Introduction

Nasopharyngeal carcinoma (NPC) originates from the epithelial cells lining the nasopharynx and is highly prevalence in specific regions, particularly Southeast Asia and Southern China [1, 2]. This malignancy is closely associated with Epstein-Barr virus (EBV) infection and often metastasizes to regional lymph nodes [3]. At the time of diagnosis, the majority of patients (approximately 70–80%) present with locally advanced NPC (LANPC) due to its subtle onset [3]. The standard treatment approach for LANPC involves induction chemotherapy (IC) followed by concurrent chemoradiotherapy (CCRT) [4]. IC is utilized to reduce tumor burden, minimize distant metastasis risk, and enhance treatment outcomes [5]. However, despite treatment advancements, LANPC prognosis remains suboptimal, with a significant portion of patients experiencing disease recurrence [3]. Therefore, there is a pressing need for innovative therapeutic approaches to improve treatment efficacy.

Nimotuzumab (NTZ), a humanized monoclonal antibody targeting the epidermal growth factor receptor (EGFR), has shown promise in treating various cancers, including LANPC [6,7,8,9]. Previous studies have indicated that combining NTZ with standard chemotherapy or radiotherapy could enhance treatment efficacy and improve survival rates in LANPC patients [10,11,12,13]. Our recent study similarly revealed that adding NTZ to CCRT improved survival outcomes compared to CCRT alone in LANPC patients who underwent IC [14]. Additionally, several studies have highlighted a correlation between response to IC and patient survival. Patients achieving complete response (CR) or partial response (PR) to IC exhibited significantly better survival outcomes than those with stable disease (SD) or progressive disease (PD) [15, 16]. Nonetheless, whether the efficacy of NTZ varies among patients with different response to IC remains unclear. Thus, this study aimed to explore the influence of IC response on survival outcomes in LANPC patients and further evaluate the efficacy of NTZ in patients with different IC response categories.

Materials and methods

Patients

Patients diagnosed with NPC between December 2012 and October 2021 were retrospectively included in this study. Inclusion criteria were as follows: (1) pathologically confirmed NPC with aged ≥ 18 years; (2) stage III-IVA disease according to the 8th edition of the American Joint Committee on Cancer staging system; (3) Eastern Cooperative Oncology Group performance status of 0 or 1; (4) receipt of two or three cycles of IC followed by CCRT with or without concurrent NTZ; (5) EGFR-positive disease, (6) available imaging evaluation after IC; (7) data on smoking and alcohol history available; (8) adequate hematologic, liver, and renal function. Patients with a history of prior malignancy or concurrent malignant diseases were excluded. Additionally, patients who did not complete radiotherapy were also excluded. This study received approval from the Ethics Committee of the First Affiliated Hospital of Xiamen University (approval number: 3502Z20244ZD1001), and informed consent was obtained from all patients.

Measures

The following patient characteristics were included in the analysis: gender, age, smoking history, alcohol history, histology, tumor (T) stage, nodal (N) stage, clinical stage, and use of NTZ. Former and current smokers were defined as individuals who had smoked within the past year or had quit smoking more than one year ago.

Induction chemotherapy

IC regimens consisted of two or three cycles of TPF (docetaxel 75 mg/m2 or paclitaxel 135 mg/m2 or nab-paclitaxel 260 mg/m2 on day 1, cisplatin 25 mg/m2 on days 1–3, and 5-FU 600–750 mg/m2 per day as a continuous 120 h infusion or S1 capsules 40 mg/m2 bid on day 1–14), TP (docetaxel 75 mg/m2 or paclitaxel 135 mg/m2 or nab-paclitaxel 260 mg/m2 on day 1, cisplatin 25 mg/m2 on days 1–3), or GP regimens (gemcitabine 1000 mg/m2 on days 1 and 8, cisplatin 25 mg/m2 on days 1–3). The choice of IC regimens was primarily based on the preferences of the treating physicians.

