Radiotherapy Alone Versus Concurrent or Adjuvant Chemoradiotherapy for Nasopharyngeal Carcinoma Patients with Negative Epstein–Barr Virus DNA after Induction Chemotherapy

Simple Summary Induction chemotherapy (IC) plus concurrent chemoradiotherapy has been recommended as the standard treatment for locoregionally advanced nasopharyngeal carcinoma (LA-NPC). However, concurrent chemotherapy was associated with increased toxicities, poor tolerance, and low completion rates. The aim of this study was to compare the efficacy and toxicity of IC + radiotherapy (RT) and IC + concurrent or adjuvant chemoradiotherapy (IC + CCRT/AC) in patients with negative post-IC EBV DNA. The results showed that IC + RT alone displayed similar efficacy to IC + CCRT/AC. The omission of concurrent or adjuvant chemotherapy did not increase locoregional or distant failure. However, patients treated with IC + RT had fewer acute toxicities than those with IC + CCRT/AC. Our finding provided evidence that the omission of concurrent or adjuvant chemotherapy may be feasible for patients with negative EBV DNA after induction chemotherapy. Abstract The purpose of this study was to compare the efficacy and toxicity of induction chemotherapy (IC) plus radiotherapy (RT) and IC plus concurrent or adjuvant chemoradiotherapy (CCRT/AC) in nasopharyngeal carcinoma (NPC) patients with negative Epstein–Barr virus DNA (EBV DNA) after IC. A total of 547 NPC patients with negative plasma EBV DNA post-IC were included. Patients were classified into the IC + RT group and the IC + CCRT/AC group. Locoregional relapse-free survival (LRFS), distant metastasis-free survival (DMFS), overall survival (OS), and progression-free survival (PFS) were estimated and compared using the Kaplan–Meier method. Propensity score matching (PSM) was performed to balance the variables. The median follow-up time was 37 months. The 3-year LRFS, DMFS, OS, and PFS rates for the whole group were 92.2%, 92.4%, 96.4%, and 84.4%, respectively. There was no significant difference in LRFS, DMFS, OS, and PFS between the IC + RT and the IC + CCRT/AC groups, both before PSM (3-year rates of 91.1% vs. 92.6%, p = 0.94; 95.6% vs. 91.5%, p = 0.08; 95.2% vs. 96.8%, p = 0.80; 85.9% vs. 84.0%, p = 0.38) and after PSM (90.7% vs. 92.7%, p = 0.77; 96.8% vs. 93.7%, p = 0.29; 94.5% vs. 93.9%, p = 0.57; 84.7% vs. 85.6%, p = 0.96). Multivariate analysis demonstrated that the treatment schedule was not an independent predictor for survival rates. Patients in the IC + RT group had fewer treatment-related acute toxicities and better tolerance. IC + RT displayed similar survival outcomes as IC + CCRT/AC for NPC patients with negative post-IC EBV DNA.


Introduction
Nasopharyngeal carcinoma (NPC) is an endemic tumor that is prevalent in southeast Asia, especially in southern China [1]. Induction chemotherapy (IC) combined with concurrent chemoradiotherapy (CCRT) is currently the standard treatment for locoregionally advanced NPC (LA-NPC) [2,3]. However, the evidence of concurrent chemotherapy (CCT) mainly comes from 2D radiotherapy (RT) [4]. In the era of intensity-modulated radiotherapy (IMRT), the value of CCT remains controversial. Moreover, CCT was associated with increased toxicities, which would compromise patients' compliance and life quality, and increase the risk of treatment-related death [4][5][6][7]. After RT, compliance with adjuvant chemotherapy (AC) was also poor [8,9]. Recently, several retrospective studies have explored the de-intensified treatment of RT alone versus CCRT after IC for LA-NPC, but the results were inconsistent [10][11][12][13]. Due to the heterogeneity of LA-NPC, de-intensified treatment for unselected patients may cause treatment failure.
EBV DNA is an important biomarker of NPC [14]. Many studies have shown that post-IC EBV DNA was an independent predictor of treatment outcome [15][16][17][18]. Undetectable post-IC EBV DNA was associated with superior survival rates. However, there is no report about de-intensified treatment based on post-IC EBV DNA status.
In this study, we aimed to compare the efficacy and toxicity of IC + RT and IC + CCRT/AC in patients with negative post-IC EBV DNA and evaluate the feasibility of omitting concurrent or adjuvant chemotherapy for these low-risk patients.

Patients
NPC patients treated with IC at our Center were retrospectively reviewed between September 2017 and November 2020. The inclusion criteria were as follows: (1) treatmentnaive, pathologically confirmed NPC patients; (2) no evidence of distant metastasis; (3) treated with IC before radical IMRT; (4) negative EBV DNA level (<500 copies/mL) after IC; and (5) without previous or concomitant malignancies. A total of 547 patients who met the criteria were included in this study ( Figure 1).

