Multi-institutional prospective observational study of radiotherapy for metastatic bone tumor

Abstract Purpose of this study is to evaluate patient characteristics, treatments and outcomes in bone metastasis radiotherapy practice. Patients for whom radiotherapy for bone metastasis was planned at 26 institutions in Japan between December 2020 and March 2021 were consecutively registered in this prospective, observational study. Study measures included patient characteristics, pain relief, skeletal-related events (SREs), overall survival and incidence of radiation-related adverse events. Pain was evaluated using a numerical rating scale (NRS) from 0 to 10. Irradiated dose was analyzed by the biologically effective dose (BED) assuming α/β = 10. Overall, 232 patients were registered; 224 patients and 302 lesions were fully analyzed. Eastern Cooperative Oncology Group Performance Status was 0/1/2/3/4 in 23%/38%/22%/13%/4%; 59% of patients had spinal metastases and 84% had painful lesions (NRS ≥ 2). BED was <20 Gy (in 27%), 20–30 Gy (24%), 30–40 Gy (36%) and ≥ 40 Gy (13%); 9% of patients were treated by stereotactic body radiotherapy. Grade 3 adverse events occurred in 4% and no grade 4–5 toxicity was reported. Pain relief was achieved in 52% at 2 months. BED is not related to pain relief. The cumulative incidence of SREs was 6.5% (95% confidence interval (CI) 3.1–9.9) at 6 months; no factors were significantly associated with SREs. With spinal lesions, 18% of patients were not ambulatory at baseline and 50% of evaluable patients in this group could walk at 2 months. The 6-month overall survival rate was 70.2% (95% CI 64.2–76.9%). In conclusion, we report real-world details of radiotherapy in bone metastasis.


INTRODUCTION
Radiotherapy plays an important role in the treatment of metastatic bone tumors.In addition to pain relief, radiotherapy is also used to improve symptoms in complicated cases such as patients with paraplegia due to spinal cord compression, to prevent bone events such as pathological fractures and paralysis, and to provide long-term local control in oligo-metastases.Numerous randomized controlled trials have demonstrated the efficacy of single irradiation for pain relief [1].However, while systemic therapy significantly improves survival, the endpoints of these trials are often limited to pain relief at 1 to 3 months and complicated lesions are often excluded, creating a gap that physicians should comprehensively consider in practice.In fact, reports have described that radiotherapy is not an option for single irradiation in a variety of scenarios [2].There is a paucity of data showing what kind of radiotherapy is given to what kind of subjects in actual clinical practice, and limited data have been reported on prospective studies of clinical outcomes.Therefore, we planned to conduct a multicenter prospective observational study mainly at Japanese Society for Radiation Oncology-accredited centers to clarify the actual conditions of radiotherapy for patients with bone metastases in Japan and the efficacy of the treatment; Clinical Trial Registration number: UMIN000042491.

METHODS
Patients who underwent radiotherapy for bone metastases at 26 centers throughout Japan between December 2020 and March 2021 were included.Patients who met eligibility criteria were consecutively and prospectively enrolled until 10 cases were enrolled at each institution.Patients were followed for 6 months from the date of enrollment of the last case.The eligibility criteria were as follows: written consent to participate in the study had been obtained from the patient and radiation therapy was planned for metastatic bone tumor.Patients who were deemed by their physicians to be unsuitable for participation in the study were excluded.
Radiotherapy was administered at each facility as clinical practice, while the dose fractionation and method were left up to the treating physician.Adjuvant and supportive care were provided by the patient's physician as needed, and there were no restrictions on post-treatment cancer therapy.We collected the following information at the end of irradiation: age, sex, and Eastern Cooperative Oncology Group Performance Status (ECOG-PS) as patient background at the time of enrollment and disease information, including the name of the primary disease, whether the primary tumor was under control, the site of the lesion other than that to be irradiated, the history of treatment for the primary disease, the Surgical Instability Neoplastic Score (SINS) [3] and Bilsky score [4] in the case of spinal metastases, and Mirels score [5].We also collected information on radiotherapy, including irradiation method and dose, pre-treatment pain score (numerical value) as a symptom-related variable, elapsed time (numerical value) since the onset of pain due to radiotherapy lesions at the time of registration, and whether the patient had been reviewed by a cancer board specific for bone metastases.Follow-up data were collected on the following items: pain, presence or absence of bone events, ambulatory status and survival status at 2 and 6 months after the start of treatment.Skeletal-related events (SREs) were defined as pathological fracture, spinal cord paralysis, surgery on the bone and re-irradiation (if applicable, presence and date of each).Adverse events were evaluated using the Common Terminology Criteria for Adverse Events v5.0 (CTCAE v5.0).Quality of life (EQ-5D-5L, EORTC-PAL15 and BM22) and employment status were also investigated, but an analysis of employment status has already been reported elsewhere [6], and QOL is currently being analyzed and is not included in this report [7].Numerical rating scale (NRS) score of pain at the region treated with radiation therapy anchored at 0 and 10 was recorded.Patients were asked to score the NRS to reflect the worst pain they had experienced within the last 3 days.Narcotic doses were converted to oral morphine doses.Patients who had experienced pain with an NRS of ≥2 at the start of treatment were evaluated for pain at 2 and 6 months after radiotherapy.Pain relief was defined as a decrease in NRS of ≥2 or a decrease in opioid use of ≥25% compared with that at baseline, and complete pain resolution was defined as NRS 0 without an increase in opioid use.An increase in NRS of ≥2 or an increase in opioid use of ≥25% was defined as worsening, and an absence of any of the above was defined as an unchanged status [8].
The study was performed in compliance with the tenets of the Declaration of Helsinki and in accordance with the explanation provided to the participants.