Assessment of induction chemotherapy response

Response after IC was assessed using the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1. All patients underwent nasopharyngeal and neck magnetic resonance imaging (MRI) evaluation one week after completing the last cycle of IC. Changes in imaging of primary nasopharyngeal tumors and metastatic cervical lymph nodes before and after IC were evaluated for CR, PR, SD, and PD.

Concurrent chemoradiotherapy and nimotuzumab treatment

All patients underwent intensity-modulated radiotherapy (IMRT). Primary nasopharyngeal tumor, cervical metastatic lymph nodes, high-risk clinical target volume, and low-risk clinical target volume received doses of 70 Gy (Gy), 66–70 Gy, 62 Gy, and 56 Gy in 33 fractions, respectively. Each patient received two cycles of cisplatin-based (80 mg/m2 every 3 weeks) or lobaplatin-based (30 mg/m2 every 3 weeks) concurrent chemotherapy. NTZ (200 mg, weekly for 7 courses) was administered during CCRT. The decision-making of the concurrent chemotherapy regimens was mainly according to physician-specific preference. NTZ has been approved for use in patients with LANPC in China. In clinical practice, NTZ was routinely recommended during CCRT for these patients, although its administration depended largely on patient preference.

Follow-up

Follow-up visits occurred every three months during the first two years, every 6 months from the 3rd to 5th year post-treatment, and annually thereafter. Physical examination, nasopharynx and neck lymph node examinations, and auxiliary examinations including nasopharynx and neck MRI, endoscopy, chest CT, abdominal ultrasound, and bone emission computerized tomography were performed. PET-CT scans were conducted as necessary. Biopsies or needle biopsies were performed if disease recurrence was suspected. The primary endpoints of this study were disease-free survival (DFS) and overall survival (OS). DFS was calculated from the time of NPC diagnosis to disease recurrence or death from any cause. OS was defined as the time of NPC diagnosis until death from any cause.

Statistical analysis

The chi-square test or chi-square test for trend (Cochran-Armitage test) was used to analyze the correlation between patient characteristics and IC response. Propensity score matching (PSM) was employed to reduce bias. Kaplan-Meier survival analysis with a log-rank test was used to depict and compare survival curves. Variables with P values < 0.10 in the univariate Cox regression model were included in the multivariate Cox proportional analysis to identify prognostic factors related to survival outcomes. Sensitivity analyses were performed to evaluate the impact of NTZ on survival outcomes after stratifying by IC response. All statistical analyses were conducted using SPSS version 26.0 (SPSS Inc., Chicago, IL, USA), with P values < 0.05 considered statistically significant.

Results

Patients characteristics

A total of 383 patients meeting the inclusion criteria were included (Table 1). Among them, 216 (56.4%) received NTZ, while 167 (43.6%) did not. The median age of patients was 48.0 years (range, 18–70 years). Of these, 278 (72.6%) were male, 338 (88.3%) had WHO III type classification, and 257 (67.1%) were at stage T3-4 disease. There were 83 (21.7%), 172 (44.9%), and 128 (33.4%) patients at N1, N2, and N3 stages, respectively.

Table 1 Patient baseline characteristics between those with CR/PR and SD/PD after induction chemotherapy
Full size table

Regarding primary nasopharyngeal tumors, the responses to IC were as follows: 79 (20.6%) for CR, 243 (63.4%) for PR, 58 (15.1%) for SD, and 3 (0.8%) for PD. For cervical lymph nodes, there were 52 (13.6%), 234 (61.1%), 93 (24.3%), and 4 (1.0%) patients with CR, PR, SD, and PD to IC, respectively. Overall, 269 (70.2%) patients achieved CR/PR, while 114 (29.8%) had SD/PD. There were no significant differences observed in terms of age (P = 0.478), gender (P = 0.188), smoking history (P = 0.430), alcohol consumption history (P = 0.678), histology (P = 0.891), T stage (P = 0.522), N stage (P = 0.059), and clinical stage (P = 0.207) between patients with CR/PR and SD/PD after IC. Notably, patients with SD/PD after IC were less likely to receive NTZ compared to those with CR/PR (46.5% vs. 60.6%, P = 0.011). In addition, patients who received gemcitabine-based IC showed a better response to IC compared to those treated with taxane-based IC (P < 0.001).

