Dose-escalated radiotherapy with simultaneous integrated boost for bone metastases in selected patients with assumed favourable prognosis

Bone metastases represent one of the most frequent metastatic sites in advanced malignant disease.^1-3 This site is associated with a wide range of symptoms including pain, hypercalcemia, increased risk of pathological fracture and neurological symptoms.^1,2,4,5 Due to their complications, bone metastases can decrease the quality of life in cancer patients.⁶ A significant number of patients with advanced malignant disease present with symptomatic bone metastases.⁷ The analgesic effect of radiotherapy for painful bone metastases has been established for years and therefore irradiation is the preferred treatment for localized bone pain in advanced malignant disease. Approximately 70– 80% of patients will respond with pain relief, up to one-third will achieve complete pain response.¹

Significant progress in systemic and supportive therapy has increased patients’ life expectancy.² Furthermore, beginning with Hellman and Weichselbaum in 1995, the hypothesis of the existence of an oligometastatic state of cancer, as an intermediate stage of cancer spread, has been established⁸ and is nowadays differentiated from widespread metastatic disease. With improvements in diagnostic modalities, oligometastatic malignant disease is being diagnosed more frequently than before⁹, resulting in earlier detection of metastases.¹⁰ However, various definitions and different cut-offs are discussed in the literature. In most studies, oligometastatic state was defined as limited number of metastases, with 1–3 or 1–5 metastatic lesions.^{9, 11, 12} Accumulating clinical evidence suggests that metastasis-directed local therapy for these patients might result in improved clinical response, prevent additional metastatic spread and delay the initiation of systemic therapies.^{13, 14} Adequate radiotherapy regimens to achieve sufficient pain relief have been discussed in the literature and different regimens in the palliative situation have been reported and summarized in various studies.^2,7,15,16 However, the optimal fractionation and dose regimen for patients with oli-gometastatic disease is still an unresolved issue. Considering improved survival for patients with oligometastatic disease, the goal of an aggressive metastases-directed approach is not only to achieve an optimal pain relief, but also long-term local control (LC).

To deliver high doses to the target, maximize targeting capabilities and minimize damage to organ at risk (OAR) or healthy tissue, stereotactic body radiotherapy (SBRT) has been introduced.¹⁷ In a systematic review published in 2019 by Spencer et al., the role of stereotactic radiotherapy in 1-6 fractions in the management of bone metastases from solid-organ tumours was examined. Excellent local control rates, as well as superior rates for pain relief (compared to conventional radiotherapy) were reported in this analysis.³ However, for some bone metastases, stereotactic radiotherapy in a few fractions might be unsuitable, due to their close proximity to OAR or size or limited definability of target volumes. For this reason, many study protocols exclude tumours within a distance of < 3 mm to the spinal cord, with the aim to respect its dose limitations.¹⁸ Various studies examined intensity modulated radiotherapy (IMRT) regimes with simultaneous integrated boost (SIB) for radiotherapy of spine metastases.18^-20 Compared to conventional IMRT, this approach should offer dose reduction in the spinal cord and dose escalation in the target volume using SIB.²¹ In our institution, a higher-dose fractionated regimen for bone metastases with 30 Gy and 40 Gy radiotherapy in 10 fractions with dose escalation by SIB (“30/40 Gy”) to treat patients oligometastatic and oligoprogressive malignant disease was introduced. This regimen enables not only a dose escalation in the target as an alternative to SBRT, but also a coverage of tumour-affected compartment (according to clinical assessment).

The aim of this study was to assess the feasibility concerning completion of treatment, acute toxicity and to evaluate oncologic outcomes after fractionated radiotherapy using this concept for bone metastases in selected patients with assumed favourable prognosis. In addition, dose constraints for palliative radiotherapy of the spine have been adapted to higher-dose radiotherapy.

The study protocol was submitted to the Ethics Committee of the Medical Faculty in our institution and approved in 2020 (990/2020B02). This study represents a single institution retrospective analysis of all consecutive patients treated with this regimen at our institution. Clinical records of all patients treated with radiotherapy of bone metastases with intensity modulated radiotherapy (IMRT) with 30/40 Gy SIB in 10 fractions between 2017 and 2020 in were evaluated. Patients treated with the evaluated regimen were not considered for SBRT, due to close proximity of the tumour to OAR, size or limited definability of target volumes. In most cases, patients included in the study had malignant disease in oligometastatic or oligoprogressive state. However, the evaluated treatment was also offered to patients with diffuse metastatic disease, in case of radioresistant histology (such as pheochromocytoma or renal cell carcinoma) or vertebral-body metastasis with intraspinal component, where improved LC with higher-dose fractionated regimen was desired (due to favourable prognosis and expected efficient systemic treatment). The indication for radiotherapy was mainly not palliative symptom control but local treatment of all macroscopic or progressive tumour localizations. Various definitions of oligometastatic disease have been described in the literature. Foster et al. reports that a definition of ≤3 metastases was used in 12/25 retrospective studies.¹¹ Therefore, for the purpose of this study, we defined an oligometastatic disease as 3 or less extracranial metastases. If patients had locally untreated organ metastases, disease was classified as diffuse metastatic disease. Oligoprogression was defined as progression of 3 or less extracranial metastases under systemic therapy. To determine the number of metastases, the last radiological imaging before radiotherapy was used.