Introduction
Nasopharyngeal carcinoma (NPC) is an endemic tumor that is prevalent in southeast Asia, especially in southern China [1]. Induction chemotherapy (IC) combined with concurrent chemoradiotherapy (CCRT) is currently the standard treatment for locoregionally advanced NPC (LA-NPC) [2,3]. However, the evidence of concurrent chemotherapy (CCT) mainly comes from 2D radiotherapy (RT) [4]. In the era of intensity-modulated radiotherapy (IMRT), the value of CCT remains controversial. Moreover, CCT was associated with increased toxicities, which would compromise patients' compliance and life quality, and increase the risk of treatment-related death [4][5][6][7]. After RT, compliance with adjuvant chemotherapy (AC) was also poor [8,9]. Recently, several retrospective studies have explored the de-intensified treatment of RT alone versus CCRT after IC for LA-NPC, but the results were inconsistent [10][11][12][13]. Due to the heterogeneity of LA-NPC, de-intensified treatment for unselected patients may cause treatment failure.
EBV DNA is an important biomarker of NPC [14]. Many studies have shown that post-IC EBV DNA was an independent predictor of treatment outcome [15][16][17][18]. Undetectable post-IC EBV DNA was associated with superior survival rates. However, there is no report about de-intensified treatment based on post-IC EBV DNA status.
In this study, we aimed to compare the efficacy and toxicity of IC + RT and IC + CCRT/AC in patients with negative post-IC EBV DNA and evaluate the feasibility of omitting concurrent or adjuvant chemotherapy for these low-risk patients.

Patients
NPC patients treated with IC at our Center were retrospectively reviewed between September 2017 and November 2020. The inclusion criteria were as follows: (1) treatmentnaive, pathologically confirmed NPC patients; (2) no evidence of distant metastasis; (3) treated with IC before radical IMRT; (4) negative EBV DNA level (<500 copies/mL) after IC; and (5) without previous or concomitant malignancies. A total of 547 patients who met the criteria were included in this study ( Figure 1). All patients were evaluated by detailed medical history, physical examination, magnetic resonance imaging (MRI) (preferred) or computed tomography (CT) of the head and neck, chest CT, bone scan, abdominal ultrasonography or whole-body fluorodeoxyglucose positron emission tomography CT (PET/CT), fiberoptic nasopharyngoscopy or indirect nasopharyngoscopy, electrocardiogram, and complete blood sampling, including plasma EBV DNA level, before the start of treatment. All patients were staged according to the 8th edition of the International Union Against Cancer/American Joint Committee on Cancer (UICC/AJCC) staging system. This study was approved by the Institutional Review Board of our Cancer Center (No. 2009224-1).

Plasma EBV DNA Detection
Peripheral venous blood (5 mL) was obtained for plasma EBV DNA detection using real-time quantitative polymerase chain reaction (qPCR). Post-IC EBV DNA was detected within one week before the start of IMRT. The EBV DNA levels were defined as positive (≥500 copies/mL) or negative (<500 copies/mL), according to the standards of our center.
IMRT was delivered following our institutional treatment protocol. The target volume was defined according to the International Commission on Radiation Units (ICRU) and Measurements Reports 59 and 62 [19]. The gross tumor volume (GTV) was based on the post-IC images for intracavity tumors and lymph nodes. For infiltration tumors (bony structures of the skull base and cervical vertebra invasion) and extracellular invasion of lymph nodes, the GTV was based on the pre-IC images. The prescribed dose was 66-70.4 Gy to GTV (including the primary tumor and positive lymph nodes), 60 Gy to CTV1 (high-risk clinical target volume), and 54 Gy to CTV2 (low-risk clinical target volume) in 30-35 fractions. The simultaneous integrated boost (SIB) technique was used for all of the target volumes. More details have been described in a previous study [20].

Follow-Up and Assessment
Patients were followed up weekly during IMRT. After treatment, patients were assessed every three months in the first two years, every six months in the third to fifth years, and then annually thereafter.
Follow-up assessments included examination of the head and neck, EBV DNA levels, and thyroid and pituitary function tests. MRI of the nasopharynx, chest CT scan, and ultrasound or CT of the neck and abdomen were performed every 6-12 months. Additional tests were recommended when clinically indicated. Acute toxicities were graded according to the National Cancer Institute Common Toxicity Criteria for Adverse Events (version 4.02). The locoregional relapse-free survival (LRFS), distant metastasis-free survival (DMFS), overall survival (OS), and progression-free survival (PFS) rates were calculated from the day of the first treatment to the day of each event occurred or the last follow-up.

