
==== Front
ESMO Open
ESMO Open
ESMO Open
2059-7029
Elsevier

S2059-7029(24)01477-7
10.1016/j.esmoop.2024.103707
103707
Original Research
Determining the optimal timing of adjuvant chemotherapy initiation after concurrent chemoradiotherapy in locoregionally advanced nasopharyngeal carcinoma
Cheng H. 12†
Chen J. 12†
Jia G. 12†
Liang Y. 12
Li Y. 12
Chen Y. 12
Lin J. 12
Wang P. 12
Chen Q. 12‡
Tang L. 12‡
Mai H. 12‡
Liu L. liult@sysucc.org.cn
12∗‡
1 State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangdong Key Laboratory of Nasopharyngeal Carcinoma Diagnosis and Therapy, Guangdong Provincial Clinical Research Center for Cancer, Sun Yat-sen University Cancer Center, Guangzhou
2 Department of Nasopharyngeal Carcinoma, Sun Yat-sen University Cancer Center, Guangzhou, People's Republic of China
∗ Correspondence to: Prof. Liting Liu, Department of Nasopharyngeal Carcinoma, Sun Yat-sen University Cancer Center, No. 651 Dongfeng East Road, Guangzhou 510060, Guangdong, People's Republic of China. Tel: +86-18688450757 liult@sysucc.org.cn
† These authors contributed equally to this work and share the first authorship.

‡ The senior authors contributed equally to this work.

09 9 2024
9 2024
09 9 2024
9 9 103707© 2024 The Author(s)
2024
https://creativecommons.org/licenses/by-nc-nd/4.0/ This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
Background

Studies on several malignancies have suggested that the time to commencement of adjuvant chemotherapy (AC) is associated with survival outcomes. There have, however, been no relevant reports of nasopharyngeal carcinoma (NPC).

Patients and methods

This clinical study examined newly diagnosed patients between April 2017 and December 2020. The primary endpoint was progression-free survival (PFS). Inverse probability of treatment weighting was used to control for confounding factors. Cox models with restricted cubic splines, Kaplan–Meier method and log-rank tests were used to evaluate the relationship between AC timing and survival.

Results

A total of 551 patients were identified [median age, 45 years (interquartile range 36-52 years); 383 (69.5%) male]. Restricted cubic splines demonstrated that the timing of AC initiation had a U-shaped association with PFS. The risk of disease progression decreased within 37 days and subsequently increased. From 37 to 90 days, each additional 7-day delay conferred worse PFS of 1.32 months {hazard ratio (HR): 1.14 [95% confidence interval (CI) 1.01-1.28], P = 0.04}. The cut-off value of the receiver operating characteristic curve for initiation was 69.5 days. At a median follow-up of 48 months, the PFS was significantly better in patients initiated within 69.5 days [HR: 2.18 (95% CI 1.17-4.06), log-rank P = 0.009], with a higher 3-year rate [78.8% (95% CI 75.1% to 82.7%) versus 59.0% (95% CI 42.2% to 82.5%)] than beyond 69.5 days. Positive results were also observed in secondary endpoints. The initiation group was an independent prognostic factor [HR: 2.28 (95% CI 1.42-3.66), P < 0.001].

Conclusions

The optimal timing of AC initiation is ∼37 days after concurrent chemoradiotherapy in patients with locoregionally advanced nasopharyngeal carcinoma. A delay beyond 69.5 days is associated with compromised survival. Efforts should be made to address the reasons for delays and ensure the timely initiation of AC.

Highlights

• There is a U-shaped association between adjuvant chemotherapy initiating time and the risk of disease progression.

• The optimal timing of adjuvant chemotherapy initiation is approximately 37 days in patients with nasopharyngeal carcinoma.

• A delay initiation beyond 69.5 days is associated with compromised survival.

Key words

locoregionally advanced nasopharyngeal carcinoma
adjuvant chemotherapy initiation
optimal timing
progression-free survival
==== Body
pmcIntroduction

Nasopharyngeal carcinoma (NPC) is an epithelial carcinoma originating from the nasopharyngeal mucosal lining, which is often observed in the pharyngeal recess.1 More than 70% of patients are diagnosed with locoregionally advanced disease.2 Concurrent chemoradiotherapy (CCRT) with platinum-based agents constitutes the backbone of treatment.3 A meta-analysis in 2022 suggested that addition of induction or adjuvant chemotherapy (AC) to CCRT improved tumor control and survival.4 Although most patients achieve complete clinical remission after definitive chemoradiotherapy, 20%-30% of patients develop disease progression within 3 years.5 Induction chemotherapy (IC) appeared to be most effective against distant metastasis, while AC showed the greatest impact on locoregional progression.6

In terms of AC, the landmark Intergroup 0099 trial was the first to report the significant therapeutic benefit of cisplatin-based CCRT plus adjuvant cisplatin–fluorouracil chemotherapy compared with radiotherapy alone.7 Subsequently, similar trials in endemic populations have achieved consistent results, making CCRT followed by AC the standard treatment of locoregionally advanced NPC (LA-NPC).8, 9, 10

The optimal timing of AC after CCRT remains poorly defined. There is a widespread clinical assumption that AC should be initiated as soon as possible. Patients vary considerably, however, in their ability to tolerate AC during recovery from CCRT. Previous studies have tended to initiate AC 28-90 days after CCRT, but there is no consensus. In other malignancies, several studies have suggested that the time to commencement of AC is associated with survival outcomes. A 2012 landmark meta-analysis of colorectal cancer reported that the initiation of postoperative AC within 3 weeks reduced the recurrence rate, and overall survival (OS) decreased by 14% for every 4-week delay.11 Salazar et al.12 concluded that AC was still effective at 7-18 weeks after the resection of non-small-cell lung cancer, challenging the general consensus that postoperative AC should be started within 6 weeks.

