
==== Front
Ann Med
Ann Med
Annals of Medicine
0785-3890
1365-2060
Taylor & Francis

39310946
10.1080/07853890.2024.2407061
2407061
Version of Record
Research Article
Oncology
Impact of the radiotherapy rhythm on prognosis in nasopharyngeal carcinoma
Y. LI et al.
Li Ying a*
Chen Xiaochuan a*
Wu Lishui a*
Huang Zongwei a
Xu Siqi a
Hong Xinyi a
Lai Jinghua a
https://orcid.org/0000-0001-7660-6955
Qiu Sufang abc
Zheng Xiong a
a Radiation Oncology Department, Clinical Oncology School of Fujian Medical University, Fujian Cancer Hospital (Fujian Branch of Fudan University Shanghai Cancer Center), Fuzhou, Fujian, China
b Fujian Key Laboratory of Translational Cancer Medicine, Fuzhou, Fujian, China
c Fujian Provincial Key Laboratory of Tumor Biotherapy, Fuzhou, Fujian, China
* These authors have contributed equally to this work.

Supplemental data for this article can be accessed online at https://doi.org/10.1080/07853890.2024.2407061.

CONTACT Xiong ZhengZhengxiong2024@126.com
Sufang Qiu sufangqiu@fjmu.edu.cn Radiation Oncology Department, Clinical Oncology School of Fujian Medical University, Fujian Cancer Hospital (Fujian Branch of Fudan University Shanghai Cancer Center), Fuzhou, Fujian 350014, China
23 9 2024
2024
23 9 2024
56 1 240706130 5 2024
1 8 2024
6 8 2024
KnowledgeWorks Global Ltd.21 9 2024
published online in a building issue21 9 2024
© 2024 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group
2024
The Author(s)
https://creativecommons.org/licenses/by-nc/4.0/ This is an Open Access article distributed under the terms of the Creative Commons Attribution-NonCommercial License (http://creativecommons.org/licenses/by-nc/4.0/), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allow the posting of the Accepted Manuscript in a repository by the author(s) or with their consent.

Abstract

Objective

The role of chronoradiobiology in nasopharyngeal carcinoma (NPC) has not been fully elucidated. We sought to investigate the impact of radiotherapy rhythm on the survival outcomes of individuals to explore a chronomodulated radiation strategy to improve prognosis of NPC.

Methods

A cohort comprising non-metastatic NPC patients subjected to intensity-modulated radiotherapy at Fujian Cancer Hospital between Jan. 2016 and Dec. 2019 was assembled. Rhythmic fluctuation of radiotherapy (RFRT) was quantified based on the temporal distribution of radiation delivery. Cox proportional hazard model was performed to explore the impact of radiotherapy rhythm on all-cause mortality. The maximally selected rank statistics method was employed to discern an optimal cutoff. Sensitivity analyses were conducted to ensure the robustness of observed associations.

Results

Our analysis encompassed 2245 patients, with a median follow-up duration of 55 months, during which 315 individuals succumbed. Multivariate Cox regression analysis unveiled a significant correlation between prolonged RFRT and heightened mortality risk in NPC patients (HR, 1.17, 95% CI, 1.07-1.27, p < .001), a relationship robust to comprehensive adjustment for confounding variables. A cutoff value of 3 h was selected for potential clinical application, beyond which patients exhibited markedly poorer survival outcomes. Subgroup analyses consistently underscored the directional consistency of observed effects.

Conclusion

Our study sheds light on the potential advantages of scheduling radiotherapy sessions at consistent times. These findings have implications for optimizing radiotherapy schedules and warrant further investigation into personalized chronotherapy approaches in NPC management.

Keywords

Nasopharyngeal carcinoma
intensity-modulated radiotherapy
chronoradiotherapy
circadian rhythm
Science and Technology Program of Fujian Province, China 2018Y2003 Fujian Provincial Clinical Research Center for Cancer Radiotherapy and Immunotherapy 2020Y2012 National Clinical Key Specialty Construction Program 10.13039/501100012232 2021 Fujian Clinical Research Center for Radiation and Therapy of Digestive, Respiratory and Genitourinary Malignancies 2021Y2014 National Natural Science Foundation of China 10.13039/501100001809 82072986 Major Scientific Research Program for Young and Middle-aged Health Professionals of Fujian Province,China 2021ZQNZD010 Science and Technology Pilot Program of Fujian Province, China 2021Y0053 Qihang Funds of Fujian Medical University 2023QH2050 2022QH2048 High-level Talent Training Program of Fujian Cancer Hospital 2022YNG07 Major Scientific Research Program for Young and Middle-aged Health Professionals of Fujian Province, China This work was supported by the grants of Science and Technology Program of Fujian Province, China (2018Y2003); Fujian Provincial Clinical Research Center for Cancer Radiotherapy and Immunotherapy (2020Y2012); Supported by the National Clinical Key Specialty Construction Program (2021); Fujian Clinical Research Center for Radiation and Therapy of Digestive, Respiratory and Genitourinary Malignancies (2021Y2014). National Natural Science Foundation of China (82072986); Major Scientific Research Program for Young and Middle-aged Health Professionals of Fujian Province,China (Grant No.2021ZQNZD010); Science and Technology Pilot Program of Fujian Province, China (2021Y0053); the Qihang Funds of Fujian Medical University (2023QH2050, 2022QH2048); and High-level Talent Training Program of Fujian Cancer Hospital (2022YNG07). Major Scientific Research Program for Young and Middle-aged Health Professionals of Fujian Province, China.
==== Body
pmcIntroduction

Nasopharyngeal carcinoma (NPC) represents a distinctive entity within head and neck malignancies, that differs in histological attributes, epidemiological trends, molecular underpinnings, and therapeutic responses, predominantly affects populations from east and southeast Asia [1]. Incremental enhancements in patient survival rates, particularly among those afflicted with locoregionally advanced disease, have been attributed to the successful implementation of chemoradiation modalities over successive decades [2]. Nevertheless, despite these strides, the heterogeneity of NPC necessitates to optimize treatment strategies by a multidisciplinary approach [3]. The imperative to optimize tumor reactivity and therapeutic efficacy remains paramount in the pursuit of favorable clinical outcomes for affected individuals.

