Radiotherapy in Combination with Systemic Therapy for Multiple Myeloma—A Critical Toxicity Evaluation in the Modern Treatment Era

Simple Summary Radiotherapy is essential for the management of symptomatic osteolytic lesions in multiple myeloma but usually requires a combination with systemic therapy. This study analyzes the acute toxicities of radiotherapy in this setting and demonstrates the feasibility of modern combined modality treatments without significant increases in hematological and non-hematological side effects. However, high-grade leukocytopenia is more frequent following radiotherapy when systemic therapy is given simultaneously. Treatment of five bones or more was associated with a significant increase in thrombocytopenia and leukocytopenia during radiotherapy. Abstract Radiotherapy (RT) is an established treatment modality in the management of patients with multiple myeloma (MM), aiming at analgesia and stabilization of osteolytic lesions. As a multifocal disease, the combined use of RT, systemic chemotherapy, and targeted therapy (ST) is pivotal to achieve better disease control. However, adding RT to ST may lead to increased toxicity. The aim of this study was to evaluate the tolerability of ST given concurrently with RT. Overall, 82 patients treated at our hematological center with a median follow-up of 60 months from initial diagnosis and 46.5 months from the start of RT were evaluated retrospectively. Toxicities were recorded from 30 days before RT up to 90 days after RT. 54 patients (65.9%) developed at least one non-hematological toxicity, with 50 patients (61.0%) showing low-grade (grade I or II) and 14 patients (17.1%) revealing high-grade (grade III and IV) toxicities. Hematological toxicities were documented in 50 patients (61.0%) before RT, 60 patients (73.2%) during RT, and 67 patients (81.7%) following RT. After RT, patients who had received ST during RT showed a significant increase in high-grade hematological toxicities (p = 0.018). In summary, RT can be safely implemented into modern treatment regimens for MM, but stringent monitoring of potential toxicities even after completion of RT has to be ensured.


Introduction
Multiple myeloma (MM) belongs to the group of plasma cell neoplasms and is characterized by the monoclonal proliferation of malignant plasma cells in the bone marrow

Clinical Data
The current study was conceptualized as a monocentric retrospective analysis and approved by the respective institutional review board (Ethikkommission der Ärztekammer Westfalen-Lippe; protocol code 2020-873-f-S; 11 December 2020). Overall, 96 patients treated with RT alone or in combination with ST at our hematological center between 1999 and 2019 with a minimum follow-up of three months were analyzed. Of these, 14 patients were excluded from the analysis due to radiation of a solitary osseous plasmacytoma (7) or due to insufficient follow-up data (7). Clinical data and patient characteristics were extracted from the hospital information system (Orbis, Dedalus Health Care, Bonn, Germany), including documentation of toxicities, doctors' letters, and treatment plans. Details on RT were provided by the RT planning system (Aria, Varian Medical Systems, Pao Alto). Acute toxicities were recorded for a period from 30 days before the start of RT to 90 days after completion and classified using the National Cancer Institute's Common Terminology Criteria for Adverse Events version 5.0 (CTCAE) [21]. Non-hematological toxicities were graded according to the documentation of the treating physicians. Hematological toxicities were graded separately based on available laboratory values, with an evaluation of the last laboratory value before RT, the lowest value during RT, and the lowest value up to 90 days after RT. In cases of insufficient data, general practitioners and radiation oncologists in private practices were contacted to provide further information. The data collection ended in October 2022. Only the first myeloma-related radiation was included in the analysis.  A total of 134 localizations were irradiated, with each patient being irradiated at a median of one site (1-4; see Table 3 for details). The median radiation dose was 40 Gray (Gy; 8-59 Gy). The majority of lesions were targeted using intensity-modulated RT (IMRT) (62 sites, 46.3%) or three-dimensional conformal RT (3D-CRT) (54 sites, 40.3%). The most frequently irradiated sites were the spine (95 sites, 70.9%) and the pelvis (34 sites, 25.4%). Fifty-four patients (65.9%) developed at least one non-hematological toxicity, with no significant difference between the RT group and the RT/ST group (65.5% vs. 66.0%, p = 1.00; see Table 4 for details). Low-grade non-hematological toxicities (grade I or II) occurred in 50 patients (61%), in comparison to 14 patients (17.1%) with high-grade toxicities (grade III or IV). No toxicity-related deaths were observed. The most common non-hematological toxicities were cutaneous (32.9% vs. 1.2% for low-grade vs. highgrade toxicities, respectively), fatigue (22.0% vs. 1.2%), mucosal (13.4% vs. 4.9%), and gastrointestinal toxicities (13.4% vs. 3.7%), with no significant difference in frequency between the RT and RT/ST groups.