Statistics
Categorical variables were summarized by count and percentage, and continuous variables by mean, median, standard deviation and range.Association between factors and pain relief was investigated by Fisher exact test.Cumulative probability of the bone-related events and their risk factors were investigated by the competing risk analysis [9].Death was treated as a competing risk.Gray test was used to compare cumulative probabilities.Survival analysis with Kaplan-Meier method and Cox proportional hazard regression was conducted to investigate overall survival.Factors included in multivariate Cox regression analysis were chosen from known risk factors.Time to SREs and overall survival time were defined as the time from enrollment to occurrence of SREs and death, respectively.Two-sided P-values <0.05 were considered to be statistically significant.Analyses were performed with the use of the R statistical package, version 4.2.2 (R Core Team [2022], www.r-project.org).

Patient characteristics and treatment
Consort of this study was already reported [6].Briefly, there were 333 patients with bone metastasis referred for radiation therapy in 26 centers during December 2020 to March 2021.Among them, 224 patients were analyzed in the study described in this report.The 2month follow-up was conducted in 186 patients and the 6-month follow-up in 131 patients.
Patient characteristics, radiotherapy method and dose and cancer treatment after radiotherapy are summarized in Table 1.Overall, 60% of the patients were in good general condition with ECOG-PS 0-1, while 16% had a decrease in ECOG-PS to 3-4, which included patients with widely varying general conditions at the time of enrollment.In terms of primary disease, lung cancer and breast cancer accounted for 50% of the total.Regarding irradiation site, 52% of all cases were in the spine; painful bone metastases with NRS ≥2 accounted for 84% of all cases.
In terms of the irradiation method, 9% of the patients were treated with high-precision irradiation such as stereotactic body radiotherapy (SBRT) and intensity-modulated radiation therapy (IMRT), while the remaining 91% were treated with conventional irradiation.Overall, 10% of the patients were irradiated as re-irradiation.Of the 26 centers, 14 (54%) had at least one case treated with high-precision irradiation such as SBRT or IMRT and 13 (50%) had at least one re-irradiation case.Twenty-four (11%) of the cases were reviewed by a cancer board dedicated to bone metastases.
For observational studies, actual radiotherapy was performed in various dose fractions, and the dose was evaluated using the biologically effective dose (BED10) with α/β = 10 [10].The doses of 8 Gy/1 dose, 4 Gy/5 doses and 3 Gy/10 doses frequently used for bone metastases were 14.4, 28 and 39 Gy, respectively.The median BED by site was 28 Gy, and 84% of all patients underwent treatment with BED of <40 Gy using the conventional method.Among patients with an ECOG-PS of 3-4 at enrollment, 61% received BED of <30 Gy with the conventional method, while among patients with an ECOG-PS of 0-1, 41% received BED of <30 Gy with the conventional method (P = 0.08, Supplemental Table 1).The results indicated that irradiation method and dose fractionation were tended to be selected according to PS at the time of enrollment.