In recent years, a circadian-based treatment, known as chronotherapy, leveraging inherent biological rhythms to augment treatment efficacy, has garnered wide attention within clinical settings and has demonstrated the effect on tumors [4]. Circadian rhythm, also known as the circadian clock, is an endogenous, continuous physiological phenomenon that shows changes in a cycle of about 24 h [5]. Perturbations in circadian homeostasis have been implicated in the etiology and progression of diverse maladies, including malignancies [6]. Previous investigations have elucidated how chemotherapy treatments could be improved by giving drugs at specific times of the day (chronomodulation) based on the circadian rhythm [7–9]. Radiotherapy also holds substantial utility in the management of cancers compared to chemotherapy, especially in NPC [10]. However, the impact of temporal considerations on radiotherapeutic interventions, and subsequent prognostic implications of NPC, remain relatively underexplored within clinical paradigms, notwithstanding the potential variability introduced by scheduling constraints during conventional working hours.

In light of the achievements in chronomodulated chemotherapy and immunotherapy, research has expanded its focus towards chronomodulated radiotherapy or ‘chronoradiotherapy’ [11]. Several investigations have surfaced elucidating the potential benefits of temporal optimization in radiotherapeutic regimens, particularly concerning the mitigation of adverse effects in patients afflicted with head and neck cancer, including NPC [12–14]. Notably, Elicin et al. and their research cohort have reported heightened acute toxicity associated with radiotherapy administration during darker periods in patients with head and neck cancer, delineated by the temporal proximity to the equinoxes in the northern hemisphere [15]. Furthermore, their findings underscored enhanced 5-year loco-regional control and progression-free survival (PFS) rates when radiotherapy was administered during these darker intervals [16]. Nonetheless, extant studies have focused on the impact of radiotherapy on treatment outcomes during the season or at daily specific times [13, 15]. The prognostic implications of temporal variations in radiotherapy in NPC remain unclear.

In this study, we gathered data pertaining to the daily timing of radiotherapy and subsequently computed radiotherapy fluctuation time for each NPC patient. We aimed to evaluate the prognostic significance of radiotherapy rhythm in NPC, thereby delving into the potential utility of a chronomodulated radiation approach aimed at refining radiotherapeutic protocols and ultimately ameliorating the prognostic outlook for individuals afflicted with NPC.

Methods

Study population

We reviewed the medical records of all NPC patients newly diagnosed at Fujian Cancer Hospital within the timeframe spanning Jan. 2016 to Dec. 2019. Eligibility criteria for inclusion comprised: (1) histologically-proven primary NPC; (2) absence of distant metastasis (M0); (3) completion of radical intensity-modulated radiotherapy (IMRT); and (4) availability of detailed data concerning the daily timing of radiation treatment. Patients with unknown disease stage; treatment interruption or abandonment; history of prior anti-neoplastic therapy; disease progression during the treatment period; and lost to follow-up were not eligible for this study. The study was conducted in accordance with the Declaration of Helsinki and ethical clearance for this investigation was obtained from the institutional review board of Fujian Cancer Hospital (K2024-286-01), with retrospective nature of the study rendering patient informed consent unnecessary.

Radiotherapy rhythm analysis

We extracted the exact daily time of radiation treatment delivery (hour, minute) for each patient from the MOSAIQ integration platform [17]. The present study defined the relative amplitude between the latest and earliest treatment times over the entire radiotherapy course as rhythmic fluctuation of radiotherapy (RFRT). For instance, a patient undergoing radiotherapy exclusively between 6:00 am and 12:00 am throughout the treatment course, with a RFRT of 6 h. Given the routine practice of conducting radiotherapy sessions across three distinct shifts-morning, midday, and evening-at our institution, individuals characterized by an RFRT exceeding 6 h were excluded from the analysis to explore the potential impact of radiotherapy rhythm on patients with NPC.

Treatment and assessments

All patients received IMRT. Patients diagnosed with stage I disease underwent IMRT as a standalone intervention, while those presenting with stage II to IVa disease were subjected to concurrent chemoradiotherapy (CCRT), induction chemotherapy (IC) followed by radiotherapy, or a combination of IC with CCRT, with or without adjunctive chemotherapy (AC). The delineation of planning target volumes (PTVs) encompassing the gross tumor volume of primary nasopharyngeal tumor (GTV-T) and involved lymph nodes (GTV-N) involved the administration of radiation doses ranging from 180 to 225 cGy per fraction for GTV-T and 160 to 230 cGy per fraction for GTV-N, delivered over 28 to 37 fractions, once daily, five times per week. In instances where residual lesions were identified via contrast-enhanced magnetic resonance imaging (MRI) conducted towards the conclusion of radiotherapy, supplementary irradiation ranging from 200 to 1000 cGy over 1 to 6 fractions was administered to target residual lesions within the primary nasopharyngeal or metastatic lymph nodes, as deemed appropriate. Commonly employed agents for induction and adjuvant chemotherapy included paclitaxel, 5-fluorouracil, or gemcitabine, in combination with platinum-based compounds. Concurrent chemotherapy during radiotherapy typically entailed the administration of cisplatin (at doses ranging from 80 to 100 mg/m2) or alternative platinum agents administered at three-week intervals. Patients identified as having a high risk of disease progression, such as those presenting with T4N2 or N3 disease, received oral maintenance chemotherapy post-treatment, with the utilization of S1 or capecitabine.

Follow-up and outcomes

All patients underwent systematic follow-up, consisting of regular physical examinations nasopharyngoscopy, and plasma Epstein-Barr virus (EBV) DNA testing conducted at three-month intervals during the initial 2-year post-radiotherapy period, transitioning to biannual assessments for years 3 to 5, and subsequently, annual evaluations thereafter. Biannual head and neck MRI scans were performed, complemented by chest computed tomography (CT) scans administered every 3 to 6 months, and whole-body bone emission computed tomography (ECT) conducted on an annual basis.18F-fluorodeoxyglucose positron emission tomography and computed tomography (PET-CT) scans were not implemented unless necessitated by inconclusive findings from the aforementioned assessments or clinical suspicion of recurrence or metastasis during follow-up. Further investigations were initiated based on pertinent clinical indicators. The confirmation of tumor recurrence or metastasis relied upon biopsy (primary lesion) or fine-needle aspiration results (metastatic lymph node). In cases where lesions were inaccessible for direct sampling, clinical diagnosis necessitated the presence of at least two radiological features evident on CT, MRI, ECT, or PET/CT scans. Management of residual disease and tumor relapse, if detected, was determined on a case-by-case basis.

