Hypofractionated palliative volumetric modulated arc radiotherapy with the Radiation Therapy Oncology Group 8502 “QUAD shot” regimen for incurable head and neck cancer

Background : To review a single institutional experience of the Radiation Therapy Oncology Group (RTOG) 8502 “QUAD shot” regimen using volumetric modulated arc radiotherapy (VMAT) for incurable head and neck cancer (HNC). Methods : Thirty-four consecutive patients with HNC were treated with at least one cycle of the RTOG 8502 regimen. Treatment plans included the use of VMAT with 6 MV photons generated by a linear accelerator. Two daily fractions of 3.7 Gy were delivered with an interval of at least 6 h for 2 consecutive days, totaling 14.8 Gy over 4 fractions. This was repeated every 3–4 weeks for a total of three cycles. No concurrent systemic therapy was performed. Results : The number of completed cycles was 1 in 6 (18%) patients, 2 in 5 (15%), and 3 in 23 (68%). Tumor response was achieved in 29 (85%) patients and symptom relief in 20 (77%) of 26 patients. Overall response (tumor response or symptom relief) was achieved in 32 (94%) patients. All patients who received 2 or more treatment cycles achieved overall response. Median overall survival (OS) was 5.7 months. Multivariate analysis revealed that completion of all three treatment cycles was significantly associated with better OS ( P = 0.002). Grade 2 toxicity was observed in four (12%) patients, but no acute Grade ≥3 or late toxicity was observed. Conclusions : The RTOG 8502 “QUAD shot” regimen using VMAT is effective for incurable HNC with highly reduced toxicity. Treatment with multiple cycles is recommended for better treatment response and/or survival. 10 (7%) 8 toxicity, 30 Gy five fractions 6 Gy response achieved in 28 (80%) 2 and 3 respectively, 2, 3 and 4 (66%), 4 (11%) and 2 (6%) respectively. regimens with

concurrent systemic therapy during treatment. The minimum interval between prior RT and reirradiation with the RTOG 8502 regimen had to be 6 months.

Radiotherapy details and technique
Patients were simulated with planning computed tomography (CT) imaging in a dedicated thermoplastic head and neck mask for immobilization prior to each RT cycle. Gross tumor volume (GTV) included primary tumor and lymph node metastases; it is defined based on the fiber scope, contrast-enhanced CT images, and magnetic resonance (MR) imaging with or without [ 18 F]-fluoro-2deoxy-D-glucose (FDG)-positron emission tomography (PET)/CT (12,13). A clinical target volume margin of 5 mm was added to the GTV for subclinical invasion. Planning target volume (PTV) margins of 3 mm were added to cover setup errors. Treatment plans included the use of VMAT (RapidArc; Varian Medical Systems, Palo Alto, CA, USA) with 6 MV photons generated by a linear accelerator (Clinac iX; Varian Medical Systems, Palo Alto, CA, USA). The plan was generated using 1 arc rotating from 181° to 179° clockwise with the dose rate varied between 0 MU/min and 600 MU/min.
RT was delivered using the RTOG 8502 "QUAD shot" regimen. Two daily fractions of 3.7 Gy were delivered with an interval of at least 6 h for 2 consecutive days, totaling 14.8 Gy over 4 fractions. This was repeated every 3-4 weeks for a total of three cycles and a total dose of 44.4 Gy. The goals of the VMAT plans for target volume were defined as follows: dose to 95% of the PTV (D 95 ) > 95% of the prescribed dose, the percentage of the PTV receiving 93% of the prescribed dose (V 93% ) > 99%, the percentage of the PTV receiving 110% of the prescribed dose (V 110% ) < 1%. The maximum allowable dose limit of the spinal cord and brainstem was defined as a total dose of 30 Gy and 50 Gy for the patients without and with prior RT, respectively. The mean dose of each treatment cycle was <5Gy for at least one parotid gland. The dose to the mucous membrane and skin, which were non-adjacent to the target volume was reduced as little as possible. The cone-beam CT (CBCT) scans were acquired using a kilovoltage CBCT scanner and images of CBCT were registered to planning CT for image guidance of each treatment session. For patients with two or more treatment cycles, adaptive radiotherapy (ART) was performed to adjust to their anatomic changes and to avoid overdosing of normal tissues and underdosing or marginal geographic misses of target volumes during a course of treatment by repeating planning CT imaging and replanning for every cycle.