Pain relief
Pain was evaluated at 2 and 6 months after radiotherapy in 189 patients who had pain rated as NRS ≥2 at the start of treatment.Overall, 144 patients were evaluable at 2 months and 95 patients at 6 months.Of the evaluable patients, pain resolved or decreased in 75 (52%) at 2 months and in 55 (58%) at 6 months.Pain completely resolved in 32 patients (22%) after 2 months and 31 patients (33%) after 6 months.Pain was analyzed by dividing subjects according to baseline pain levels into mild, moderate and severe pain, which showed a pain-reducing effect regardless of the intensity of baseline pain (Table 2).The transition of pain score is shown in Fig. 1.Among 189 patients who had NRS ≥2 at the start of treatment, mean score (95% confidence interval;CI) was 6.2 (5.8-6.5),2.5 (2.1-2.9) and 1.7 (1.3-2.1) at baseline at 2 and 6 months, respectively.Among 95 patients who were evaluable 6 months after initiation of the radiation therapy, mean score (95%CI) was 5.5 (5.0-6.0),2.0 (1.5-2.4) and 1.7 (1.3-2.1) at baseline at 2 and 6 months, respectively.In patients with pain rated as NRS ≥2, systemic chemotherapy or hormone therapy was found to be associated with pain relief at 2 months, but no other factors showed any association (Table 3).Of the 17 patients who had pain of NRS ≥2 and were treated with re-irradiation, 10 (59%) had pain reduction or resolution after 2 months.Of the 127 patients who were not re-irradiated, pain had been reduced or eliminated in 65 (51%) after 2 months.There was no statistically significant difference between the two groups in this regard.Eighty-seven patients had severe pain with an NRS of 7 or greater at baseline.Of these, 63% were using opioid analgesics.At baseline, 100 of 224 patients (45%) were using opioids, but, at 2 and 6 months, 82 of 186 patients (44%) and 43 of 130 patients (33%), respectively, were using opioids.Of these 100 patients, 68 were evaluable for pain after 2 months.Of these, 26(38%) had a 25% or greater reduction in opioid use, and 25 had eliminated or reduced pain (Supplemental Table 2).There was no significant difference in pain relief (CR + PR) or CR (data not shown) between SBRT and conventional RT in this study.

Skeletal-related events (SREs)
Of the 224 patients analyzed, SREs (spinal cord paralysis, pathological fracture, surgery at the irradiated site or re-irradiation) occurred in 14 patients (Supplemental Table 3), with 60-and 180-day cumulative probabilities of occurrence of 2.4% (95% confidence interval (CI) 0.3-4.4%)and 6.5% (95% CI 3.1-9.9%),respectively (Fig. 2).Re-irradiation was performed in eight patients (3.6%).Unfortunately, data have not been collected on which site SREs occurred when they occurred in patients who were irradiated in multiple sites.Therefore, if an SRE was reported during the follow-up of a patient who had been irradiated in the spine or extremities, the event was calculated as if it had occurred in each case.Treatment for spinal metastases was performed in 132 cases.Of these, 42 patients (32%) had Bilsky grade 2-3 spinal cord compression of the tumor.Of the 25 patients who were ambulatory before radiotherapy, 19 were ambulatory 2 months later, 1 was not ambulatory, and 5 were not evaluable.Of the 17 patients who were not ambulatory before radiotherapy, 7 recovered their walking ability at 2 months, 4 did not and 6 could not be evaluated (Table 4).

Survival period
Overall survival, assessed by the Kaplan-Meier method, did not reach the median, with 2-and 6-month survival rates of 90.2% (95% CI 86.3-94.3%)and 70.2% (95% CI 64.2-76.9%),respectively (Fig. 3a).Survival time differed according to baseline PS (P < 0.001) (Fig. 3b).Univariate and multivariate analyses of survival showed that the poor prognostic factors were poor PS, regional lymph node metastases, bone metastases except for irradiated bone and the primary tumor being hepatobiliary, pancreas or colorectal, while the good prognostic factor was the primary tumor being breast cancer (Table 5).

Adverse events
Adverse events were generally mild, with grade 3 lymphopenia occurring in five patients (2%), and fracture, ileus, pharyngeal mucositis and dysphagia in only one patient (0.4%) each.