The final assessment point occurred on Feb. 20, 2024. Loss to follow-up was defined as a survival duration of less than 3 months. The primary endpoint of this investigation was overall survival (OS), delineated as the duration from the completion date of radiotherapy to the occurrence of all-cause mortality. Secondary endpoints encompassed locoregional failure-free survival (LRFS) and distant metastasis failure-free survival (DMFS), denoting the duration from the conclusion of radiotherapy to the onset of relapse or distant metastasis and/or demise from any cause, as well as PFS, characterized by the interval until the manifestation of locoregional failure, distant metastasis, or all-cause mortality.

Covariates

Data pertaining to demographic profiles, tumor attributes, and therapeutic regimens of study participants were systematically compiled. Demographic variables encompassed age and gender. Tumor characteristics encompassed the TNM stage, which underwent reassessment utilizing the 8th edition American Joint Committee on Cancer (AJCC) stage classification by two radiologists, and the plasma EBV DNA levels upon admission and post-radiotherapy. Pre-treatment EBV DNA levels were stratified dichotomously using a threshold of 4,000 copies/mL, previously identified as indicative of a ‘high-risk’ profile for disease progression [18]. Post-treatment EBV DNA levels (detectable vs. undetectable) were analyzed as categorical variables.

Statistical analysis

Continuous variables underwent distribution assessment using Kolmogorov-Smirnov tests, with median and interquartile range (IQR) reported as descriptive statistics. Absolute values and proportions were provided for categorical variables. Statistical comparisons between patient cohorts were conducted utilizing Kruskal-Wallis tests for continuous variables and Chi-squared tests for categorical variables.

Cox proportional hazard regression models were employed to estimate hazard ratios (HRs) and corresponding 95% confidence intervals (CIs) elucidating the associations between RFRT and other clinical parameters with outcomes. The assumption of proportional hazards was evaluated using the Schoenfeld residual method. Covariates demonstrating significance in univariate analysis (p < 0.05) were integrated into multivariate analysis. Restricted cubic splines (RCS) were employed to visualize the relationship between RFRT and OS in NPC patients. Three regression models were constructed, each adjusting for distinct covariates to mitigate potential confounding biases. Model 1 adjusted for demographic variables including age and gender; Model 2 additionally adjusted for tumor characteristics encompassing T stage, N stage, and plasma EBV DNA levels; and Model 3 incorporated variables from Model 2 along with antineoplastic data such as irradiation fractional dose and treatment modality. The optimal cutoff for RFRT was determined utilizing the maximally selected rank statistics method [19]. Kaplan-Meier analysis generated survival curves, with comparisons conducted via the Log-rank test. Subgroup analyses of clinical variables and interaction tests were performed to ascertain the stability of the effect size and direction of association between survival and RFRT across various subpopulations. Statistical analyses were executed using R software v4.3.0, supplemented by Zstats v0.90 (www.medsta.cn/software) and SPSS Statistics v25.0. A significance threshold of two-sided p < 0.05 was applied.

Results

Clinical characteristics

From the initial cohort of 3,469 NPC patients at study commencement, we systematically excluded 148 individuals with M1 stage, 40 with indeterminate stage, 108 with a history of prior anti-neoplastic therapy, 335 with treatment discontinuation or abandonment, 41 lost to follow-up, 6 undergoing accelerated hyperfractionation, and 4 experiencing disease progression during treatment. Moreover, 254 cases lacking complete daily radiotherapy timing records, and 288 patients exhibiting RFRT exceeding 6 h were excluded from the analysis. Ultimately, a total of 2,245 participants were retained for subsequent analyses. Figure 1 described the research profile of this study. At baseline, the median age was 49 years (IQR 42, 56), with 1,633 (72.74%) of the cohort comprising male individuals. The median RFRT was 3.33 h (IQR 2.30, 4.42). Treatment modalities encompassed 118 patients (5.26%) undergoing IMRT alone, 551 (24.54%) receiving IC in conjunction with IMRT, 204 (9.09%) subjected to CCRT, and 1,372 (61.11%) undergoing IC combined with CCRT. Over a median follow-up duration of 55 months (IQR 43, 66), 315 participants (14.03%) succumbed to mortality. This subset exhibited a significantly higher prevalence of advanced disease stages, characterized by T3-4, N2-3, and stage III-IVa classifications, elevated plasma EBV DNA levels (ps < .001), and longer RFRT (p = .003). Additionally, deceased individuals were older (p < .001) and exhibited a higher proportion of male gender representation (p = .021). Further baseline characteristics of the study population are summarized in Table 1.

Figure 1. Study profile for the rhythm of radiotherapy in nasopharyngeal carcinoma. RFRT: rhythmic fluctuation of radiotherapy.

Table 1. Demographic and medical history baseline characteristics of NPC patients.