Survival analysis

The median follow-up time was 41.2 months (range, 5-127 months), with 3-year DFS and OS rates of 77.3% and 88.5%, respectively. The 3-year DFS for patients with CR/PR and SD/PR was 80.3% and 70.6%, respectively (P = 0.031) (Fig. 1A), while the 3-year OS rates were 90.9% and 83.2%, respectively (P = 0.038) (Fig. 1B). Additionally, the 3-year DFS for patients with and without NTZ were 81.5% and 71.5%, respectively (P = 0.016) (Fig. 2A), and the 3-year OS rates were 91.4% and 85.6%, respectively (P = 0.070) (Fig. 2B).

Fig. 1
figure 1

Comparison of disease-free survival (A) and overall survival (B) between those with CR/PR and SD/PD after induction chemotherapy

Full size image
Fig. 2
figure 2

Comparison of disease-free survival (A) and overall survival (B) between those with and without nimotuzumab during concurrent chemoradiotherapy

Full size image

Prognostic analysis

Cox regression analysis was performed to investigate prognostic factors influencing DFS and OS (Table 2). Factors with P-values less than 0.10 in the univariate Cox regression analysis for DFS included age, T stage, N stage, response to IC, and receipt of NTZ. These factors were incorporated into the multivariate Cox regression model, revealing the T stage, N stage, and response to IC as independent prognostic factors affecting survival. Patients with SD/PD after IC had significantly worse DFS compared to those with CR/PR (hazard ratio [HR] 1.610, 95% confidence interval [CI] 1.041–2.490, P = 0.032).

Table 2 Univariate and multivariate Cox regression analysis on the prognostic factors associated with disease-free survival and overall survival
Full size table

For OS, factors with P-values less than 0.10 in the univariate Cox regression analysis included age, T stage, response to IC, and receipt of NTZ. These factors were included in the multivariate Cox regression model, indicating response to IC as an independent prognostic factor affecting OS. Patients with SD/PD after IC had significantly worse OS compared to those with CR/PR (HR 1.741, 95% CI 1.039–2.917, P = 0.035).

Sensitivity analyses

Sensitivity analyses were conducted to evaluate the impact of NTZ on survival outcomes after stratifying by treatment response to IC. Among patients achieving CR/PR, NTZ treatment was identified as an independent prognostic factor influencing both DFS and OS (Table 3). Patients receiving NTZ exhibited better DFS (HR 0.459, 95% CI 0.263–0.799, P = 0.006) and OS (HR 0.429, 95% CI 0.194–0.949, P = 0.037). Furthermore, 3-year DFS and OS rates were significantly higher for patients receiving NTZ compared to those not receiving it (DFS: 85.4% vs. 72.1%, P = 0.005; OS: 94.4% vs. 87.1%, P = 0.020) (Fig. 3A and B). T stage, N stage, and age were also identified as independent prognostic factors affecting survival. A 1:1 PSM was employed to reduce potential bias, including variables such as age, gender, smoking history, alcohol history, histology, T stage, N stage, and IC regimens. A total of 80 pairs of patients were fully matched (Table 4). Results further indicated that patients receiving NTZ treatment exhibited better DFS (P = 0.036) (Fig. 3C) and OS (P = 0.015) (Fig. 3D).

Table 3 Univariate and multivariate Cox regression analysis on the prognostic factors associated with disease free survival and overall survival in patients with CR/PR to induction chemotherapy
Full size table
Fig. 3
figure 3

Comparison of disease-free survival (A, before propensity score matching; C, after propensity score matching) and overall survival (C, before propensity score matching; D, after propensity score matching) between those with and without nimotuzumab during concurrent chemoradiotherapy in the CR/PR groups before and after propensity score matching

Full size image
Table 4 Patient baseline characteristics between those with and without nimotuzumab during concurrent chemoradiotherapy in patients with CR/PR to induction chemotherapy after propensity score matching
Full size table

For patients with SD/PD after IC, no significant correlation was found between those with and without NTZ regarding DFS (3-year DFS: 69.7% vs. 70.6%) (HR 1.015, 95% CI 0.520–1.983, P = 0.964) (Fig. 4A) and OS (3-year OS: 83.1% vs. 83.1%) (HR 1.098, 95% CI 0.491–2.457, P = 0.820) (Fig. 4B).