All patients with documented laboratory values developed at least one hematological toxicity (73/82 patients, 89%; see Table 5, Table S1 and Figure 1 for more details). Of these, 37 patients (45.1%) also suffered from at least one high-grade toxicity (grade III or IV). There was no significant difference in overall high-grade toxicities between the RT and RT/ST groups (27.6% vs. 54.7%; p = 0.349). While 61.0% of patients (50 patients) already showed at least one hematological toxicity in the period of 30 days before RT, this proportion increased to 73.2% (60 patients) during RT and 81.7% (67 patients) after RT. High-grade toxicities were seen in seven patients (8.5%) before RT, 16 patients (19.5%) during RT, and 31 patients (37.8%) after RT. Despite no significant difference in high-grade toxicities between the RT and RT/ST groups before and during RT (before RT: 6.9% vs. 9.4%, p = 1.00; during RT: 13.8% vs. 22.6%, p = 0.754), a significant increase in high-grade toxicities was observed in the combined-modality group (17.2% vs. 49.1%, p = 0.018) after RT. Table 4. Non-Hematological toxicities given in absolute numbers and percentages (in parentheses). Grading based on the National Cancer Institute's Common Terminology Criteria for Adverse Events version 5.0 (CTCAE) [21]. Low-grade toxicities were defined as grade I or II adverse events; highgrade refers to grade III or IV. No grade V toxicities (death) were observed. For each patient, only the severest grade of toxicity from the respective category was considered.
Hematological and non-hematological toxicities were analyzed for each substance class if data on at least 10 patients was available. No decisive analysis of the use of glucocorticoids was performed, as all patients treated with simultaneous ST also received a glucocorticoid (see Tables 4 and 5 for details). The concurrent use of proteasome inhibitors showed a significant increase in high-grade hematological toxicities after the end of RT (p = 0.027; Figure 2a), especially regarding thrombocytopenia (p = 0.034; Figure 2b). With the use of alkylating agents, there was a significant increase in leukocytopenia during radiotherapy (p = 0.005; Figure 2c). A further significant increase in hematological or non-hematological toxicities during or after RT could not be shown for the different substance classes. Table 5. Hematological toxicities given in absolute numbers and percentages (in parentheses). Grading based on the National Cancer Institute's Common Terminology Criteria for Adverse Events version 5.0 (CTCAE) [21]. Information on hematological toxicities was available for a variable number of patients depending on the accessibility of laboratory values. Percentage values are adjusted relative to the basic population. High-grade toxicities were seen in seven patients (8.5%) before RT, 16 patients (19.5%) during RT, and 31 patients (37.8%) after RT. Despite no significant difference in highgrade toxicities between the RT and RT/ST groups before and during RT (before RT: 6.9% vs. 9.4%, p = 1.00; during RT: 13.8% vs. 22.6%, p = 0.754), a significant increase in highgrade toxicities was observed in the combined-modality group (17.2% vs. 49.1%, p = 0.018) after RT. Hematological and non-hematological toxicities were analyzed for each substance class if data on at least 10 patients was available. No decisive analysis of the use of glucocorticoids was performed, as all patients treated with simultaneous ST also received a glucocorticoid (see Tables 4 and 5 for details). The concurrent use of proteasome inhibitors showed a significant increase in high-grade hematological toxicities after the end of RT (p = 0.027; Figure 2a), especially regarding thrombocytopenia (p = 0.034; Figure 2b). With the use of alkylating agents, there was a significant increase in leukocytopenia during radiotherapy (p = 0.005; Figure 2c). A further significant increase in hematological or non-hematological toxicities during or after RT could not be shown for the different substance classes.  Hematological and non-hematological toxicities were analyzed for each substance class if data on at least 10 patients was available. No decisive analysis of the use of glucocorticoids was performed, as all patients treated with simultaneous ST also received a glucocorticoid (see Tables 4 and 5 for details). The concurrent use of proteasome inhibitors showed a significant increase in high-grade hematological toxicities after the end of RT (p = 0.027; Figure 2a), especially regarding thrombocytopenia (p = 0.034; Figure 2b). With the use of alkylating agents, there was a significant increase in leukocytopenia during radiotherapy (p = 0.005; Figure 2c). A further significant increase in hematological or non-hematological toxicities during or after RT could not be shown for the different substance classes. Dichotomous stratification between <40 Gy and ≥40 Gy revealed an increase in lowgrade non-hematological toxicities in the ≥40 Gy group (p = 0.004), with no dose-side-effect relationship for high-grade non-hematological toxicities or hematological (low-and highgrade) toxicities. Concerning the number of bones irradiated, it could be shown that simultaneous irradiation of five or more bones led to a significantly higher rate of thrombocytopenia (p = 0.019) and leukocytopenia (p = 0.033) during RT (see Figure 3 for details). This effect was not seen regarding the rate of anemia (p = 0.994). Dichotomous stratification between <40 Gy and ≥40 Gy revealed an increase in low-grade non-hematological toxicities in the ≥40 Gy group (p = 0.004), with no doseside-effect relationship for high-grade non-hematological toxicities or hematological (low-and high-grade) toxicities. Concerning the number of bones irradiated, it could be shown that simultaneous irradiation of five or more bones led to a significantly higher rate of thrombocytopenia (p = 0.019) and leukocytopenia (p = 0.033) during RT (see Figure 3 for details). This effect was not seen regarding the rate of anemia (p = 0.994).