To date, there have been no studies on the optimal timing of AC initiation in patients with LA-NPC. We conducted this study to examine the association between the timing of AC initiation and survival outcomes, with the hypothesis that an optimal threshold could be determined.

Methods

Study population

The Single Disease Database of the Sun Yat-sen University Cancer Center was searched for newly diagnosed patients between April 2017 and December 2020. Our eligibility criteria included: histologically confirmed NPC of non-keratinizing carcinoma [World Health Organization (WHO) types II and III]; staged III-IVa diseases according to the eight edition of the American Joint Committee on Cancer staging system; aged 18-70 years old; receipt of platinum-based CCRT followed by AC; Eastern Cooperative Oncology Group (ECOG) performance status score of 0-1; satisfactory bone marrow, liver, and kidney function; no history of other malignancies; complete treatment information; and available results for Epstein–Barr virus (EBV) DNA level. The key exclusion criteria included metastasis or recurrence of NPC at the time of diagnosis; lactation or pregnancy; insufficient clinical data; or severe coexisting illnesses. To minimize the influence of outliers, patients within the top and bottom 2% of the time to AC were excluded,12 resulting in the final distribution as shown in Figure 1.Figure 1 Flow chart of the patient selection process. AC, adjuvant chemotherapy; CCRT, concurrent chemoradiotherapy; NPC, nasopharyngeal carcinoma; SYSUCC, Sun Yat-sen University Cancer Center; WHO, World Health Organization.

All patients underwent pretreatment evaluation, including complete patient history, physical examination, hematology and biochemistry profiles, magnetic resonance imaging (MRI) of the nasopharynx and neck, whole-body computed tomography (CT), chest radiography, abdominal sonography, and bone scintigraphy or positron emission tomography-CT (PET-CT).

Treatment

All patients received CCRT followed by AC. The patients underwent intensity-modulated radiotherapy. The gross tumor volume included the primary tumor and enlarged lymph nodes. Five daily fractions of an accumulated dose of 68-70 Gy were prescribed per week. Concurrent platinum-based chemotherapy (100 mg/m2 every 3 weeks or 40 mg/m2 every week) was administered simultaneously with radiotherapy. The AC was administered either intravenously (i.v.) or orally. I.V. AC regimens were as follows: PF [cisplatin (80 mg/m2, day 1), 5-fluorouracil (4 g/m2, continuous i.v. infusion for 96 h), every 4 weeks] and GP [gemcitabine (1.0 g/m2, days 1 and 8), cisplatin (80 mg/m2, day 1), every 3 weeks] for up to three cycles. Oral administration included capecitabine (1000 mg/m2 b.i.d., days 1-14) and S-1 (40-60 mg b.i.d., days 1-14) for at least two cycles. The dose of S-1 was determined according to the body surface area (BSA): 40 mg twice a day for BSA <1.25 m2; 50 mg twice a day for 1.25 m2 ≤BSA <1.5 m2; and 60 mg twice a day for BSA ≥1.5 m2. Chemotherapy dose adjustments were permitted for adverse events [Common Terminology Criteria for Adverse Events (CTCAE), version 5.0], but AC would not be initiated until the adverse events recover to grade <2.

A total of 184 participants (33.3%) in the cohort received IC. The IC regimens were listed as follows: TPF [docetaxel (60 mg/m2, day 1), cisplatin (60 mg/m2, day 1), and 5-fluorouracil (3 g/m2, continuous i.v. infusion for 120 h)], PF [cisplatin (75 mg/m2, day 1), 5-fluorouracil (4 g/m2)], TP [docetaxel (75 mg/m2, day 1) or paclitaxel (135 mg/m2, day 1) and cisplatin (75 mg/m2, day 1)], GP [gemcitabine (1.0 g/m2, days 1 and 8), cisplatin (80 mg/m2, day 1)] for two to three cycles.

Follow-up and outcomes

Patients were followed up every 3 months in the first 3 years and then every 6 months until 31 December 2023. Posttreatment locoregional recurrence or distant metastasis were confirmed by pathological examination or imaging (MRI, CT, PET-CT, abdominal sonography, and bone scintigraphy).

The primary endpoint was progression-free survival (PFS), defined as the time interval between the initial date of diagnosis and disease progression or death from any cause. The secondary endpoints were OS (interval from the first date of diagnosis to death from any cause), locoregional recurrence-free survival (LRRFS), and distant metastasis-free survival (DMFS), which corresponded to the time interval between the initial date of diagnosis and the first locoregional recurrence or distant metastasis or death from any cause.