Characteristic	Overall
N = 2245	Surviving
N = 1930	Dead
N = 315	p-value	
Age (IQR, years)	49 (42, 56)	48 (41, 55)	53 (46, 63)	<0.001	
Gender, n (%)	 	 	 	0.021	
Male	1633 (72.74)	1387 (71.87)	246 (78.10)	 	
Female	612 (27.26)	543 (28.13)	69 (21.90)	 	
T stage, n (%)	 	 	 	<0.001	
T0-2	779 (34.70)	701 (36.32)	78 (24.76)	 	
T3-4	1466 (65.30)	1229 (63.68)	237 (75.24)	 	
N stage, n (%)	 	 	 	<0.001	
N0-1	1063 (47.35)	950 (49.22)	113 (35.87)	 	
N2-3	1182 (52.65)	980 (50.78)	202 (64.13)	 	
Clinical stage, n (%)	 	 	 	<0.001	
Stage I-II	326 (14.52)	307 (15.91)	19 (6.03)	 	
Stage III-IVa	1919 (85.48)	1623 (84.09)	296 (93.97)	 	
EBV DNApre (copies/mL)	 	 	 	<0.001	
≤ 4000	1289 (57.42)	1139 (59.02)	150 (47.62)	 	
> 4000	944 (42.05)	779 (40.36)	165 (52.38)	 	
missing	12 (0.53)	12 (0.62)	0 (0.00)	 	
EBV DNApost	 	 	 	<0.001	
Undetectable	1881 (83.79)	1663 (86.17)	218 (69.21)	 	
Detectable	333 (14.83)	238 (12.33)	95 (30.16)	 	
missing	31 (1.38)	29 (1.50)	2 (0.63)	 	
RFRT (IQR, h)	3.33 (2.30, 4.42)	3.28 (2.28, 4.40)	3.67 (2.52, 4.53)	0.003	
Irradiation fractional dose
(IQR, cGy/fraction)	 	 	 	 	
GTV-T-f	212 (205, 215)	212 (205, 215)	212 (200, 212)	<0.001	
GTV-N-f	210 (200, 215)	210 (200, 215)	205 (200, 212)	<0.001	
IC cycle	 	 	 	0.339	
0	322 (14.34)	278 (14.40)	44 (13.97)	 	
≤ 3	1303 (58.04)	1109 (57.46)	194 (61.59)	 	
> 3	620 (27.62)	543 (28.13)	77 (24.44)	 	
Received CC, n (%)	 	 	 	0.107	
No	669 (29.80)	563 (29.17)	106 (33.65)	 	
Yes	1576 (70.20)	1367 (70.83)	209 (66.35)	 	
CC cycle (IQR)	2 (0, 2)	2 (0, 2)	2 (0, 2)	0.012	
Received AC, n (%)	 	 	 	0.042	
No	2030 (90.42)	1755 (90.93)	275 (87.30)	 	
Yes	215 (9.58)	175 (9.07)	40 (12.70)	 	
Treatment modality, n (%)	 	 	 	0.029	
RT alone	118 (5.26)	93 (4.82)	25 (7.94)	 	
IC + RT	551 (24.54)	470 (24.35)	81 (25.71)	 	
CCRT	204 (9.09)	185 (9.59)	19 (6.03)	 	
IC + CCRT	1372 (61.11)	1182 (61.24)	190 (60.32)	 	
Continuous variables were represented as median with IQR, with Kruskal-waills test for comparing; while categorical data were represented as absolute values and proportions with Chi-squared test. AC: adjuvant chemotherapy; CC: concurrent chemotherapy; CCRT: concurrent chemoradiotherapy; EBV DNA: Epstein-Barr virus DNA; EBV DNApre: plasma EBV DNA level at admission; EBV DNApost: post-radiotherapy plasma EBV DNA level; GTV-N-f: irradiation fractional dose of gross tumor volume of lymph nodes; GTV-T-f: irradiation fractional dose of gross tumor volume of primary nasopharyngeal tumor; IC: induction chemotherapy; IQR: interquartile range; NPC: nasopharyngeal carcinoma; RFRT: rhythmic fluctuation of radiotherapy; RT: radiotherapy.

Association between RFRT and OS in NPC patients

We used multivariate Cox regression to adjust for potential confounders of OS, including age, gender, T stage, N stage, pre- and post-treatment plasma EBV DNA level, RFRT, irradiation fractional dose of GTV-T and GTV-N and chemotherapy timing. The results were presented in Figure 2, showing age (HR, 1.04, 95% CI, 1.03-1.05, p < .001), T stage (HR, 1.65, 95% CI, 1.26-2.16, p < .001), N stage (HR, 1.68, 95% CI, 1.31-2.14, p < .001), post-treatment plasma EBV DNA level (HR, 2.61, 95% CI, 2.04-3.33, p < .001), and RFRT (HR, 1.17, 95% CI, 1.07-1.27, p < .001) were significantly correlated with all-cause mortality. RCS suggested the relationship between RFRT and OS was linear (p for nonlinearity = .576, Figure S1A), indicating the risk of mortality increased with the increase of RFRT. To further elucidate the relationship between RFRT and OS in NPC, we divided the RFRT into 6 equal parts by time. Compared with participants with RFRT within 1h (R1), segmented Cox regression analysis showed those within 2 to 6 had an increased risk of mortality (HRs > 1, unadjusted model, Table 2). Upon adjusting for demographic and tumor characteristics, participants with RFRT of 3-4h (R4), 4-5h (R5) and 5-6h (R6) were significantly associated with increasing mortality in Model 1 and Model 2 (HR, 2.12, 95% CI, 1.02-4.40, p = .044 for R4; HR, 2.18, 95% CI, 1.05-4.55, p = .037 for R5; HR, 2.69, 95% CI, 1.27-5.69, p = .010 for R6). When adjusted for other confounders of treatment, participants with R6 were still associated with a higher risk of mortality among NPC patients (Model 3, HR, 2.65, 95% CI, 1.25-5.62, p = .011). These findings underscore patients with shorter RFRT predicting better survival outcome in NPC.

Figure 2. Univariable and multivariable Cox proportional hazards regression of overall survival in patients with nasopharyngeal carcinoma. P values in bold are statistically significant. AC: adjuvant chemotherapy; CC: concurrent chemotherapy; CI: confidence interval; EBV DNA: Epstein-Barr virus DNA; EBV DNApre: plasma EBV DNA level at admission; EBV DNApost: post-radiotherapy plasma EBV DNA level; GTV-N-f: irradiation fractional dose of gross tumor volume of lymph nodes; GTV-T-f: irradiation fractional dose of gross tumor volume of primary nasopharyngeal tumor; HR: hazard ratio; IC: induction chemotherapy; RFRT: rhythmic fluctuation of radiotherapy.

Table 2. Association of rhythmic fluctuation of radiotherapy and all-cause mortality in patients with nasopharyngeal carcinoma.