Data were collected retrospectively and abstracted by chart review. Feasibility was defined as conducting radiotherapy without interruption and no toxicity ≥ grade 3 (Common Terminology Criteria for Adverse Events [CTCAE] V 4.0). Due to the retrospective study design, pain response to radiotherapy was evaluated based on clinical records and therefore not graded. Overall survival (OS) and progression-free survival (PFS) were evaluated per patient based on the follow up scans and prostate-specific antigen (PSA) measures (for prostate cancer). OS was defined as the time from the date of the end of radiotherapy to the last contact or death. LC and PFS were defined as the time from the end of radiotherapy to last follow-up or to the diagnosis of local progression for LC and local progression or distant progression for PFS. LC was calculated for each irradiated metastasis. In patients with prostatic cancer, in case of no PSA-elevation and no progression of clinical symptoms (such as pain or neurological symptoms connected to irradiated localization), no radiological imaging was performed during follow-up. PSA-level was used as a measurement to assess LC and PFS in these patients.

IMRT was planned based on a three-dimensional planning CT using 3 mm slice thickness, 4-dimensional-CT (4D-CT) was used for metastases of the ribs. Similar to the regimen described by Guckenberger et al.²⁰, we generated multiple target volumes to receive different doses per fraction and maintain the same number of fractions. Gross tumour volume (GTV), i.e. the macroscopic metastasis, was contoured on the planning CT by the aid (and in most cases co-registration) of diagnostic imaging. Clinical target volume (CTV) included GTV and was delineated depending on the localization: the whole vertebral body for spine, or additional assumed subclinical expansion (e.g. along affected ribs). Planning target volume (PTV) for the spine (PTV30) for the 30 Gy-volume was CTV plus 5 mm margin. Planning target volume (PTV40) was generated with 0-2 mm margin around the GTV for the 40 Gy-volume, depending on the localization with 0 mm next to the spinal cord. For metastases in ribs, GTV was contoured as macroscopic tumour in 4D-CT. Internal target volume (ITV) was generated by the aid of 4D-CT to incorporate all potential locations of the tumour. CTV included GTV and 2–3 mm in craniocaudal extension, as well as the whole affected rib on the metastasis level in transverse plane. Additional 6 mm margins were used on CTV to generate PTV30 for metastases in ribs. For metastases in other non-vertebrae bones (sacral bone, sternum, femur), CTV was generated to involve the whole affected bone for sacral bone and sternum (due to large metastasis-size), as well as assumed sub-clinical expansion along affected long bone. PTV30 was generated with different margins (5–15 mm), dependant on the size of the metastasis and considering positioning inaccuracies. Dose prescription according to International Commission on Radiation Units and Measurements (ICRU)50 was aimed at for the GTV with prioritization of limited dose to the spinal cord. Maximal tolerated dose was 107%. The PTV30 should have been covered with ≥ 90% of the prescribed dose to 98% of the contoured volume (D98) and ≤ 107% of the prescribed dose to 2% of the contoured volume (D2). An example of a treatment plan for radiotherapy with SIB with 30/40 Gy in 10 fractions is demonstrated in Figure 1. Spinal cord was limited to 34 Gy total dose, i.e. 50 Gy equivalent dose (2 Gy) (EQD2), estimated by the linear quadratic model with an alpha/beta = 0.87 Gy for spinal cord, according to QUANTEC.²² An EQD2 of 60 Gy (alpha/beta = 2 Gy) was allowed for metastases localized at the level of the cauda equina. Target volumes were delineated using Monaco planning system, version 5.11.03 or Oncentra Masterplan treatment planning system 4.3 (both Elekta AB, Stockholm, Sweden). Treatment planning (optimization) was performed by the above-mentioned version of Monaco or the inhouse product Hyperion 2.4.5, respectively. Treatment was delivered by 6 MV Elekta linear accelerators and image-guided radiotherapy (IGRT) with positioning controls using cone-beam CT and daily online corrections. No specific patient immobilisation was needed, due to daily IGRT controls and corrections.