Statistical analysis

To model the association between AC initiation time and PFS, we developed a multivariable Cox proportional hazards model with restricted cubic splines (RCS). RCS is a smooth connection of polynomial functions under the assumption of a non-linearity relationship.13 With outstanding advantages in the recognition of risk function inflection points, while adjusting for confounding factors, the RCS has been widely applied to survival data analysis.14, 15, 16 We tested non-linearity using the Martingale residuals (Supplementary Figure S1, available at https://doi.org/10.1016/j.esmoop.2024.103707), and defined the spline with four knots.17 The following clinical characteristics were included in the multivariate analysis: sex, age [median age 45 years (interquartile range, IQR 36-52 years)], family history of NPC, tumor stages (T and N stages separately, and overall stage), schemes (with or without IC), AC regimens (i.v. or oral administration), pretreatment plasma EBV DNA (categorized in ≤4000 copies/ml and >4000 copies/ml18), pre-AC plasma EBV DNA.

The cohort was then divided into early and late groups based on the cut-off value of the time to AC initiation, which was implemented through a receiver operating characteristic (ROC) curve. The distribution of clinical factors between the two groups was compared using the chi-square test. To minimize the bias arising from clinical characteristics, we used inverse probability of treatment weighting (IPTW) in our study.19 Survival curves and outcomes were analyzed using the Kaplan–Meier method and the log-rank test. Hazard ratios (HRs) were calculated using univariate Cox regression analysis. A multivariable Cox proportional hazards model was used to evaluate the independence of the prognostic values. P values <0.05 (two-sided) were considered statistically significant. Statistical analyses were carried out using SPSS 27.0 (IBM Corp., Armonk, NY) and R software 4.3.2 (R Core Team, Vienna, Austria), R packages: rms, version 6.7-1; plotRCS, version 0.1.5; survival, version 3.5-7; RISCA, version 1.0.4; and survminer, version 0.4.9.

Results

Patient characteristics of the total cohort

A total of 551 patients were enrolled in this study. The median age was 45 years (IQR 36-52 years) and 383 (69.5%) participants were men. The median total cisplatin dose was 440 mg/m2 (IQR 300-540 mg/m2), while the median concurrent cisplatin dose was 240 mg/m2 (IQR 200-300 mg/m2).20,21 Before AC initiation, a total of 336 patients (61%) experienced grade 3 or higher adverse events (CTCAE 5.0). Approximately 85% of patients adhered fully to the clinician’s recommendations on initiation timing, whereas 9% delayed AC for personal reasons. After a median follow-up period of 48 months (IQR 38-59 months), there were 132 (24.0%) cases of disease progression and 53 (9.6%) deaths. The failure site distribution included 36 locoregional recurrences, 80 distant metastases alone, and 16 combined cases. The median time to AC initiation was 36 days (IQR 30-48 days). To explore the relationship between AC timing and PFS in patients with LA-NPC, an adjusted Cox model was created with AC timing separated into 7-day intervals (Figure 2A). Patients in the 36-day (median time) interval were used as references. No linear correlation was found between progression risk and AC timing.Figure 2 Relationship between time to adjuvant chemotherapy initiation and progression risk. (A) Progression risk associated with weeks to adjuvant chemotherapy initiation. The reference period was 6 weeks. Dots represent the log of hazard ratios and whiskers represent 95% CIs. (B) Restricted cubic spline modeling of the relationship between days to adjuvant chemotherapy initiation and progression risk. The risk function shows an inflection point on day 37. Hazard ratios derived from the multivariate Cox model. The shaded areas represent the 95% CIs of the adjusted hazard ratios. (C and D) Restricted cubic spline modeling of patients with N3 stage (C), and pretreatment EBV DNA >4000 copies/ml (D). The risk functions show inflection point at 37 days. Hazard ratios derived from the multivariate Cox model, and the shaded areas represent the 95% CIs of the adjusted hazard ratios. CI, confidence interval; EBV, Epstein–Barr virus.

RCS identifying the association between adjuvant timing and PFS

A multivariable-adjusted Cox hazard model with RCS demonstrated an inflection point in the risk function on day 37, which was considered the optimal timing (Figure 2B). A U-shaped association was observed between the time of AC initiation and PFS. The hazard of disease progression decreased rapidly before the model-derived 37 days but increased with the time of the delay to AC afterward. Among the patients who received AC between 37 and 90 days, each additional 7-day delay conferred a worse PFS of 1.32 months {HR: 1.14 [95% confidence interval (CI) 1.01-1.28], P = 0.04}. We carried out subgroup analysis to explore whether there was a similar trend in optimal AC timing for high-risk patients. The RCS demonstrated U-shaped associations with inflection points at 37 days for patients with N3 stage, and pretreatment EBV DNA >4000 copies/ml, respectively (Figure 2C and D).

