5 x 4 Gy with respect to functional outcome. Thus, patients with a poor survival prognosis (eg, < 4 to 6 months), in whom bone recalcification is not important, could be treated with 1 x 8 Gy to keep the overall treatment time as short as possible, resulting in less discomfort for the patient. For patients with a better prognosis (expected survival of at least 4 to 6 months), a more protracted RT schedule should be applied to achieve better bone recalcification and fewer recurrences. Because our results demonstrated a comparable effect on functional outcome for the three more protracted regimens, 10 x 3 Gy should be used for such patients because it is associated with the shortest overall treatment time and the least discomfort for the patient. According to our data, survival after RT of MSCC depends on various prognostic factors such as performance status, number of involved vertebra, interval from cancer diagnosis to MSCC, ambulatory status before RT, type of primary tumor, time of developing motor deficits before RT, and response to RT.
The five groups compared in this study were balanced with respect to the seven prognostic factors examined (Table 1). Therefore, these factors did not have a relevant impact on the dose response. The optimum dose of corticosteroids could not be investigated because patients received comparable doses of dexamethasone during RT. In a randomized trial comparing high-dose dexamethasone (96 mg daily) to no corticosteroids during RT, gait function after RT was obtained in 81% of the patients who had received high-dose dexamethasone compared with 63% of the patients who did not receive corticosteroids.23 A historical case-control series that compared high-dose with moderate-dose (16 mg daily) dexamethasone demonstrated a significantly higher rate of serious side effects for the high-dose regimen (14% v 0%, respectively).24 The patients in the present series received dexamethasone at a moderate to intermediate dose level. There is a need for further investigation to determine the optimum dose of dexamethasone or other corticosteroids.25
Because each series contributed by the participating centers represented an unselected group of patients treated in a specific time period, the potential selection bias was excluded. The administration of the five schedules (two or three schedules per center) was related to the waiting list of patients and not to other factors, such as type of tumor histology or performance status. This makes the possibility of bias unlikely. Between the centers contributing patients, there was no relevant difference regarding RT techniques and the prescription of RT doses. Again, a selection bias seemed unlikely.
In RT oncology, the effect of a RT schedule on tumor control and on late toxicity depends on both the total dose and the dose per fraction. The five schedules can be compared with the equivalent dose in 2 Gy fractions (EQD2), which is calculated by the following equation: EQD2 = D x [(d + /)/(2 Gy + /)], which is derived from the linear-quadratic model; D represents total dose, d represents dose per fraction, represents linear (first-order, dose-dependent) component of cell killing, represents quadratic (second-order, dose-dependent) component of cell killing (more reparable), and / ratio represents the dose at which both components of cell killing are equal.26,27
The / ratio suggested for tumor effect is 10 Gy, resulting in an EQD2 of 12 Gy for the schedule of 1 x 8 Gy, 23.3 Gy for the schedule of 5 x 4 Gy, 32.5 Gy for the schedule of 10 x 3 Gy, 39 Gy for the schedule of 15 x 2.5 Gy, and 40 Gy for the schedule of 20 x 2 Gy. Thus, 15 x 2.5 Gy and 20 x 2 Gy would be expected to be the most effective schedules for tumor effect and functional outcome. However, our results did not reveal a significant difference between the five RT schedules in their effect on functional outcome or ambulatory status.
In addition to the five fractionation schedules, this study investigated seven potential prognostic factors for functional outcome in MSCC. A significant impact was found for the length of time of developing motor deficits before RT, the histology of the primary tumor, the ECOG performance status, the interval from cancer diagnosis to MSCC, the pretreatment ambulatory status, and the number of involved vertebra. The histology of the primary tumor and the pretreatment ambulatory status have been recognized earlier as prognostic factors.6-12 The ECOG performance status has not yet been established as a relevant prognostic factor, but it is correlated with the ambulatory status. Patients who are not able to walk, even with aid, have an ECOG performance status of either 3 or 4. A potential prognostic impact of the number of involved vertebra has been mentioned by Huddart et al22 in a series of 67 prostate cancer patients. The time of developing motor deficits before RT was already suggested to be a relevant prognostic factor in MSCC treatment.4,5 The poor functional outcome after a rapid development of motor dysfunction can be explained by disruption of the arterial blood flow by a rapidly constricted spinal cord, which leads to spinal cord infarction. A slower growing lesion is suggested to first cause venous congestion, which is more reversible.28 Another possible explanation for longer lasting symptoms predicting better functional outcome may be a reflection of tumor biology. More aggressive tumors would be expected to grow faster and cause a faster course of motor dysfunction. The aggressiveness of a tumor may also be reflected by the interval from cancer diagnosis to MSCC, which has been previously described as a potential prognostic factor.7 A longer interval is suggested to predict better outcome. This is in accordance with our results that demonstrated a significant association between the interval from first diagnosis of cancer to MSCC.