Hematological
therapy concurrent with radiotherapy (RT). Grade of toxicities (a) and thrombocytopenia after RT (b) for the patient group treated with proteasome inhibitors (PI) in comparison to patients without. (c) Grade and percentage of leukocytopenia during RT depending on the use of alkylating agents. Error bars present 95% confidence intervals.
Dichotomous stratification between <40 Gy and ≥40 Gy revealed an increase in lowgrade non-hematological toxicities in the ≥40 Gy group (p = 0.004), with no dose-side-effect relationship for high-grade non-hematological toxicities or hematological (low-and highgrade) toxicities. Concerning the number of bones irradiated, it could be shown that simultaneous irradiation of five or more bones led to a significantly higher rate of thrombocytopenia (p = 0.019) and leukocytopenia (p = 0.033) during RT (see Figure 3 for details). This effect was not seen regarding the rate of anemia (p = 0.994).

Discussion
The presented analysis outlines important data on the use of RT as a treatment element for MM in the modern era. It emphasizes that (1) RT is a tolerable and efficient treatment for MM with a low rate of high-grade toxicities; (2) combined-modality treatment may be established safely in most cases; (3) an increase in high-grade hematological toxicities after RT, especially leukocytopenia, was shown for the combined modality group requiring careful laboratory monitoring; (4) large RT fields encompassing five or more bones lead to an increase in leukocytopenia and thrombocytopenia and demand a carefully considered indication.
Overall, both hematological and non-hematological toxicities are common side effects of treatment in MM. Hematological toxicities may be caused not only by MM itself but also by treatment modalities. Depending on the treatment regimen and indication (induction, consolidation, or maintenance), high-grade hematological toxicities were observed in 30% up to 94.9% of patients in the literature [24][25][26][27][28][29][30][31]. Attal et al. described particularly high values for an induction therapy consisting of lenalidomide, bortezomib, and dexamethasone, followed by stem cell therapy and one-year lenalidomide maintenance therapy (94.9%) [24].
This raises the question of whether RT leads to a further increase in toxicities, which has been discussed in the literature with unclear results [17][18][19][20]. Evidence is sparse, with some analyses lacking control groups [20] or only investigating typical radiotherapy-associated toxicities without consideration of potential aggravation of (hematological) toxicities due to systemic therapy [19].
In this regard, all patients in our analysis with documented laboratory values developed at least one low-grade hematological toxicity until 90 days after completion of RT. Concerning high-grade hematological toxicities, 45.1% of patients were affected, with a steady increase over time (8.5% before RT up to 37.8% after RT). This is mirrored by a continued rise in the use of ST (47.6% before RT vs. 64.6% during and 91.5% after RT). While no significant difference in the degree of hematological toxicity could be demonstrated between the RT and RT/ST groups during RT, there was a significant increase in high-grade toxicities in the RT/ST group after the completion of RT.