Comparison of baseline characteristics between patients with early and late initiation

Based on the cut-off value of 69.5 days determined by the ROC curve, 468 patients were assigned to the early group (≤69.5 days) and 83 patients to the late group (>69.5 days). Compared with the early initiation group, patients in the late initiation group were more likely to receive IC (P < 0.001), and oral adjuvant regimens (P < 0.001). They also had a higher proportion of pretreatment plasma EBV DNA >4000 copies/ml (P < 0.001). After IPTW adjustment, however, all listed co-variates were well-balanced between groups (Table 1).Table 1 Baseline characteristics of the early and late initiation groups before and after weighting

Characteristic	Unweighted, n %	Weighted, n %	
Early initiation	Late initiation	P value	Early initiation	Late initiation	P value	
n = 468	n = 83	
Sex			0.05			0.20	
 Female	135 (29)	33 (40)		(31)	(44)		
 Male	333 (71)	50 (60)		(69)	(56)		
Age			0.47			0.76	
 <45 years	229 (49)	37 (45)		(48)	(45)		
 ≥45 years	239 (51)	46 (55)		(52)	(55)		
Family history of NPC			0.78			>0.99	
 No	430 (92)	77 (93)		(92)	(92)		
 Yes	38 (8)	6 (7)		(8)	(8)		
T stagea			0.82			0.07	
 T1-2	47 (10)	9 (11)		(10)	(5)		
 T3-4	421 (90)	74 (89)		(90)	(95)		
N stagea			0.09			0.23	
 N0-1	101 (22)	25 (30)		(23)	(35)		
 N2-3	367 (78)	58 (70)		(77)	(65)		
Total stagesa			0.08			0.23	
 III	206 (44)	28 (34)		(42)	(54)		
 IVa	262 (56)	55 (66)		(58)	(46)		
Schemes			<0.001			0.40	
 CCRT + AC	346 (74)	21 (25)		(66)	(59)		
 IC + CCRT + AC	122 (26)	62 (75)		(34)	(41)		
Adjuvant regimen			<0.001			0.59	
 Intravenous administration	277 (59)	8 (10)		(52)	(46)		
 Oral administration	191 (41)	75 (90)		(48)	(54)		
Pretreatment EBV DNA			<0.001			0.12	
 ≤4000 copies/ml	334 (71)	42 (51)		(68)	(52)		
 >4000 copies/ml	134 (29)	41 (49)		(32)	(48)		
Pre-AC EBV DNA			0.48			0.24	
 0 copies/ml	429 (92)	78 (94)		(92)	(80)		
 >0 copies/ml	39 (8)	5 (6)		(8)	(20)		
The chi-square test was used to calculate P values. All variables were transformed into categorical variables.

AC, adjuvant chemotherapy; CCRT, concurrent chemoradiotherapy; EBV, Epstein-Barr virus; IC, induction chemotherapy; NPC, nasopharyngeal carcinoma.

a According to the eighth edition of UICC/AJCC staging system.

Prognostic value of timing for initiating AC

At the time of analysis, the IPTW-adjusted Kaplan–Meier curves illustrated that PFS at 3 years was 78.8% (95% CI 75.1% to 82.7%) in the early group, compared with 59.0% (95% CI 42.2% to 82.5%) in the late group [HR: 2.18 (95% CI 1.17-4.06), log-rank P = 0.009]. Significantly higher 3-year OS rate [93.4% versus 88.0%, HR: 2.43 (95% CI 1.07-5.53), log-rank P = 0.047], LRRFS rate [87.9% versus 73.4%, HR: 2.22 (95% CI 1.15-4.31), log-rank P = 0.03], and DMFS rate [83.0% versus 72.5%, HR: 1.97 (95% CI 0.97-4.03), log-rank P = 0.04] were also observed in the early initiation group (Figure 3).Figure 3 Kaplan–Meier curves of survival outcomes. Kaplan–Meier curves of PFS (A), OS (B), LRRFS (C), and DMFS (D). The inverse probability of treatment weighting-adjusted Kaplan–Meier curves of PFS (E), OS (F), LRRFS (G), and DMFS (H). A univariate Cox proportional hazards model was used to calculate the HRs and 95% CIs. CI, confidence interval; DMFS, distant metastasis-free survival; HR, hazard ratio; LRRFS, locoregional recurrence-free survival; OS, overall survival; PFS, progression-free survival.

After weighting, the initiation time significantly affected PFS [HR: 2.28 (95% CI 1.42-3.66), P < 0.001], OS [HR: 2.44 (95% CI 1.31-4.56), P = 0.005], LRRFS [HR: 2.89 (95% CI 1.52-5.51), P = 0.001], and DMFS [HR: 1.80 (95% CI 1.01-2.88), P = 0.02]. In addition, the overall stage [HR: 2.54 (95% CI 1.40-4.60), P = 0.002], schemes [HR: 0.55 (95% CI 0.32-0.92), P = 0.02], pretreatment plasma EBV DNA level [HR: 2.12 (95% CI 1.25-3.60), P = 0.005], and pre-AC plasma EBV DNA level [HR: 4.27 (95% CI 2.06-8.85), P < 0.001] were also independent for PFS (Table 2). Detailed information on the IPTW-adjusted multivariate and univariate Cox regression analysis are summarized in Supplementary Tables S1 and S2, available at https://doi.org/10.1016/j.esmoop.2024.103707.Table 2 Multivariate Cox regression analysis of PFS before and after weighting