Characteristic	Non-adjusted	Model 1	Model 2	Model 3	
HR (95% CI)	p-value	HR (95% CI)	p-value	HR (95% CI)	p-value	HR (95% CI)	p-value	
RFRT	1.13 (1.04-1.23)	0.003	1.17 (1.08-1.27)	<0.001	1.17 (1.08-1.27)	<0.001	1.17 (1.07-1.27)	<0.001	
R1	1.00 (Reference)	 	1.00 (Reference)	 	1.00 (Reference)	 	1.00 (Reference)	 	
R2	1.37 (0.65-2.92)	0.409	1.55 (0.72-3.32)	0.262	1.55 (0.72-3.32)	0.262	1.52 (0.71-3.26)	0.283	
R3	1.07 (0.51-2.25)	0.852	1.48 (0.70-3.13)	0.302	1.48 (0.70-3.13)	0.302	1.45 (0.68-3.06)	0.333	
R4	1.62 (0.78-3.35)	0.192	2.12 (1.02-4.40)	0.044	2.12 (1.02-4.40)	0.044	2.03 (0.98-4.23)	0.057	
R5	1.73 (0.83-3.58)	0.142	2.18 (1.05-4.55)	0.037	2.18 (1.05-4.55)	0.037	2.07 (0.99-4.33)	0.052	
R6	1.90 (0.90-4.02)	0.091	2.69 (1.27-5.69)	0.010	2.69 (1.27-5.69)	0.010	2.65 (1.25-5.62)	0.011	
EBV DNA: Epstein-Barr virus DNA; EBV DNApre: plasma EBV DNA level at admission; EBV DNApost: post-radiotherapy plasma EBV DNA level; GTV-N-f: irradiation fractional dose of gross tumor volume of lymph nodes; GTV-T-f: irradiation fractional dose of gross tumor volume of primary nasopharyngeal tumor; RFRT: rhythmic fluctuation of radiotherapy.

Model 1 was adjusted for age and gender.

Model 2 was adjusted for age, gender, T stage, N stage, EBV DNApre, and EBV DNApost.

Model 3 was adjusted for age, gender, T stage, N stage, EBV DNApre, and EBV DNApost, GTV-T-f, GTV-N-f, and Treatment modality.

R1 ∼ 6, relative amplitude of RFRT between 0 ∼ 1h, 1 ∼ 2h, 2 ∼ 3h, 3 ∼ 4h, 4 ∼ 5h, and 5 ∼ 6h.

Subgroup analysis according to RFRT

Using the maximally selected rank statistics, we identified the optimal cut-off value for high RFRT to predict OS as 3.42 h (Figure S1B). Furthermore, the RCS curve showed that the RFRT may have a piecewise effect on survival status with a distinct inflection point (inflection point: 3.33 h, Figure S1A). With clinical applicability in mind, we established the RFRT cut-off value at 3 h, leading to the categorization of patients into two groups: 945 cases (42.09%) in the RFRT ≤ 3-hour group and 1300 cases (57.91%) in the RFRT > 3-hour group. Analysis of clinical-pathological parameters, encompassing age, gender, TNM stage, plasma EBV DNA level, and treatment details, did not reveal statistically significant differences between the two groups (ps > .05, Table S1). Kaplan-Meier survival curves illustrated that NPC patients with RFRT ≤ 3 h exhibited more favorable outcomes in terms of OS (Log-rank p = .001), PFS (Log-rank p = .048), LRFS (Log-rank p = .005), and DMFS (Log-rank p = .022) compared to those with RFRT > 3 h (Figure 3).

Figure 3. Survival curves of (A) overall survival, (B) progression-free survival, (C) locoregional recurrence-free survival, and (D) distant metastasis-free survival in patients with rhythmic fluctuation ≤ 3h and > 3h. DMFS, distant metastasis-free survival; LRFS: locoregional recurrence-free survival; OS: overall survival; PFS: progression-free survival; RFRT: rhythmic fluctuation of radiotherapy.

Sensitivity analyses

The findings depicted in Figure 4 outline the outcomes of the segmented subgroup analysis. No significant interactions between clinical variables and RFRT were observed. The associations of RFRT with OS remained consistent when considering additional subgroup participants, and were stable among male, and patients with T3-4, N2-3, stage III-IVa, pre-treatment EBV DNA > 4000 copies/mL and undetectable post-treatment EBV DNA level and receiving IC (ps < .05). Moreover, associations between RFRT and OS were evident among patients both with and without CC or AC. Given the independent prognostic significance of age in NPC (HR, 1.05, 95% CI, 1.04-1.06, p < .001 in univariable Cox regression; HR, 1.04, 95% CI, 1.03-1.05, p < .001 in multivariable Cox regression, Figure 2), a sensitivity analysis was conducted based on the median age. As depicted in Table S2, the association between RFRT and OS remained consistent across age strata (HR, 1.58, 95% CI, 1.06-2.35, p = .025 in patients aged < 50; HR, 1.45, 95% CI, 1.08-1.93, p = .012 in patients aged ≥ 50).

Figure 4. Forest Plot of overall survival HRs by patient subgroup. HRs and 95% CIs of rhythmic fluctuation > 3h versus rhythmic fluctuation ≤ 3h (univariable Cox models). P values in bold are statistically significant. AC: adjuvant chemotherapy; CC: concurrent chemotherapy; CI: confidence interval; EBV DNA: Epstein-Barr virus DNA; EBV DNApre: plasma EBV DNA level at admission; EBV DNApost: post-radiotherapy plasma EBV DNA level; HR: hazard ratio; IC: induction chemotherapy; RFRT: rhythmic fluctuation of radiotherapy.

Discussion

While previous studies have explored chronomodulation to mitigate the adverse effects of radiation therapy, there remains a paucity of research investigating the impact of temporal variations in radiotherapy on the prognosis of NPC [13]. Our study assessed the broad range of treatment time intervals, enabling the identification of critical time frames that delineate significant differences in survival outcomes based on the daily timing of treatment administration. We observed that prolonged RFRT correlated significantly with diminished OS. Through adjustment for confounding factors, multivariate Cox proportional hazard regression analysis reaffirmed the robustness of the association between RFRT and OS, highlighting a trend toward increased mortality risk with each additional hour of RFRT. A cutoff value of 3 h was selected for potential clinical application. Patients undergoing RFRT for longer than 3 h exhibited significantly poorer OS, PFS, LRFS, and DMFS. Furthermore, the significant associations observed between radiotherapy rhythm and patient outcomes were further supported by sensitivity and subgroup analyses, demonstrating the consistency and stability of our findings across various patient populations.