Non-hematological toxicities were low-grade in most cases, with only a small proportion of patients revealing grade III or IV toxicities without any grade V toxicity up to 90 days after RT. The most common non-hematological adverse events were dermatitis, fatigue, nausea, and gastrointestinal and mucosal toxicities, consistent with previous studies [17][18][19][20] (see Table 6 for an overview). In these publications, the rate of non-hematological toxicities varied between 14.6% and 45.9%, with no evidence for an increase in non-hematological toxicities with the combined modality treatment of RT and ST. In a small cohort of 39 patients, 17 of whom had received at least one novel agent, Shin et al. were unable to show increased rates of non-hematological toxicities for a combination therapy of RT and ST compared to RT alone. Only about 30% of patients developed any non-hematological toxicity, mainly dermatitis, diarrhea, or fatigue [17]. Similar results were shown by Salgado et al. in a cohort of 130 patients, with no significant difference in the occurrence of acute (within four weeks) and subacute (within six months) non-hematological toxicities [18]. In accordance, Guerini et al. failed to reveal significant differences in acute non-hematological side effects (during RT, one month after, and three months after) between RT alone, conventional system therapy, and ST with novel agents in a cohort of 312 patients. In relation to the number of RT series, non-hematological side effects occurred in 41% of cases during RT, which were primarily low-grade (grade I or II) [19]. Regarding radiation treatment, various RT fractionation regimens have been postulated in the literature, differing between a single 8 Gy fraction, hypofractionated regimes, and up to 50 Gy in normofractionation [32][33][34][35]. Comparing these fractionations, previous studies could elaborate on the superiority of a hypofractionated (30 Gy in 10 fractions) RT treatment compared to 8 Gy in one fraction concerning quality of life [33]. Higher doses have been correlated with significantly better pain relief [34,36] and motor function gain after spinal cord compression [35]. Guidelines from the International Lymphoma Radiation Oncology Group (ILROG) recommend the use of hypofractionated regimens with a dose of 8 to 30 Gy for osseous lesions, whereby a single 8 Gy fraction should be considered in patients with poor prognosis [32]. Prompted by the COVID-19 pandemic and limited treat-ment resources [16], recent publications point towards the use of low-dose RT regimens as a feasible treatment alternative. In a retrospective analysis, Price et al. compared low-dose radiation with 12 Gy or less to higher doses and found no significant differences in the duration of analgetic response [37]. A prospective multi-institutional study by the ILROG investigating the use of a 2 × 2 Gy regimen is currently in progress [38].
In this regard, the total dose used in the current analysis is high to achieve optimal recalcification and stabilization. This is supported by Stölting et al., who were able to show a significant increase in recalcification of osteolytic lesions using doses of ≥50 Gy compared to <30 Gy in a cohort of 138 patients [34]. Similar results were shown by Matuschek et al. in a cohort of 69 patients treated with doses of 20 to 60 Gy, in which the use of higher total doses led to a significantly higher rate of recalcification [36]. In view of further studies, total doses of >30 Gy [6,39] also seem to lead to higher recalcification rates than hypofractionated regimens [33].
As combinations of RT and ST are essential to providing long-term responses in MM, new therapeutic agents are constantly developed. Immunotherapies with daratumumab, an anti-CD38 antibody, have been introduced as first-line therapy [28] and are combined with autologous stem cell transplantation in transplant-eligible patients. For therapy-refractory MM chimeric-antigen receptor t-cells targeted against the B-cell maturating antigen or antibodies such as teclistamab represent promising therapeutic perspectives [40,41].
This analysis bears some limitations that are inherent to its monocentric and retrospective character. Consequently, the RT and RT/ST groups were not matched, with imbalances in terms of stem cell conditioning (RT: 6.9%; RT/ST: 20.8%) and worse ECOG (11 of 12 patients with ECOG grade ≥2 in the RT/ST group). Data on comorbidities and MM immunotypes were not collected, which may impair generalization. Strikingly, the proportion of patients who received stem cell transplantation up to 3 months after completion of RT was small (8 patients, 9.8%). Although all RT indications are discussed carefully in our interdisciplinary myeloma tumor board and intended not to impair subsequent (high-dose) therapies, other studies report on a delay in stem cell mobilization and collection by RT [14]. This effect (or a selection bias) cannot be excluded beyond doubt in the present analysis. Recently developed agents like daratumumab have not been considered in our analysis but may lead to new challenges in toxicity management [28].
Additionally, variations in RT concepts and doses are limited, which restricts application to a broader group. Furthermore, no large-field conditioning concepts prior to stem cell transplantation, e.g., total body irradiation (TBI), were examined. In the past, TBI has been used in combination with melphalan but has been shown to be inferior regarding toxicity and hematological recovery, with a trend towards inferior survival in comparison to a high-dose melphalan-only conditioning strategy [42]. Modern approaches with selective targeting of bone marrow (total marrow irradiation) are now being tested in this situation [43,44].
Future studies will elaborate on new, synergistic combinations of RT and ST. With the advent of chimeric antigen receptor t-cells, RT may be implemented both as a bridging and consolidation therapy [45].

Conclusions
Radiotherapy is a safe component of the multimodal therapy of multiple myeloma. The combination of radiotherapy and systemic therapy did not lead to a significant increase in non-hematological or most hematological toxicities. The only exception was a significant increase in leukocytopenia after the completion of radiotherapy. Consequently, careful clinical and laboratory monitoring during and after radiotherapy has to be implemented. The irradiation of five or more bones led to a significant rise in thrombocytopenia and leukocytopenia during radiotherapy. In this treatment situation, a critical evaluation of the size of the target volume has to be performed, along with a discussion of the sequential radiation treatment of one/several target locations.