Evaluation of treatment response and toxicity
Tumor response, symptom relief, and toxicity were assessed every 2 weeks until 1 month after the final course of treatment, and every 3 to 4 weeks thereafter until patients died or were no longer able to comply (5). Tumor response was evaluated by physical examination and/or radiographic tumor response, and defined as objective shrinkage of GTV (2, 6). Symptom relief was defined as subjective reduction of the presenting symptom(s) (2,8). Overall response was defined as tumor response or symptom relief. Toxicity was scored by the Common Terminology Criteria for Adverse Events version 5.0. Acute toxicity was defined as occurring up to 3 months after treatment completion (2).

Statistical analysis
Overall survival (OS) and progression-free survival (PFS) rates were calculated from initiation of RT using the Kaplan-Meier method. Candidate variables for prognostic factors of OS and PFS, including age, Eastern Cooperative Oncology Group performance status, tumor site, histology, clinical stage, prior RT at the palliative site, and treatment cycles were evaluated by univariate analysis using logrank statistics. To determine the independent significance of variables, multivariate analyses were performed using the Cox proportional hazards model, by selecting significant variables on univariate analysis. Differences with P-values < 0.05 were considered statistically significant. Statistical calculations were performed using SPSS software, version 26.0 (SPSS Inc., Chicago, IL, USA).

Survival
At the time of analysis, 32 (94%) patients had died. The median follow-up duration was 5.8 (range: 1.0-18.9) months. Median OS was 5.7 (range: 1.0-18.9) months. Univariate analysis showed that clinical stage II-III (P = 0.046) and the completion of all three treatment cycles (P = 0.003) were significantly associated with better OS (Table 3, Figure 1). These two factors remained as independent variables for OS in a multivariate analysis (P = 0.023 and P = 0.002, respectively). Median PFS was 4.4 (range: 0.8-15.9) months. The univariate analysis revealed that only the completion of all three treatment cycles was significantly associated with better PFS (P = 0.045, Table 3, Figure 1).

Toxicity
Overall, Grade 1 acute toxicity was observed in nine (26%) patients, with the most common being mucositis (N = 7) and dry mouth (N = 3). Grade 2 acute toxicity was observed in four (12%) patients and consisted of mucositis (N = 4) and dry mouth (N = 1) ( Table 2). No acute Grade ≥3 or late toxicity was observed. Of the seven patients who had received prior RT at the palliative sites, Grade 1 acute toxicity was observed in one (14%) patient with dry mouth. Grade 2 acute toxicity was observed in two (26%) patients and consisted of mucositis (N = 2) and dry mouth (N = 1).