DISCUSSION
This study involved the prospective collection of data on radiotherapy for bone metastases, revealing the reality of treatment in the real world.Although the main role of radiotherapy for bone metastases is undoubtedly pain palliation, in some cases physicians administer radiotherapy with the intention of local control and reduction of bonerelated events in addition to pain palliation.Overall, 84% of the patients in this study had painful bone metastases with NRS ≥2, indicating that the majority of patients were treated for pain relief.
The pain response rate for evaluable patients based on ICPRE is reported to be 60% [11], and this study also showed 52% pain relief at 2 months, with the pain-relieving effects replicating those previously reported [1,11].The present study was an observational study with wide-ranging eligibility criteria, and patients with various systemic conditions and complicated cases were also enrolled.Nevertheless, the pain-relieving effects reported in clinical trials were reproduced, indicating that the pain-relieving effects of radiotherapy are broadly applicable to patients with painful bone metastases.The pain-relieving effect after 2 months was better in patients who received post-irradiation systemic drug therapy (Table 3).To the best of our knowledge, there are no reports suggesting that systemic therapy after palliative radiotherapy may be effective in relieving pain from bone metastases.Although it is not intended to have a therapeutic effect on bone metastases, it may result in an enhanced antitumor effect on metastatic lesions and might provide pain relief by shrinking osteolytic lesions and promoting bone remodeling at the site of irradiation.This study showed that radiotherapy can be expected to provide pain relief even for bone metastases with high baseline pain (Table 2).Meanwhile, 44% of patients complained of pain from lesions that were not the radiotherapy target lesions, and more than half of these complaints involved tumor-related pain (Table 1).In these cases, radiotherapy alone is not sufficient to relieve pain, indicating that analgesics should be used in combination with radiotherapy.Although reports have been published stating that SBRT is more effective in relieving pain than conventional RT [12,13], there was no significant difference in pain relief between SBRT and conventional RT in this study.
In this study, there were cases in which radiotherapy was applied to prevent the occurrence of symptoms and bone-related events.A recently reported randomized phase II trial of RT for asymptomatic In the cohort in this study, no SRE occurred in cases of oligometastases or in cases treated with SBRT or IMRT.Long-term follow-up data on spinal SBRT also indicate better local control with SBRT [16], and appropriate selection of cases is warranted.Some SBRT cases had BED of <50 Gy.The risk of recurrence may be a concern with a longer follow-up.It is hoped that the dose will be optimized in the future as guidelines for SBRT become more widely available.
To make appropriate clinical judgments, it is desirable for cases to be reviewed by a cancer board when the expected prognosis is long, when multidisciplinary treatment such as surgery and systemic therapy is necessary, and when high-precision treatment such as SBRT or IMRT is being considered.In the current study, only 11% of cases with bone metastases were actually reviewed by the cancer board, which should be improved in the future.
There are several limitations to this study.First, although data were collected from 26 facilities throughout Japan, all of them were boardcertified facilities, which means that selection bias of the facilities could not be eliminated.In addition, the number of cases was limited because only 10 consecutive cases of radiotherapy for bone metastases were enrolled during the study period due to the burden on participating institutions.Another limitation is that, although follow-up data were obtained prospectively, follow-up data could not be obtained when the follow-up observation at the facility where the treatment was performed had been completed.In addition, although the 6-month survival rate was 70%, longer-term follow-up was not possible.
We believe that the findings of this study are valuable not only for pain relief, but also for prospective and comprehensive evaluation of radiotherapy for bone metastases.The effect of pain relief for bone metastases is well known and consisted with this study, but since it has been shown that SRE is less common in irradiated cases, it will be possible to explain less occurrence of SRE to patients based on this evidence.In conclusion, this work reveals that radiation oncologists are comprehensively evaluating patients with bone metastases and determining the actual radiotherapy regimen.The efficacy in relieving pain and preventing SRE shown in this prospective observational study was promising, indicating high efficacy and low adverse events associated with radiotherapy.These results may be useful for patients with bone metastases and physicians in making treatment choices in clinical practice.We believe that the prospective clinical data obtained here can be used as a benchmark for new intervention studies.

Fig. 1 .
Fig. 1.NRS score of pain at the region treated with radiation therapy.The bars represent mean and its 95% CI.(a) Pain scores in patients who were evaluable before treatment and at 2 and 6 months after initiation of the radiation therapy.(b) Pain score in patients who were evaluable 6 months after initiation of the radiation therapy.CI = confidence interval.

Table 1 .
Patient characteristics and treatment summary

Table 2 .
Pain relief for patients who experienced pain of NRS ≥2 at the start of treatment NRS = numerical rating scale Percentages were calculated using the number of evaluable cases at that time as the denominator.

Table 5 .
Risk factors for overall survival