Relevant acute or late RT-related toxicity was not observed in our series. The tolerance dose (5% late toxicity within 5 years) for RT myelopathy is considered to be 45 to 50 Gy for conventional fractionation (dose per fraction of 1.8 to 2 Gy).29 If the end point is myelopathy, the EQD2 has to be calculated with an / ratio of 2 Gy, resulting in an EQD2 of 20 Gy for the schedule of 1 x 8 Gy, 30 Gy for the schedule of 5 x 4 Gy, 37.5 Gy for the schedule of 10 x 3 Gy, 42 Gy for the schedule of 15 x 2.5 Gy, and 40 Gy for the schedule of 20 x 2 Gy.26,27 These doses are below the tolerance dose, making late toxicity unlikely. The limited survival associated with MSCC also makes late toxicity unlikely. However, toxicity might be enhanced if concurrent chemotherapy is administered during or immediately after RT. Wallington et al12 reported six (18%) of 34 deaths in a series of lymphoma and myeloma patients to be related to toxicity from chemotherapy. In the present study, patients with concurrent chemotherapy were excluded from the analysis.
The five RT schedules were also compared with respect to in-field recurrences. Recurrences occurred significantly more frequent after 1 x 8 Gy and 5 x 4 Gy than after 10 x 3 Gy, 15 x 2.5 Gy, and 20 x 2 Gy. However, a bias may have been introduced as a result of the fact that more patients may have had in-field recurrences but were not referred for reirradiation because of poor prognosis or the wrong impression that reirradiation is not effective. The reirradiation rate may have also been influenced by the concern about myelopathy, which becomes more of an issue after initial RT with higher doses such as 30 to 40 Gy. This question can only be properly answered in a prospective manner, where all patients who are treated with different RT schedules are assessed the same way at the same time points.
In our series, approximately every fifth patient treated with 1 x 8 Gy, and approximately every 10th patient treated with 10 x 3 Gy, 15 x 3 Gy, or 20 x 2 Gy, needed re-treatment within 1 year after initial RT. Taking into account the EQD2 for late toxicity and the tolerance dose for myelopathy, it seems that reirradiation can be performed more safely after 1 x 8 Gy and 5 x 4 Gy than after a more protracted schedule with a higher total dose (EQD2). This has been considered by the contributing centers. Reirradiation was performed in almost every patient with a recurrence after 1 x 8 Gy and 5 x 4 Gy, whereas only 12 (35%) of the 34 patients initially treated with 10 x 3 Gy, 15 x 2.5 Gy, or 20 x 2 Gy received a second RT treatment. However, in these 12 patients, RT-related myelopathy was not observed after a median follow-up of 8 months (range, two to 29 months). This finding was supported by a recently presented analysis that demonstrated that reirradiation of the spinal cord seemed possible with an acceptable risk of myelopathy, also after initial treatment with 10 x 3 Gy, 15 x 2.5 Gy, or 20 x 2 Gy.30 The most effective treatment for in-field recurrence was decompressive surgery. However, it was only performed in selected patients with only one involved vertebra. The value of surgery in case of involvement of only one vertebra has recently been shown by Patchell et al.2
Our data demonstrated that the five investigated RT schedules were comparably effective with respect to functional outcome in MSCC. The schedule with the shortest overall treatment time (1 x 8 Gy) should be administered to patients with a poor estimated survival (eg, < 4 to 6 months). These short programs are associated with good initial results in terms of improved functionality.
More protracted schedules (10 x 3 Gy, 15 x 2.5 Gy, and 20 x 2 Gy) seem to be associated with fewer relapses, although this finding may have been confounded by a potential bias because of concerns about reirradiation if the primary RT was performed with comparably high total doses. This question needs to be clarified in a prospective manner.
Furthermore, more protracted RT schedules are suggested to result in better recalcification than 1 x 8 Gy and 5 x 4 Gy. Because significant bone recalcification occurs only several months after RT, more protracted schedules can be recommended for patients with a longer expected survival (eg, longer than 4 to 6 months). Because no significant difference was observed between the three more protracted schedules investigated in this study, 10 x 3 Gy would be the best option because it offers the shortest overall treatment time and the lowest costs. This study supports the application of a more standardized RT approach for MSCC by using only two schedules, 1 x 8 Gy or 10 x 3 Gy, depending on the estimated patient survival. Because of the retrospective nature of our study, the results need to be confirmed in a prospective, randomized, multicenter trial. Such a comparative trial is underway in our institutions.