Figure 1

Additionally, dose constraints for OAR and radiotherapy data were evaluated for all irradiated metastatic sites. Mean values for the volume, D2% (D2) and D98% (D98) for GTV40, CTV30, PTV40 and PTV30 were evaluated. Dose values for spinal cord were calculated for patients with metastases in the vertebral body. Mean value for the maximal point dose in the spinal cord (Dmax), D2 and D0.5ccm were reported. The mean dose (Dmean) values were analyzed for kidneys. Patients with Dmean for kidneys below 1 Gy were excluded from this part of the analysis (target volumes far away).

Statistical analysis was performed with IBM SPSS Version 26. Means were compared by two-sided Student’s t-test. Survival times were examined using Kaplan-Meier estimator and compared using the log-rank test. Chi-square test was used to describe correlations between categorized variables. Significance was considered in case of p < 0.05 and 0.05 < p < 0.1 was defined as a trend to statistical significance. Pearson´s correlations coefficient was used to measure the statistical relationship between two continuous variables. Pearson`s correlation coefficient 0.4 < r < 0.7 was defined as moderate correlation, coefficient ≥0.7 was defined as a strong correlation.

Sufficient LC-rates were demonstrated with the evaluated regimen in our cohort, with LC at 1 and 2 years of 90.0 ± 6.7% and 83.3 ± 15.2%, respectively. SBRT regimen with SIB for patients with spinal bone metastases have been increasingly studied^18-21, and although inclusion criteria and dosing varied between studies, our 1-year LC-rate is in line with reported data.¹⁹ In a prospective study published by Guckenberger et al., spinal metastases were irradiated with SBRT regimen with SIB with either 48.5/30 Gy or 35/20 Gy in 10 fractions.²⁰ Lubgan et al. reports good LC-rates after irradiation of spinal metastases using various SBRT regimen with SIB (median dose of 42.0/ 32.39 Gy in 10– 12 fractions.¹⁹ Comparable to our data, both studies reported good feasibility and no radiation-induced myelopathy as long-term side effect.²⁴

Although radiotherapy for the treatment of painful bone metastases has been established²⁵, optimal fractionation and dose regimens for patients with oligometastatic disease seem to remain challenging and are still an unresolved issue. Various fractionation and dose schedules for palliative radiotherapy for bone metastases have been examined and can be divided broadly into two categories: short-course radiotherapy (delivered in up to five fractions) and long-course radiotherapy (delivered in 10 or more fractions).²⁶ Different studies found no difference in pain relief^{26, 27} or toxicity rates between short-course and long-course therapies.²⁶ However, conventional radiotherapy with 8 Gy single dose is associated with shorter pain relief (3–6 months) and can be insufficient for patients with longer life expectancy.²⁴ Furthermore, accumulating clinical data suggest better local control rates after irradiation of bone metastases with long-course radiotherapy. Improved 1-year local control rates for spinal metastases with spinal canal compression in patients with breast and prostate cancer were reported after long-course radiotherapy, compared to short-course palliative radiotherapy with 8 Gy in one fraction.¹ In addition, the incidence of repeated irradiation to the same metastatic site is lower in patients treated with longer fractionated schedules, compared to patients treated with 8 Gy in one fraction.^{25, 27}

With improved survival rates for patients with oligometastatic disease and predominant bone metastases²⁶, aggressive metastasis-directed therapy has been proposed to improve clinical response and eventually delay the initiation of systemic therapies.^{13, 14, 28} In a study published in 2011 by Rades et al., improved local control, as well as survival benefit were demonstrated for patients with favourable survival prognoses after radiotherapy with total dose escalated beyond 30 Gy (40 Gy in 20 fractions or 37.5 Gy in 15 fractions).²⁹ A fractionated regimen with SIB (to escalate the dose in the target volume and reduce the dose for organs at risks) for radiotherapy of spine metastases was examined in various studies.^{18, 19} Various regimen for palliative radiotherapy in patients with spinal bone metastases are being evaluated in one ongoing prospective study (30 Gy in 10 fractions, 30/40 Gy in 10 fractions, 20 Gy in 5 fractions and 20/30 Gy in five fractions).²¹ To increase the duration of pain relief, achieve better local control, deliver higher dose to the target volume with proper sparing of organs at risks, a higher-dose IMRT fractionated regime with 30/40 Gy with SIB was introduced in our institution. This regimen differs from stereotactic radiotherapy not only in its dose, but also in target volume delineation concept, as it integrates two target volumes (macroscopic tumour and localized adjuvant region within the affected bone).