Research in cancer biology has found the correlations between circadian rhythm and cancer, including cancer incidence, chemotherapeutic treatment and now radiotherapy [11]. The strategic timing of anti-cancer treatments has garnered increasing attention as investigations using both animal and human models unravel the molecular intricacies governing circadian regulation of biological pathways [20]. Given the specific modality of administration (daily dosages within a concise timeframe and targeted tissue specificity), radiotherapy presents an opportune avenue for chronotherapeutic interventions [12]. Although chronoradiotherapy has been explored by numerous research groups as a potential strategy to mitigate radiotherapy-associated toxicities and enhance tumor control, only limited studies have probed whether timing-specific radiotherapy could ameliorate treatment toxicity in NPC patients [13, 14]. Research suggests that patients with head and neck cancer receiving radiotherapy during fall/winter season may have better outcomes than those treated during spring/summer season, possibly secondary to seasonal variations in cell cycle progression [15]. Although season of radiotherapy had no impact on prognosis in two large, prospective datasets of head and neck cancer [21]. In addition, the timing of radiotherapy (morning radiotherapy) could enhance survival in non-metastatic NPC in our cohort that will be expounded upon in a separate article. Our observations hint at the potential advantages of scheduling radiotherapy sessions at consistent times. This possibly attributable to the influence of radiotherapy on cell cycle dynamics [22, 23]. Indeed, circadian rhythm-dependent cell cycle progression can yield variations in the response to radiotherapeutic interventions, potentially mediated by factors such as the circadian rhythm of tumor hypoxia, cell cycle progression affected by seasonal changes, and DNA damage [12, 15, 21, 23–26]. Studies by Xian et al. have demonstrated significant variations in the distribution of cells across G1, S, and G2-M phases relative to tumor sampling time in NPC, underscoring the importance of administering radiotherapy during the appropriate phases of the cell cycle [14, 27]. Furthermore, maintaining a regular radiotherapy schedule may mitigate the likelihood of surviving tumor cells undergoing redistribution during the cell cycle intervals between successive radiation deliveries, thereby optimizing treatment efficacy [28].

The timing of radiotherapy administration is often overlooked in clinical settings as it varies along the daytime working hours. Current investigations in chronotherapy have predominantly concentrated on delineating specific time points to demarcate radiotherapy periods, such as morning and afternoon groups [29, 30]. Given the aim of translating the cellular-level biological effects of the circadian cycle into therapeutic responses, Bermúdez-Guzmán et al. advocated for delineating specific time intervals for distinct patient cohorts as a superior approach [26]. In our investigation, we discerned that NPC patients undergoing radiotherapy, wherein the daily variation in treatment delivery was constrained to less than 3 h, exhibited reduced risks of mortality and adverse events. The value was identified by the maximally selected rank statistics and RCS, enhances the precision of our analyses [19, 31]. Notably, the intricate interplay between the circadian cycle, host physiology, cancer biology, and treatment outcomes are further modulated by inter-individual disparities and variations in clock gene polymorphisms [32–35]. It is possible that the sensitivity to radiotherapy across a 24-hour period may conform to a cyclical pattern akin to a cosine curve, mirroring the typical rhythmic expression of circadian clock genes [36]. Controlling the radiotherapy rhythm within 3 h may cause part of the time curve to fall within a periodic pattern of increased radiosensitivity to allow patient’s benefit from chrono-tailored radiotherapy, although further studies are warranted to verify such a pattern.

Furthermore, it has been suggested in previous studies to strive for a cohort of patients with maximal homogeneity when investigating the impact of chronoradiotherapy on patient outcomes [26]. Indeed, concerning treatment failure in NPC, it has been established that treatment response and clinical outcomes are influenced by various prognostic factors, including the number of tumor lesions, functional status, age, and comorbidities, among others [37–40]. In the present study, age, T stage, N stage, and post-radiotherapy plasma EBV DNA level emerged as independent prognostic factors for OS in NPC patients, consistent with findings from many prognostic investigations in NPC [41, 42]. Notably, even after adjusting for these confounding factors, a positive association between RFRT and heightened mortality risk in the NPC population persisted. This observed correlation remained robust across diverse NPC patient subgroups, including age stratifications. Interestingly, this trend was not evident in female patients, potentially attributable to the predominance of male patients in our cohort. A more representative cohort may be imperative to delineate potential sex- and age-based disparities in radiotherapy chronomodulation [43, 44].

Our study benefits from a large and well-characterized patient cohort, allowing for robust multivariate analyses that control for potential confounders. Despite our endeavors, it is imperative to acknowledge certain limitations in our study. Firstly, the inherent bias characteristic of cross-sectional studies is unavoidable; each daily timing of radiotherapy was documented via the MOSAIQ integration platform, introducing the potential for reporting and information biases. Secondly, while we conducted comprehensive adjustments for covariates to mitigate confounding factors, the presence of residual confounders cannot be entirely ruled out in the associations between radiotherapy rhythm and survival outcomes. Thirdly, the absence of data concerning radiotherapy-related adverse reactions precluded an exploration of the relationship between adverse reactions and radiotherapy rhythm in our analysis. Consequently, the findings of this retrospective study warrant cautious interpretation, particularly given the unclear elucidation of potential biological mechanisms of action. Subsequent studies are needed to better characterize and further optimize radiation therapeutic strategy, necessitating appropriately designed clinical trials to corroborate our observations.

Conclusion

To the best of our knowledge, this is the first study reporting the prognostic impact of radiotherapy rhythm in patients with NPC, revealing enhanced survival outcomes associated with radiotherapy administered within a regulated 3-hour window compared to prolonged intervals. This would allow an almost cost-free modification to treatment that would improve curative effect.

Supplementary Material

Supplemental Material

Acknowledgments

The flowchart presented in this article was created using Biorender, for which we extend our sincere gratitude.

Authors contributions

Y. Li: Writing-original draft, Formal analysis, Visualization; X. Chen: Formal analysis, Software, Visualization; L. Wu: Data curation, Investigation; Z. Huang: Methodology, Validation; S.Xu: Validation; X. Hong: Validation; J. Lai: Resources; S. Qiu: Project administration, Supervision; X. Zheng: Funding acquisition, Writing-review & editing. All authors contributed to the article and approved the final manuscript.