Characteristic	Unweighted	Weighted	
HR (95% CI)	P value	HR (95% CI)	P value	
Sex	1.35 (0.93-1.97)	0.11	0.75 (0.46-1.23)	0.26	
 Female	Reference	
 Male		
Age (years)	0.97 (0.70-1.35)	0.86	1.12 (0.68-1.87)	0.65	
 <45	Reference	
 ≥45		
Family history of NPC	0.96 (0.54-1.71)	0.88	0.75 (0.40-1.41)	0.37	
 None	Reference	
 Yes		
T stagea	1.38 (0.76-2.53)	0.29	0.97 (0.47-2.00)	0.93	
 T1-2	Reference	
 T3-4		
N stagea	1.40 (0.88-2.22)	0.16	0.93 (0.46-1.91)	0.85	
 N0-1	Reference	
 N2-3		
Total stagesa	1.63 (1.13-2.37)	0.009	2.54 (1.40-4.60)	0.002	
 III	Reference	
 IVa		
Schemes	0.77 (0.49-1.20)	0.25	0.55 (0.32-0.92)	0.02	
 CCRT + AC	Reference	
 IC + CCRT + AC		
Adjuvant regimen	1.05 (0.68-1.62)	0.82	0.76 (0.42-1.39)	0.37	
 Intravenous administration	Reference	
 Oral administration		
Pretreatment EBV DNA (copies/ml)	1.36 (0.96-1.92)	0.08	2.12 (1.25-3.60)	0.005	
 ≤4000	Reference	
 >4000		
Pre-AC EBV DNA (copies/ml)	2.63 (1.64-4.20)	<0.001	4.27 (2.06-8.85)	<0.001	
 0	Reference	
 >0		
Group	1.91 (1.21-3.00)	0.005	2.28 (1.42-3.66)	<0.001	
 Early initiation	Reference	
 Late initiation		
A Cox proportional hazards regression model was used to perform multivariate analysis. HRs and their 95% CIs were calculated for sex (male versus female); age in years (≥45 versus <45); family history of NPC (yes versus no); T stage (3-4 versus 1-2); N stage (2-3 versus 0-1); overall stage (IVa versus III); schemes (IC + CCRT + AC versus CCRT + AC); adjuvant regimen (oral administration versus intravenous administration); pretreatment EBV DNA (copies/ml) (>4000 versus ≤4000); pre-AC EBV DNA (copies/ml) (>0 versus =0); initiation group (late initiation versus early initiation).

AC, adjuvant chemotherapy; CCRT, concurrent chemoradiotherapy; CI, confidence interval; EBV, Epstein-Barr virus; HR, hazard ratio; IC, induction chemotherapy; NPC, nasopharyngeal carcinoma; PFS, progression-free survival.

a According to the eighth edition of UICC/AJCC staging system.

Discussion

To our knowledge, this is the first study to explore the association between the time to AC initiation and prognosis in LA-NPC. We determined that the optimal initiation time was 37 days, and delaying beyond 69.5 days was detrimental to survival outcomes.

Since the Intergroup 0099 trial, our understanding of AC has evolved over the past few decades. The combination of CCRT + AC is the IIA recommendation of the National Comprehensive Cancer Network (NCCN) guidelines (version 2.2024). In 2016, after evaluating the seven most common treatments for LA-NPC, a network meta-analysis emphasized that the concurrent adjuvant sequence achieved the highest survival benefit and consistent improvement in all survival outcomes.22 Although Chen et al.23 reported that adjuvant cisplatin–fluorouracil chemotherapy did not significantly improve failure-free survival in their previous study, they recently suggested that the addition of metronomic oral capecitabine following CCRT achieved improved survival outcomes and a manageable safety profile.24 Two other studies drew similar conclusions regarding the use of adjuvant capecitabine and S-1 in patients with high-risk factors, respectively.25,26 In 2023, our previous study revealed encouraging results for the adjuvant gemcitabine–cisplatin regimen following CCRT over traditional cisplatin–fluorouracil.27 The NPC-0502 trial, however, did not confirm the superiority of adjuvant gemcitabine–cisplatin chemotherapy in patients with detectable plasma EBV DNA after 6-8 weeks of CCRT.28 But there was a 12-week interval between the end of radiotherapy and AC initiation, which might not have been sufficient to eradicate occult recurrence and metastasis. Despite numerous prospective studies and meta-analyses, the results are inconsistent, partly because of heterogeneity among studies. Therefore, the undefined timing of AC initiation, different patients with variable tumor stages and EBV statuses, diverse systemic schemes, and radiation techniques should be considered.29,30