Statistical Analysis
The χ 2 test or Fisher's exact test was used to compare clinicopathologic and treatment characteristics between groups. The LRFS, DMFS, OS, and PFS rates were estimated and compared using the Kaplan-Meier method and log-rank tests. The Cox regression model was used for multivariate analyses. Patients were classified into the IC + RT group and the IC + CCRT/AC group. The IC + CCRT/AC group included patients who were treated with IC + CCRT and IC + RT + AC. One-to-one propensity score matching was performed using logistic regression, with a nearest-neighbor caliper width of 0.1 to optimize the comparability of different treatment groups. Covariates used for matching included gender, age, T category, N category, clinical stage, pre-IC EBV DNA levels, IC regimen, IC cycle, and targeted therapy. All the statistical analysis was performed using the Statistical Package for Social Sciences (SPSS version 20.0) and R (version 4.2.2). Two-sided p values < 0.05 were considered statistically significant.

Patient and Treatment Characteristics
The baseline characteristics of the 547 patients are summarized in Table 1. Among them, 115 patients received IC + RT, and 432 received IC + CCRT/AC. Before PSM, patients treated with IC + CCRT/AC tended to be younger (<45 years, p = 0.002) and have more advanced diseases (N2-3, p = 0.023; stage IVa, p = 0.007) than those with IC + RT. IC regimen and cycle were also unbalanced, with more patients receiving a GP regimen in the IC + CCRT/AC group (p < 0.001), and most of them received 2 cycles (p < 0.001). There was no significant difference in gender, T category, pre-IC EBV DNA levels, and targeted therapy. After one-to-one propensity score matching, 99 patients from each group were enrolled in the study. All variables were well-balanced between the two groups ( Table 1).   In the IC + CCRT/AC group, a total of 205 patients received AC, of which 147 patients (71.7%) completed 2 cycles, 6 patients (2.9%) completed 3 cycles, and 52 patients (25.4%) completed only 1 cycle due to intolerable toxicities or refusal by patients. A total

Survival Outcomes
With

Treatment Failures
Patterns of treatment failure were analyzed in the original cohort. At the time of the last follow-up, a total of 34 (7.9%) patients in the IC + CCRT/AC and 9 (7.8%) in the IC + RT group had locoregional recurrences. Distant failure was found in 35 (8.1%) and 4 (3.5%) patients in the IC + CCRT/AC and IC + RT groups, respectively. Twenty-five patients died. Cancer-specific death accounted for 64% of all death. Details are shown in Table 2. There was no significant difference in the cumulative incidence of locoregional relapse (LRR) and distant metastasis (DM) between the two groups. In the original cohort, the 3year cumulative incidence of LRR and DM among patients who underwent IC + CCRT/AC and IC + RT was 7.9% vs. 7.8% (p = 0.977) and 8.1% vs. 3.5% (p = 0.085), respectively ( Figure 3A). In the PSM cohort, the 3-year cumulative incidence of LRR and DM for patients treated with IC + CCRT/AC and IC + RT was 7.1% vs. 8.1% (p = 0.804) and 6.1% vs. 3.0% (p = 0.485), respectively ( Figure 3B). respectively ( Figure 3A). In the PSM cohort, the 3-year cumulative incidence of LRR and DM for patients treated with IC + CCRT/AC and IC + RT was 7.1% vs. 8.1% (p = 0.804) and 6.1% vs. 3.0% (p = 0.485), respectively ( Figure 3B).

Toxicities
Treatment-related acute toxicities for different treatment groups in the PSM cohort were evaluated. Grade 3-4 hematological toxicities were more frequently observed in the IC + CCRT group than in the IC + RT group. Patients in the IC + CCRT group were more likely to suffer from grade 3-4 leukocytopenia (13.6 vs. 2%, p = 0.017) and anemia (6.8 vs. 0%, p = 0.046), as compared with those in the IC + RT alone group. There was no significant difference in grade 3-4 dermatitis and mucositis between the two groups. However, grade 2 mucositis was more frequently observed in the IC + CCRT group than in the IC + RT group (43.2% vs. 24.2%, p = 0.023). Treatment-related toxicities are shown in Table 3.

Toxicities
Treatment-related acute toxicities for different treatment groups in the PSM cohort were evaluated. Grade 3-4 hematological toxicities were more frequently observed in the IC + CCRT group than in the IC + RT group. Patients in the IC + CCRT group were more likely to suffer from grade 3-4 leukocytopenia (13.6 vs. 2%, p = 0.017) and anemia (6.8 vs. 0%, p = 0.046), as compared with those in the IC + RT alone group. There was no significant difference in grade 3-4 dermatitis and mucositis between the two groups. However, grade 2 mucositis was more frequently observed in the IC + CCRT group than in the IC + RT group (43.2% vs. 24.2%, p = 0.023). Treatment-related toxicities are shown in Table 3.