Authors' Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
NOTES
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Poortmans P, Vulto A, Raaijmakers E: Always on a Friday Time pattern of referral for spinal cord compression. Acta Oncol 40:88-91, 2001
Patchell R, Tibbs PA, Regine WF, et al: A randomized trial of direct decompressive surgical resection in the treatment of spinal cord compression caused by metastasis. Proc Am Soc Clin Oncol 22:1, 2003 (abstr 2)
Ampil F: Epidural compression from metastatic tumor with resultant paralysis. J Neurooncol 7:129-136, 1989
Rades D, Karstens JH, Alberti W: Role of radiotherapy in the treatment of motor dysfunction due to metastatic spinal cord compression: Comparison of three different fractionation schedules. Int J Radiat Oncol Biol Phys 54:1160-1164, 2002
Rades D, Heidenreich F, Karstens JH: Final results of a prospective study for the prognostic value of the time of developing motor deficits before irradiation in metastatic spinal cord compression. Int J Radiat Oncol Biol Phys 53:975-979, 2002
Helweg-Larsen S: Clinical outcome in metastatic spinal cord compression. Acta Neurol Scand 94:269-275, 1996
Helweg-Larsen S, Srensen PS, Kreiner S: Prognostic factors in metastatic spinal cord compression: A prospective study using multivariate analysis of variables influencing survival and gait function in 153 patients. Int J Radiat Oncol Biol Phys 46:1163-1169, 2000
Kim RY, Smith JW, Spencer SA, et al: Malignant epidural spinal cord compression associated with a paravertebral mass: Its radiotherapeutic outcome on radiosensitivity. Int J Radiat Oncol Biol Phys 27:1079-1083, 1993
Leviov M, Dale J, Stein M, et al: The management of metastatic spinal cord compression: A radiotherapeutic success ceiling. Int J Radiat Oncol Biol Phys 27:231-234, 1993
Maranzano E, Latini P: Effectiveness of radiation therapy without surgery in metastatic spinal cord compression: Final results from a prospective trial. Int J Radiat Oncol Biol Phys 32:959-967, 1995
Srensen PS, Brgesen SE, Rohde K, et al: Metastatic epidural spinal cord compression. Cancer 65:1502-1509, 1990
Wallington M, Mendis S, Premawardhana U, et al: Local control and survival in spinal cord compression from lymphoma and myeloma. Radiother Oncol 42:43-47, 1997
Hoskin PJ, Grover A, Bhana R: Metastatic spinal cord compression: Radiotherapy outcome and dose fractionation. Radiother Oncol 68:175-180, 2003
Tomita T, Galicich JH, Sundaresan N: Radiation therapy for spinal epidural metastases with complete block. Acta Radiol Oncol 22:135-143, 1983
Kaplan EL, Meier P: Non parametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958
Trotti A, Byhardt R, Stetz J, et al: Common toxicity criteria: Version 2.0—An improved reference for grading the acute effects of cancer treatment: Impact on radiotherapy. Int J Radiat Oncol Biol Phys 47:13-47, 2000
Niewald M, Tkocz HJ, Abel U, et al: Rapid course radiation therapy vs. more standard treatment: A randomised trial for bone metastases. Int J Radiat Oncol Biol Phys 36:1085-1089, 1996
Bone Pain Trial Working Party: 8 Gy single fraction radiotherapy for the treatment of metastatic skeletal pain: Randomised comparison with a multifraction schedule over 12 months of patient follow-up. Radiother Oncol 52:111-121, 1999
Price P, Hoskin PJ, Easton D, et al: Prospective randomised trial of single and multifraction radiotherapy schedules in the treatment of painful bony metastases. Radiother Oncol 6:247-255, 1986
Steenland E, Leer J, van Houwelingen H, et al: The effect of a single fraction compared to multiple fraction on painful bone metastases: A global analysis of the Dutch Bone Metastasis Study. Radiother Oncol 52:101-109, 1999
Koswig S, Budach V: Re-calcification and pain relief following radiotherapy for bone metastases: A randomised trial of 2 different fractionation schedules (10 x 3 Gy vs 1 x 8 Gy). Strahlenther Onkol 175:500-508, 1999
Huddart RA, Rajan B, Law M, et al: Spinal cord compression in prostate cancer: Treatment outcome and prognostic factors. Radiother Oncol 44:229-236, 1997
Srensen PS, Helweg-Larsen S, Mouridsen H, et al: Effect of high-dose dexamethasone in carcinomatous metastatic spinal cord compression treated with radiotherapy: A randomized trial. Eur J Cancer 30A:22-27, 1994
Heimdal K, Hirschberg H, Slettebo H, et al: High incidence of serious side effects of high-dose dexamethasone treatment in patients with epidural spinal cord compression. J Neurooncol 12:141-144, 1992
Loblaw DA, Lapierre NJ: Emergency treatment of malignant extradural spinal cord compression: An evidence-based guideline. J Clin Oncol 16:1613-1624, 1998
Barendsen GW: Dose fractionation, dose rate and iso-effect relationships for normal tissue responses. Int J Radiat Oncol Biol Phys 8:1981-1997, 1982
Joiner MC, Van der Kogel AJ: The linear-quadratic approach to fractionation and calculation of isoeffect relationships, in Steel GG (ed): Basic Clinical Radiobiology. New York, NY, Oxford University Press, 1997, pp 106-112
Manabe S, Tanaka H, Higo Y, et al: Experimental analysis of the spinal cord compressed by spinal metastasis. Spine 14:1308-1315, 1989
Emami B, Lyman J, Brown A, et al: Tolerance of normal tissue to therapeutic irradiation. Int J Radiat Oncol Biol Phys 21:109-122, 1991
Andratschke N, Nieder C, Grosu AL, et al: A risk score for human spinal cord re-irradiation based on clinical data from 40 patients. Radiother Oncol 73:S207, 2004 (abstr 466)(Dirk Rades, Lukas J.A. St)