The latest meta-analysis concluded that AC showed the greatest impact on locoregional progression, while IC appeared to be the most effective against distant metastasis.6 In 2019, Zhang et al.5 reported a superior 3-year PFS rate with gemcitabine plus cisplatin combined with CCRT group compared with standard CCRT (85% versus 77%), as well as an improved OS rate (95% versus 90%). Among the existing large clinical trials, the 3-year PFS rates in the experimental arms were all ∼85%.5,24,27 Despite the focus on patients with LA-NPC, the eligibility criteria for these studies were heterogenous. The trial by Zhang et al.5 included patients with stage III to IVB disease (American Joint Committee on Cancer, seventh edition), excluding those with N0 stage.5 The trial by Chen et al. further excluded low-risk patients with T3N1 stage,24 while our previous study only included high-risk patients staged N2-3.27 As there was insufficient evidence from a head-to-head comparison between IC and AC, it remains unclear which treatment is superior. The preliminary results of our studies (NCT03306121), directly compared IC + CCRT with CCRT + AC based on PF regimen, and found no significant difference in survival between the two groups.

The precise time for the commencement of AC has not yet been established. For a similar adjuvant regimen of gemcitabine–cisplatin, the positive results of our trial countered those of the HK-NPC-0502 trial, and the differentiating factor was posited to be the initiation time (4 weeks in our study versus 12 weeks in Chan’s trial).31 The NCCN guidelines (version 2.2024) recommend a 4-week interval between CCRT and AC. Due to the complexity and diversity of clinical practice, however, AC initiation is often delayed. Possible factors include slow recovery from the cumulative toxicities of CCRT, increased age, and poor compliance. Currently, without reliable guidelines and references, oncologists empirically initiate AC within a time frame of 28 to 90 days. As evidence from retrospective studies has accumulated for other cancers,32, 33, 34 the optimal timing for LA-NPC requires investigation.

In this retrospective study, we constructed an adjusted RCS model to examine the relationship between the time to AC initiation and PFS. A U-shaped association was observed, with the lowest risk occurring at 37 days, which was much alike to the postoperative AC timing in non-small-cell lung cancer.12 A similar trend of optimal timing was observed in patients with high risk factors (N3 stage or high pretreatment EBV DNA level). A reduction in hazard was observed within 5 weeks after CCRT, which was attributed to the gradual recovery from cumulative toxicities. This finding indicates that it may be better to fully optimize patients to increase the probability of completing adjuvant courses. From 5 to 12 weeks, as the risk increased, PFS decreased by 1.32 months for each week of delay. To further explore the relationship between delayed AC and survival, we assigned the participants to early and late initiation groups based on the cut-off value of 69.5 days (10 weeks) derived from the ROC curve. After IPTW adjustment, the differences between the two groups were eliminated, and the initiation group was still an independent prognostic factor for all survival endpoints. In addition, the few cases of T1-2 stages in our locoregionally advanced cohort limited the predictive power of T stages for LRRFS. Regarding the IPTW-adjusted survival outcomes, patients who received early AC within 69.5 days had significantly better PFS, OS, LRRFS, and DMFS than those who procrastinated beyond this threshold. This finding was consistent with the 10-year outcome of 9901 and 9902 combined analysis, which concluded that CCRT + AC significantly improved OS and disease control.35 A widely accepted assumption for these mechanisms is the eradication of micro-metastatic deposits in a proportion of patients who would otherwise develop recurrence. According to the Gompertzian growth curve,36 early administration is associated with limited tumor burden and ideal chemotherapeutic sensitivity. Furthermore, because a delay in initiation has been correlated with poor survival, identifying the reasons for the delay is essential. Through medical records and telephone follow-ups, we found that ∼6% of patients had not fully recovered to grade <2 adverse events 37 days after CCRT. Additionally, 85% of patients reported initiating AC based on clinician advice, while 9% experienced delays for personal reasons (living far away or individual schedules). Recent evidence has emphasized the importance of nutritional intervention,37 which may contribute to organ function improvement and toxicity recovery, thus avoiding unnecessary AC delays. Other relating factors, however, require further investigation.

This study has several limitations that should be addressed. Firstly, this was carried out at a single center and was retrospective in nature, which might restrict the generalizability of the results. Secondly, the number of patients in the late initiation group was insufficient, so validation with a larger sample size is required. Thirdly, the inability to compare specific chemotherapeutic regimens should be considered. Finally, there were various reasons for the delays, which may have affected the results.

In conclusion, the optimal timing of AC initiation was ∼37 days after CCRT for LA-NPC. Delayed administration beyond 69.5 days led to inferior survival outcomes. Efforts should be made to address the reasons for delays and ensure the timely initiation of AC. Further studies are required to confirm these findings.

Supplementary data

Supplementary data

Acknowledgements

We thank all the patients who participated in the study and their families. We thank Elsevier language editing service for manuscript editing during drafting.