Ethics statement

The study was conducted in accordance with the Declaration of Helsinki and ethical clearance for this investigation was obtained from the institutional review board of Fujian Cancer Hospital (K2024-286-01), with retrospective nature of the study rendering patient informed consent unnecessary.

Disclosure statement

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability statement

Data that supports the findings of this study are available from the corresponding author upon request.
==== Refs
References

1 Chen YP, Chan ATC, Le QT, et al. Nasopharyngeal carcinoma. Lancet. 2019;394 (10192 ):64–80. doi: 10.1016/S0140-6736(19)30956-0.31178151
2 Juarez-Vignon Whaley JJ, Afkhami M, Onyshchenko M, et al. Recurrent/metastatic nasopharyngeal carcinoma treatment from present to future: where are we and where are we heading? Curr Treat Options Oncol. 2023;24 (9 ):1138–1166. doi: 10.1007/s11864-023-01101-3.37318724
3 De Felice F, Tombolini V, de Vincentiis M, et al. Multidisciplinary team in head and neck cancer: a management model. Med Oncol. 2018;36 (1 ):2. doi: 10.1007/s12032-018-1227-z.30426243
4 Damato AR, Herzog ED. Circadian clock synchrony and chronotherapy opportunities in cancer treatment. Semin Cell Dev Biol. 2022;126 :27–36. doi: 10.1016/j.semcdb.2021.07.017.34362656
5 Patke A, Young MW, Axelrod S. Molecular mechanisms and physiological importance of circadian rhythms. Nat Rev Mol Cell Biol. 2020;21 (2 ):67–84. doi: 10.1038/s41580-019-0179-2.31768006
6 Nelson N, Lombardo J, Matlack L, et al. Chronoradiobiology of breast cancer: the time is now to link circadian rhythm and radiation biology. Int J Mol Sci. 2022;23 (3 ):1331. doi: 10.3390/ijms23031331.35163264
7 Printezi MI, Kilgallen AB, Bond MJG, et al. Toxicity and efficacy of chronomodulated chemotherapy: a systematic review. Lancet Oncol. 2022;23 (3 ):e129-43. doi: 10.1016/S1470-2045(21)00639-2.35240088
8 Liu L, Luo X, Wu W, et al. Long-term survival, toxicities, and the role of chrono-chemotherapy with different ­infusion rates in locally advanced nasopharyngeal carcinoma patients treated with intensity-modulated radiation therapy: a retrospective study with a 5-year follow-up. Front Oncol. 2024;14 :1371878. doi: 10.3389/fonc.2024.1371878.38585011
9 Sancar A, Van Gelder RN. Clocks, cancer, and chronochemotherapy. Science. 2021;371 (6524 ):eabb0738. doi: 10.1126/science.abb0738.33384351
10 Tang L-L, Guo R, Zhang N, et al. Effect of radiotherapy alone vs radiotherapy with concurrent chemoradiotherapy on survival without disease relapse in patients with low-risk nasopharyngeal carcinoma: a randomized clinical trial. JAMA. 2022;328 (8 ):728–736. doi: 10.1001/jama.2022.13997.35997729
11 Harper E, Talbot CJ. Is it time to change radiotherapy: the dawning of chronoradiotherapy? Clin Oncol. 2019;31 (5 ):326–335. doi: 10.1016/j.clon.2019.02.010.
12 Gu F, Farrugia MK, Duncan WD, et al. Daily time of radiation treatment is associated with subsequent oral mucositis severity during radiotherapy in head and neck cancer patients. Cancer Epidemiol Biomarkers Prev. 2020;29 (5 ):949–955. doi: 10.1158/1055-9965.EPI-19-0961.32098893
13 Lv J, Liao S, Li B, et al. Scheduling radiotherapy for ­patients with nasopharyngeal carcinoma in the corresponding time window can reduce radiation-induced oral mucositis: a randomized, prospective study. Cancer Med. 2023;12 (15 ):16032–16040. doi: 10.1002/cam4.6252.37537945
14 Zhang Y, Chen X, Ren P, et al. Synergistic effect of combination topotecan and chronomodulated radiation therapy on xenografted human nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys. 2013;87 (2 ):356–362. doi: 10.1016/j.ijrobp.2013.05.047.23906928
15 Brolese EK, Cihoric N, Bojaxhiu B, et al. The impact of delivery daytime and seasonality of radiotherapy for head and neck cancer on toxicity burden. Radiother Oncol. 2021;158 :162–166. doi: 10.1016/j.radonc.2021.02.039.33667582
16 Elicin O, Koller Brolese E, Bojaxhiu B, et al. The prognostic impact of daytime and seasonality of radiotherapy on head and neck cancer. Radiother Oncol. 2021;158 :293–299. doi: 10.1016/j.radonc.2021.04.004.33848563
17 Yang X, He Z, Jiang X, et al. Development and application of MOSAIQ integration platform based on radiotherapy workflow. Chinese Journal of Radiation Oncology. 2017;6 :918–923.
18 Tang SQ, Chen L, Li WF, et al. Identifying optimal clinical trial candidates for locoregionally advanced nasopharyngeal carcinoma: analysis of 9468 real-world cases and validation by two phase 3 multicentre, randomised controlled trial. Radiother Oncol. 2022;167 :179–186. doi: 10.1016/j.radonc.2021.12.029.34971660
19 Woo S, Kim S, Kim J. Determining cutoff values of prognostic factors in survival data with competing risks. Comput Stat. 2016;31 (1 ):369–386. doi: 10.1007/s00180-015-0582-x.
20 Dakup PP, Porter KI, Little AA, et al. The circadian clock regulates cisplatin-induced toxicity and tumor regression in melanoma mouse and human models. Oncotarget. 2018;9 (18 ):14524–14538. doi: 10.18632/oncotarget.24539.29581861
21 Limkin E, Blanchard P, Lacas B, et al. Season of radiotherapy and outcomes of head & neck cancer patients in the MACH-NC & MARCH meta-analyses. Radiother Oncol. 2024;190 :110011. doi: 10.1016/j.radonc.2023.110011.37956890