Discussion
RT for incurable HNC has been demonstrated to be an effective palliative modality, even for patients who have received prior RT (2,14). Currently, no consensus exists for appropriate palliative RT regimen in HNC. In general, a once-daily hypofractionated RT regimen of 30 Gy/10 fractions is commonly performed as palliative RT regardless of the tumor site; however, this treatment regimen is 8 inappropriate for HNC because of the acute adverse effects. The reported frequency of ≥Grade 3 acute toxicity with this treatment regimen for patients with HNC was >40% (6). Other  (16). Tumor response was achieved in 28 (80%) patients. Grade 2 and 3 mucositis were reported in 13 (37%) and 9 (26%) patients, respectively, and Grade 2, 3 and 4 dysphagias were reported in 23 (66%), 4 (11%) and 2 (6%) patients, respectively. These RT regimens provide certain palliative response; however, acute adverse effects that may decrease patients' QOL are still relatively strong. Palliative RT should be considered for relief or prevention of locoregional symptoms; however, severe toxicity should be avoided (9). Our results suggest that RTOG 8502 regimen using VMAT is one of the strongest candidates of palliative RT regimens with good treatment response and low toxicity.
In earlier studies, RTOG 8502 regimen was performed using a two-dimensional (2D) RT. RT field was typically defined as the gross symptomatic disease plus a 1-2 cm margin based on physical examination (4,5). Nowadays, modern diagnostic imaging of MR and FDG/PET-CT, which achieve precise target definition and reduced target volume, has been included in RT planning (12,(17)(18)(19).
Furthermore, the technical development of RT techniques in the last 2 decades, such as 3D-CRT, IMRT, VMAT, ART, and image-guided RT (IGRT) based on CT images, provides an enhanced dose concentration to the target volumes, reduces dose to OARs, and promises precise RT delivery (11,(20)(21)(22)(23)(24). These sophisticated treatment techniques are of significant value for not only definitive RT but also palliative RT with regard to treatment response and toxicity.
The clinical outcome of RT with RTOG 8502 regimen for patients with head and neck tumors is summarized in Table 4 The previous reported symptom response and overall response of the RTOG 8502 regimen with or without systemic therapy approximated 55%-100% and 65%-85%, respectively (4,5,7,8) (Table 4).
Our results of symptom response were comparable to these results. Furtheremore, our results of overall response were superior to these results. RTOG 8502 regimen using VMAT provides appropriate treatment response without using systemic therapy: this treament strategy may improve or maintain patient QOL.
The previously reported median OS and PFS of the RTOG 8502 regimen with or without systemic therapy approximated 4-7 months and 3-4 months, respectively (4,5,7,8) (Table 4). Our results were similar with these results. Considering that the prognosis of patients with HNC who undergo noncurative treatment is poor with approximate survival time of 2-4 months (25,26), palliative RT may contribute to a certin degree of prolonged survival. However, more than half of the patients die within 6 months even if they undergo palliative RT, including RTOG 8502 regimen (4,5,7,8,15,16). Therefore, a smaller number of fractions such as RTOG 8502 regimen is feasible for the palliative RT.
In our series, all patients who received 2 or more treatment cycles achieved overall response.
Furthermore, completion of all three treatment cycles is significantly associated with better OS and PFS, which is consistent with previous reports (2,8). Treatment with multiple cycles is recommended for better treatment response and/or survival.
The incidence of the Grade 2 and 3 toxicities in patients with HNC treated with RTOG 8502 regimen was reported as approximately 20%-40% and 0%-10%, respectively (2, 5, 7, 8) ( Table 4). Our results of toxicities were much lower than those of previous reports. Furthermore, toxicity was acceptable even in the patients who had received prior RT at the palliative sites. The primary reason for highly reduced toxicity in our patients may be attributed to the use of VMAT, ART, and IGRT. Another reason is that we did not perform concurrent systemic therapy. We recommend the introduction of these sophisticated treatement techniques into palliative RT regimen with RTOG 8502 not only because of their excellent palliative response but because of the highly reduced toxicity.
However, this was a retrospective study based on a relatively small number of patients. The potential for selection bias exists, which may influence the results of the treatment outcomes and analysis.
Similarly, we could not evaluate the influence of palliative RT on overall QOL of the patients because our study was retrospective. A prospective trial with a larger cohort should be performed to further evaluate the value of introducing modern sophisticated RT techniques into RTOG 8502 regimen for HNC patients. Further investigations are underway to assess this concern. Conclusions The RTOG 8502 "QUAD shot" regimen using VMAT is effective for incurable HNC with highly reduced  Table   Table 1 Patient characteristics (N = 34)