Funding

This work was supported by grants from the 10.13039/501100012166 National Key Research and Development Program of China [grant numbers 2022YFC2505800, 2022YFC2705005], 10.13039/501100001809 National Natural Science Foundation of China [grant numbers 32200651, 82203776, 82203125, 82222050, 82272739, 82272882, 82173287, 82073003, 82003267, 82002852, 82373258, 82361168664, 82303967], 10.13039/501100021171 Guangdong Basic and Applied Basic Research Foundation [grant number 2021B1515230002], Science and Technology Program of Guangzhou [grant numbers 202201011561, 2023A04J2127, 2024B03J1248], 10.13039/501100002402 Sun Yat-sen University Clinical Research 5010 Program [grant numbers 201315, 2015021, 2017010, 2019023], 10.13039/100020732 Innovative Research Team of High-level Local Universities in Shanghai [grant number SSMU-ZLCX20180500], Postdoctoral Innovative Talent Support Program [grant number BX20220361], Planned Science and Technology Project of Guangdong Province [grant number 2019B020230002], Key Youth Teacher Cultivating Program of 10.13039/501100002402 Sun Yat-sen University [grant number 20ykzd24], and Fundamental Research Funds for the Central Universities.

Disclosure

The authors have declared no conflicts of interest.

Data sharing

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Ethics approval and consent to participate

The study was approved by the Research Ethics Committee of the Cancer Center of Sun Yat-sen University (Guangdong, China), and carried out according to the ethical principles of the Declaration of Helsinki.
==== Refs
References