Fig. 4
figure 4

Comparison of disease-free survival (A) and overall survival (B) between those with and without nimotuzumab during concurrent chemoradiotherapy in the SD/PD groups before propensity score matching

Full size image

Discussion

In this study, we investigated the impact of response to IC on survival outcomes in patients with LANPC. Furthermore, we examined whether the addition of NTZ during CCRT could enhance survival benefits for patients exhibiting different responses to IC. Our findings indicated that patients achieving CR/PR after IC experienced significantly better DFS and OS compared to those with SD/PD after IC. Moreover, our findings showed that the addition of NTZ during CCRT conferred survival benefits exclusively to patients who achieved CR/PR, but not to those with SD/PD.

IC represents a standard treatment for LANPC [4]. Although NPC is typically responsive to chemotherapy, not all patients exhibit sensitivity to these treatments. In our study, 70.2% of patients achieved CR/PR after IC, while 29.8% showed SD/PD. Prospective study using the TPF regimen reported a CR/PR rate of 90.9%, with 9.1% exhibiting SD [15]. Another prospective study using the GP regimen found a CR/PR rate of 94.5% after IC, with SD and PD rates of 4.2% and 1.3%, respectively [16]. However, in real-world studies like Liu et al., observed a CR/PR rate of 76.9% after IC, with a SD/PD rate of 23.1% [24]. Additionally, Jiang et al. reported CR/PR and SD/PD rates of 68.3% and 31.7% following IC, respectively [19], which was consistent with our findings. Furthermore, our study indicated that IC regimens predominantly including gemcitabine resulted in significantly higher tumor response rates compared to those incorporating taxanes. Therefore, variations in patient demographics and treatment protocols may contribute to differing outcomes following IC.

The secondary analyses from the aforementioned two prospective studies consistently demonstrated a strong correlation between IC response and patient survival in LANPC. Patients achieving CR/PR after IC exhibited significantly better outcomes compared to those with SD/PD after IC [15, 16]. This correlation was consistently observed across retrospective studies conducted in both endemic and non-endemic areas [17,18,19,20]. Our study underscores the critical role of IC response in determining survival outcomes of LANPC patients. Specifically, patients achieving CR/PR showed significantly improved DFS and OS, aligning with previous studies that highlight the crucial role of favorable response to IC in improving LANPC prognosis. The primary objective of IC is to reduce tumor size, minimize distant metastasis risk, and potentially improve locoregional control efficacy. Patients demonstrating favorable IC responses generally experience better long-term outcomes. Therefore, IC response could serve as a potential surrogate endpoint for LANPC. Conversely, patients with SD/PD following IC require aggressive treatment strategies, highlighting the importance of closely monitoring treatment response during IC to facilitate timely adjustments in treatment plans.

The mechanism of action of NTZ involves EGFR pathway blockade, which is pivotal for cancer cell growth and survival [6, 21]. By inhibiting EGFR activation, NTZ enhances the cytotoxic effects of chemotherapy and radiotherapy [22, 23]. This dual-targeting approach likely contributes to the observed survival benefits in patients receiving NTZ. Notably, NTZ has shown promising efficacy across various cancers, including LANPC, head and neck cancer, cervical, and esophageal cancers, with a favorable safety profile [6,7,8,9]. Our study showed that adding NTZ to CCRT improves survival outcomes in chemotherapy-sensitive patients who respond favorably to IC, suggesting its potential as a valuable adjunctive therapy. This finding suggests that NTZ may enhance treatment efficacy by inhibiting EGFR. Importantly, tumor response to IC may inform survival outcomes and guide personalized treatment strategies, particularly for patients achieving CR/PR.