We included patients with favourable prognostic factors (e.g. number of metastases, systemic treatment options) and assumed longer life expectancy. Oligometastatic disease and oligoprogression under systemic therapy were the indication for this regimen for most of the patients (79.2%). Patients with diffuse metastatic disease were included in case of assumed radioresistant disease or vertebral metastasis with intraspinal component (if the patients had favourable prognosis and efficient systemic therapy options), where the higher-dose regime was applied to achieve better local control. This assumption was supported by a systematic literature review published in 2009 by Gerszten et al. They defined tumour histology as a prognostic factor in treatment response after conventional radiotherapy of spine metastases.³⁰

Although the reported dataset is limited with number of patients and limited follow up, we observed sufficient LC rates using this regime. However, 83.4% of the patients received systemic therapy directly before, in parallel to or after radiation therapy, which might have influenced our LC-rates with its synergistic effect. In comparison to our results, a retrospective study published by Makita et al. reported a 1-year LC-rates of 60% for biological effective dose (BED)₁₀ < 39.0 Gy (= 1 x 8 Gy, 5 x 4 Gy, 4 x 5 Gy or 10 x 2.5 Gy) and 80% for BED₁₀ = 39.0 Gy (= 10 x 3 Gy).³¹ Two patients in our analysis developed local progression. In both cases, local progression seems to be limited to area with sufficient target volume coverage. These cases included radioresistant tumour histology (clear cell renal carcinoma and urothelial carcinoma) which indicates that this regimen should be evaluated in larger series for patients with radioresistant malignancies. Furthermore, feasibility of this regimen was good, with all patients completing the treatment and no patients developing acute toxicity beyond grade 1. Grade 1 acute toxicity was documented for 36.0% of the patients and included mild urinary or gastrointestinal toxicity, dysphagia and nausea. Assuming the extended life expectancy for most patients with oligometastatic disease, late side effects are much more clinically relevant than acute toxicity. Thus, one patient was operated due to painful bone instability 4 months after the end of radiation therapy. No pathological fractures and no neurologic toxicity were observed. However, these results might be limited with absence of imaging during follow-up in some patients with prostatic cancer, where no imaging was performed in case of no PSA-elevation and no progression of clinical symptoms connected to irradiated localization. We adjusted dose constraints for spinal cord using our institutional constraints for normofractionated radiotherapy for vertebral body. Respecting the dose limitation for spinal cord was priority, which led to underdosage in target volume coverage in selected cases. Thus, with this approach, clinically satisfying results were achieved regarding late neurologic toxicity as well as LC-rate.

One patient with urothelial carcinoma and metastasis in distal femur required surgery due to bone instability (rated by orthopaedic surgeons as instability due to the metastasis and not as radiotherapy-induced osteonecrosis). In our analysis, no exceed in ICRU recommendations in radiation plan for this patient was observed. This patient had the largest PTV30 in the whole patient cohort. However, this was the only metastasis in a long bone and is therefore hardly comparable to the spine and ribs volumes (where no osteonecrosis or pathological fractures were detected). This indicates that a further evaluation of this regimen for metastases therapy in long bones is needed, as there might be additional factors to be considered in radiation therapy planning for this localization (e.g. functional load). Pain as an initial symptom was reported in 13/25 patients. Pain relief was documented for 69.2% patients at some point during follow-up, which is comparable to another study that examined pain response after IMRT with 30 Gy in 10 fractions for spinal metastases.³² However, due to the retrospective study design, no pain grading or accurate analyses of pain relief duration was possible.

In summary, higher-dose IMRT fractionated regimen with 30/40 Gy with SIB is a safe and feasible treatment regimen for selected patients with bone metastases, with all patients completing all therapy sessions with no acute radiation toxicity > grade 1. With limited number of patients and follow-up, as well as methodological limitations of a retrospective study, good LC-rates were demonstrated in our cohort. Using this treatment method, we managed to deliver a high radiation dose to the target volume and simultaneously achieve proper sparing of organs at risk. This intermediate-dose regimen represents a therapy in between clear palliative schedules and stereotactic body radiation therapy (SBRT) in few fractions and might be the preferred option for patients with oligometastatic or oligoprogressive disease and long-life expectancy, if SBRT cannot be applied. Furthermore, this treatment can be convenient for bone metastases with intraspinal component, when improved LC-rate might be achieved using this higher-dose fractionated regime. However, late toxicity after this treatment concept and special combinations of metastasis localization and histology warrants further evaluation.