1 Chen Y.P. Chan A.T.C. Le Q.T. Blanchard P. Sun Y. Ma J. Nasopharyngeal carcinoma Lancet 394 2019 64 80 31178151
2 Mao Y.P. Xie F.Y. Liu L.Z. Re-evaluation of 6th edition of AJCC staging system for nasopharyngeal carcinoma and proposed improvement based on magnetic resonance imaging Int J Radiat Oncol Biol Phys 73 2009 1326 1334 19153016
3 Chen Y.P. Ismaila N. Chua M.L.K. Chemotherapy in combination with radiotherapy for definitive-intent treatment of stage II-IVA nasopharyngeal carcinoma: CSCO and ASCO guideline J Clin Oncol 39 2021 840 859 33405943
4 Blanchard P. Lee A.W.M. Carmel A. Meta-analysis of chemotherapy in nasopharynx carcinoma (MAC-NPC): an update on 26 trials and 7080 patients Clin Transl Radiat Oncol 32 2022 59 68 34935776
5 Zhang Y. Chen L. Hu G.Q. Gemcitabine and cisplatin induction chemotherapy in nasopharyngeal carcinoma N Engl J Med 381 2019 1124 1135 31150573
6 Petit C. Lee A. Ma J. Role of chemotherapy in patients with nasopharynx carcinoma treated with radiotherapy (MAC-NPC): an updated individual patient data network meta-analysis Lancet Oncol 2 4 2023 611 623
7 Al-Sarraf M. LeBlanc M. Giri P.G. Chemoradiotherapy versus radiotherapy in patients with advanced nasopharyngeal cancer: phase III randomized Intergroup study 0099 J Clin Oncol 16 1998 1310 1317 9552031
8 Lee A.W. Lau W.H. Tung S.Y. Preliminary results of a randomized study on therapeutic gain by concurrent chemotherapy for regionally-advanced nasopharyngeal carcinoma: NPC-9901 trial by the Hong Kong Nasopharyngeal Cancer Study Group J Clin Oncol 23 2005 6966 6975 16192584
9 Wee J. Tan E.H. Tai B.C. Randomized trial of radiotherapy versus concurrent chemoradiotherapy followed by adjuvant chemotherapy in patients with American Joint Committee on Cancer/International Union against cancer stage III and IV nasopharyngeal cancer of the endemic variety J Clin Oncol 23 2005 6730 6738 16170180
10 Chen Y. Liu M.Z. Liang S.B. Preliminary results of a prospective randomized trial comparing concurrent chemoradiotherapy plus adjuvant chemotherapy with radiotherapy alone in patients with locoregionally advanced nasopharyngeal carcinoma in endemic regions of China Int J Radiat Oncol Biol Phys 71 2008 1356 1364 18472356
11 Biagi J.J. Raphael M.J. Mackillop W.J. Kong W. King W.D. Booth C.M. Association between time to initiation of adjuvant chemotherapy and survival in colorectal cancer: a systematic review and meta-analysis JAMA 305 2011 2335 2342 21642686
12 Salazar M.C. Rosen J.E. Wang Z. Association of delayed adjuvant chemotherapy with survival after lung cancer surgery JAMA Oncol 3 2017 610 619 28056112
13 Molinari N. Daures J.P. Durand J.F. Regression splines for threshold selection in survival data analysis Stat Med 20 2001 237 247 11169599
14 Sun Z. Adam M.A. Kim J. Determining the optimal timing for initiation of adjuvant chemotherapy after resection for stage II and III colon cancer Dis Colon Rectum 59 2016 87 93 26734965
15 Bhaskaran K. Dos-Santos-Silva I. Leon D.A. Douglas I.J. Smeeth L. Association of BMI with overall and cause-specific mortality: a population-based cohort study of 3.6 million adults in the UK Lancet Diabetes Endocrinol 6 2018 944 953 30389323
16 Lee D.H. Keum N. Hu F.B. Predicted lean body mass, fat mass, and all cause and cause specific mortality in men: prospective US cohort study BMJ 362 2018 k2575
17 Durrleman S. Simon R. Flexible regression models with cubic splines Stat Med 8 1989 551 561 2657958
18 Li X.Y. Luo D.H. Guo L. Deintensified chemoradiotherapy for pretreatment Epstein-Barr virus DNA-selected low-risk locoregionally advanced nasopharyngeal carcinoma: a phase II randomized noninferiority trial J Clin Oncol 40 2022 1163 1173 34990291
19 Takada K. Shimokawa M. Takamori S. Clinical impact of probiotics on the efficacy of anti-PD-1 monotherapy in patients with nonsmall cell lung cancer: A multicenter retrospective survival analysis study with inverse probability of treatment weighting Int J Cancer 149 2021 473 482 33720422
20 Lv J.W. Qi Z.Y. Zhou G.Q. Optimal cumulative cisplatin dose in nasopharyngeal carcinoma patients receiving additional induction chemotherapy Cancer Sci 109 2018 751 763 29266526
21 Liu S.L. Sun X.S. Yan J.J. Optimal cumulative cisplatin dose in nasopharyngeal carcinoma patients based on induction chemotherapy response Radiother Oncol 137 2019 83 94 31078941
22 Ribassin-Majed L. Marguet S. Lee A.W.M. What is the best treatment of locally advanced nasopharyngeal carcinoma? An individual patient data network meta-analysis J Clin Oncol 35 2017 498 505 27918720
23 Chen L. Hu C.S. Chen X.Z. Concurrent chemoradiotherapy plus adjuvant chemotherapy versus concurrent chemoradiotherapy alone in patients with locoregionally advanced nasopharyngeal carcinoma: a phase 3 multicentre randomised controlled trial Lancet Oncol 13 2012 163 171 22154591
24 Chen Y.P. Liu X. Zhou Q. Metronomic capecitabine as adjuvant therapy in locoregionally advanced nasopharyngeal carcinoma: a multicentre, open-label, parallel-group, randomised, controlled, phase 3 trial Lancet 398 2021 303 313 34111416
25 Miao J. Wang L. Tan S.H. Adjuvant capecitabine following concurrent chemoradiotherapy in locoregionally advanced nasopharyngeal carcinoma: a randomized clinical trial JAMA Oncol 8 2022 1776 1785 36227615
26 Liu Y.C. Wang W.Y. Twu C.W. Prognostic impact of adjuvant chemotherapy in high-risk nasopharyngeal carcinoma patients Oral Oncol 64 2017 15 21 28024719
27 Liu L.T. Liu H. Huang Y. Concurrent chemoradiotherapy followed by adjuvant cisplatin-gemcitabine versus cisplatin-fluorouracil chemotherapy for N2-3 nasopharyngeal carcinoma: a multicentre, open-label, randomised, controlled, phase 3 trial Lancet Oncol 24 2023 798 810 37290468
28 Chan A.T.C. Hui E.P. Ngan R.K.C. Analysis of plasma Epstein-Barr virus DNA in nasopharyngeal cancer after chemoradiation to identify high-risk patients for adjuvant chemotherapy: a randomized controlled trial J Clin Oncol 2018 10.1200/JCO.2018.77.7847
29 Chua M.L.K. Circulating tumor DNA to personalize treatment in nasopharynx cancer – time to look "ahead"? Int J Radiat Oncol Biol Phys 104 2019 362 364 31047626
30 Park J.C. Chan A.W. Wirth L.J. Controversies regarding the management of locoregionally advanced nasopharyngeal carcinoma: are we asking the right questions? J Oncol Pract 14 2018 603 605 30312565
31 Yip P.L. Lee A.W.M. Chua M.L.K. Adjuvant chemotherapy in nasopharyngeal carcinoma Lancet Oncol 24 2023 713 715 37290469
32 Chavez-MacGregor M. Clarke C.A. Lichtensztajn D.Y. Delayed initiation of adjuvant chemotherapy among patients with breast cancer JAMA Oncol 2 2016 322 329 26659132
33 Bos A.C. van Erning F.N. van Gestel Y.R. Timing of adjuvant chemotherapy and its relation to survival among patients with stage III colon cancer Eur J Cancer 51 2015 2553 2561 26360411
34 Jue J.S. Koru-Sengul T. Miao F. Timing of adjuvant chemotherapy and overall survival following radical cystectomy Urol Oncol 38 2020 75.e15 75.e22
35 Ng W.T. Tung S.Y. Lee V. Concurrent-adjuvant chemoradiation therapy for stage III-IVB nasopharyngeal carcinoma-exploration for achieving optimal 10-year therapeutic ratio Int J Radiat Oncol Biol Phys 101 2018 1078 1086 29885997
36 Gompertzian growth curve Schwab M. Encyclopedia of Cancer 2011 Springer Berlin, Heidelberg 1576
37 Deng J. He Y. Sun X.S. Construction of a comprehensive nutritional index and its correlation with quality of life and survival in patients with nasopharyngeal carcinoma undergoing IMRT: a prospective study Oral Oncol 98 2019 62 68 31541928