22 Ashok Kumar PV, Dakup PP, Sarkar S, et al. It’s about time: advances in understanding the circadian regulation of DNA damage and repair in carcinogenesis and cancer treatment outcomes. Yale J Biol Med. 2019;92 (2 ):305–316.31249491
23 Leung C, Gérard C, Gonze D. Modeling the circadian control of the cell cycle and its consequences for cancer chronotherapy. Biology. 2023;12 (4 ):612. doi: 10.3390/biology12040612.37106812
24 Pawlik TM, Keyomarsi K. Role of cell cycle in mediating sensitivity to radiotherapy. Int J Radiat Oncol Biol Phys. 2004;59 (4 ):928–942. doi: 10.1016/j.ijrobp.2004.03.005.15234026
25 Shostak A. Circadian clock, cell division, and cancer: from molecules to organism. Int J Mol Sci. 2017;18 (4 ):873. doi: 10.3390/ijms18040873.28425940
26 Bermúdez-Guzmán L, Blanco-Saborío A, Ramírez-Zamora J, et al. The time for chronotherapy in radiation oncology. Front Oncol. 2021;11 :687672. doi: 10.3389/fonc.2021.687672.34046365
27 Xian L-J, Jian S, Cao Q-Y, et al. Circadian rhythms of DNA synthesis in nasopharyngeal carcinoma cells. Chronobiol Int. 2002;19 (1 ):69–76. doi: 10.1081/cbi-120002590.11962687
28 He H, Lin K, Zou C, et al. Knockdown of annexin A2 enhances radiosensitivity by increasing G2/M-phase arrest, apoptosis and activating the p38 MAPK-HSP27 pathway in nasopharyngeal carcinoma. Front Oncol. 2022;12 :769544. doi: 10.3389/fonc.2022.769544.35371986
29 Karaboué A, Collon T, Pavese I, et al. Time-dependent efficacy of checkpoint inhibitor nivolumab: results from a pilot study in patients with metastatic non-small-cell lung cancer. Cancers. 2022;14 (4 ):896. doi: 10.3390/cancers14040896.35205644
30 Catozzi S, Assaad S, Delrieu L, et al. Early morning ­immune checkpoint blockade and overall survival of patients with metastatic cancer: an In-depth ­chronotherapeutic study. Eur J Cancer. 2024;199 :113571. doi: 10.1016/j.ejca.2024.113571.38301362
31 Molinari N, Daures J-P, Durand J-F. Regression splines for threshold selection in survival data analysis. Statist Med. 2001;20 (2 ):237–247. doi: 10.1002/1097-0258(20010130)20:2<237::AID-SIM654>3.0.CO;2-I.
32 Lévi F, Focan C, Karaboué A, et al. Implications of circadian clocks for the rhythmic delivery of cancer therapeutics. Adv Drug Deliv Rev. 2007;59 (9-10 ):1015–1035. doi: 10.1016/j.addr.2006.11.001.17692427
33 Münch M, Kramer A. Timing matters: new tools for personalized chronomedicine and circadian health. Acta Physiol (Oxf). 2019;227 (2 ):e13300. doi: 10.1111/apha.13300.31102499
34 Rahman S, Kraljević Pavelić S, Markova-Car E. Circadian (de)regulation in head and neck squamous cell carcinoma. Int J Mol Sci. 2019;20 (11 ):2662. doi: 10.3390/ijms20112662.31151182
35 Ortega-Campos SM, Verdugo-Sivianes EM, Amiama-Roig A, et al. Interactions of circadian clock genes with the hallmarks of cancer. Biochim Biophys Acta Rev Cancer. 2023;1878 (3 ):188900. doi: 10.1016/j.bbcan.2023.188900.37105413
36 Bjarnason GA, Jordan RC, Wood PA, et al. Circadian ­expression of clock genes in human oral mucosa and skin: association with specific cell-cycle phases. Am J Pathol. 2001;158 (5 ):1793–1801. doi: 10.1016/S0002-9440(10)64135-1.11337377
37 Xiao X-T, Wu Y-S, Chen Y-P, et al. Patterns and prognosis of regional recurrence in nasopharyngeal carcinoma after intensity-modulated radiotherapy. Cancer Med. 2023;12 (2 ):1399–1408. doi: 10.1002/cam4.5020.35822664
38 Lu T-Z, Zeng F-J, Hu Y-J, et al. Anatomic prognostic factors and their potential roles in refining M1 classification for de novo metastatic nasopharyngeal carcinoma. Cancer Med. 2023;12 (24 ):22091–22102. doi: 10.1002/cam4.6816.38073447
39 Dongxiang W, Liting L, Yujing L, et al. Prediction of outcomes in patients with local recurrent nasopharyngeal carcinoma: development and validation of a four-factor prognostic model integrating baseline characteristics and [18F]FDG PET/CT parameters. Eur Radiol. 2023;33 (4 ):2840–2849. doi: 10.1007/s00330-022-09232-1.36422647
40 Huang Y, Chen W, Haque W, et al. The impact of comorbidity on overall survival in elderly nasopharyngeal carcinoma patients: a National Cancer Data Base analysis. Cancer Med. 2018;7 (4 ):1093–1101. doi: 10.1002/cam4.1377.29493889
41 Neo J, Yip PL, Ong EHW, et al. Longitudinal post-radiotherapy plasma Epstein-Barr virus DNA trends inform on optimal risk stratification in endemic nasopharyngeal carcinoma. Oral Oncol. 2024;148 :106655. doi: 10.1016/j.oraloncology.2023.106655.38056062
42 Gou X, Duan B, Shi H, et al. The relations of dosimetric parameters with long-term outcomes and late toxicities in advanced T-stage nasopharyngeal carcinoma with IMRT. Head Neck. 2020;42 (1 ):85–92. doi: 10.1002/hed.25986.31650657
43 Lévi FA, Okyar A, Hadadi E, et al. Circadian regulation of drug responses: toward sex-specific and personalized chronotherapy. Annu Rev Pharmacol Toxicol. 2024;64 (1 ):89–114. doi: 10.1146/annurev-pharmtox-051920-095416.37722720
44 Cao Y, Wang RH. Associations among metabolism, circadian rhythm and age-associated diseases. Aging Dis. 2017;8 (3 ):314–333. doi: 10.14336/AD.2016.1101.28580187