For patients showing suboptimal response to IC, several studies have explored various strategies, including increasing cisplatin dosage during CCRT, employing dual-drug combination chemotherapy during CCRT, and escalating cycles of IC. However, none have shown a clear survival advantage [24,25,26]. Therefore, further investigation into methods to enhance efficacy is warranted for these patients. In our study, patients with SD/PD after IC were less likely to receive NTZ compared to those with CR/PR (46.5% vs. 60.6%, P = 0.011). The underlying reasons for this phenomenon remain unclear. Factors such as physician preferences, patient preferences, or treatment accessibility may influence treatment decisions. Similarly, patients experiencing SD/PD after IC often exhibit chemotherapy insensitivity, representing a subgroup with the highest rate of disease progression. Identifying more suitable treatment options for these patients is imperative. Thus, incorporating an EGFR inhibitor during CCRT could present a viable therapeutic avenue. A study by Niu et al. found that adding NTZ during CCRT in IC-resistant LANPC patients resulted in 3-year DFS and OS of 79.3% and 94.0%, respectively, with predominantly mild hematologic acute toxicities and rare severe toxicities [27]. However, the absence of a control group in this study precludes definitive determination of the exact efficacy of NTZ in patients with suboptimal response to IC. Contrarily, our study found no survival improvement with NTZ addition during CCRT in patients assessed as SD/PD after IC. Notably, patients exhibiting unsatisfactory tumor responses after IC demonstrated poorer prognoses and inferior response to NTZ compared to those with satisfactory tumor responses. Hence, clinical attention should be focused on patients experiencing SD/PD after IC.

NPC patients often harbor several drug-resistant genes [28, 29]. While several preclinical studies have attempted to reverse chemoresistance [30, 31], no formal drugs currently enhance cisplatin sensitivity. Immune checkpoint inhibitors may offer additional survival benefits to IC-insensitive patients. A recent study found that incorporating sintilimab (a PD-1 inhibitor) during IC, CCRT, and adjuvant therapy significantly improved event-free survival compared to those treated with IC and CCRT alone in LANPC [32]. However, studies involving avelumab (a PD-L1 inhibitor) or pembrolizumab (a PD-1 inhibitor) alongside standard-of-care CCRT did not enhance survival outcomes relative to CCRT alone in head and neck cancer patients [33, 34]. Therefore, further exploration is warranted to optimize the integration of immune checkpoint inhibitors into the treatment of LANPC at the optimal time. Several studies have found that adding immune checkpoint inhibitors during the IC phase could increase the CR rate to approximately 40–50% [35,36,37,38], significantly higher than the rates achieved with TPF or GP regimens (typically around 10%) [15, 16]. However, identifying the optimal patients who may benefit from immune checkpoint inhibitors during IC necessitates further investigation. Radiomics has been employed in several studies to predict IC response [39,40,41], though additional research is required to validate these findings. Moreover, carbon-ion radiotherapy may represent a potential treatment avenue for IC-insensitive patients [42, 43].

It is crucial to acknowledge the limitations of our study. Firstly, its retrospective nature renders it susceptible to inherent biases. Confirming our findings and establishing the causal relationship between treatment response and survival outcomes necessitate randomized controlled trials. Secondly, the small sample size in our study may limit the generalizability of our findings. Future studies with larger cohorts are necessary to validate our findings. Thirdly, we did not include EBV-DNA levels for analysis in this study because EBV-DNA testing was not conducted at our institution before 2017. However, in our previous studies, we have found that EBV-DNA levels before treatmet, residual EBV-DNA levels after IC, and residual EBV-DNA levels after IMRT have a significant impact on the survival outcomes of NPC patient [44, 45].

Conclusions

In conclusion, our study highlights the significance of treatment response to IC in LANPC and underscores the potential benefits of NTZ during CCRT for improving survival outcomes in patients achieving CR/PR to IC. Tailoring individualized treatment strategies for patients with SD/PD to IC warrants further exploration. These findings emphasize the significance of closely monitoring treatment response and adjusting treatment strategies accordingly.