INCENP and CDCA8 predict neoadjuvant chemotherapy response and outcomes in esophageal squamous cell carcinoma

Wang, Xiangyu; Wang, Ting; Wang, Keke; Zhang, Chengjuan; Li, Zhibo; Liu, Feifei; Tian, Xueli; Shi, Xiaodan; Zhang, Zihan; Wang, Rui; Jia, Ludan; Laster, Kyle Vaughn; Xia, Qingxin; Zhao, Simin; Dong, Zigang

doi:10.1038/s41467-026-68371-x

Download PDF

Article
Open access
Published: 21 January 2026

INCENP and CDCA8 predict neoadjuvant chemotherapy response and outcomes in esophageal squamous cell carcinoma

Xiangyu Wang^1,2^na1,
Ting Wang^1,2^na1,
Keke Wang²,
Chengjuan Zhang³,
Zhibo Li^1,2,
Feifei Liu^1,2,
Xueli Tian^2,4,
Xiaodan Shi^1,2,
Zihan Zhang^1,2,
Rui Wang^1,2,
Ludan Jia^1,2,
Kyle Vaughn Laster ORCID: orcid.org/0000-0002-8469-0015²,
Qingxin Xia³,
Simin Zhao^3,5,6 &
…
Zigang Dong ORCID: orcid.org/0000-0002-4174-4028^1,2,3,5,6

Nature Communications volume 17, Article number: 1672 (2026) Cite this article

3594 Accesses
Metrics details

Subjects

Abstract

Neoadjuvant chemotherapy (NACT), a key strategy for various cancers, markedly improves patient prognosis and 5-year survival rates. However, numerous patients develop resistance to NACT and thus fail to benefit from it. Therefore, identifying reliable biomarkers to predict patient responsiveness to NACT remains a critical challenge. Here, we demonstrate that elevated expression of INCENP and CDCA8 contributes to poor NACT responsiveness across multiple cancers. Mechanistically, the 5′UTR (GGACT at position 113) of INCENP and the 3′UTR (GGACT at position 1041) of CDCA8 undergo m⁶A methylation and are recognized by YTHDF3, which facilitates their translation through interaction with eIF3A, ultimately driving poor response to NACT. Moreover, inhibition of INCENP and CDCA8 enhances NACT sensitivity by promoting multipolar spindle formation. Collectively, our findings establish that INCENP and CDCA8 serve as crucial biomarkers for predicting NACT responsiveness and as potential therapeutic targets for combination therapy with NACT to improve patient survival.

Dissecting the single-cell transcriptome network in patients with esophageal squamous cell carcinoma receiving operative paclitaxel plus platinum chemotherapy

Article Open access 26 October 2021

Neoadjuvant chemotherapy reduces the levels of HMGB1 and E-cadherin in patients with breast cancer

Article Open access 08 September 2023

The impact of neoadjuvant chemotherapy on the tumor microenvironment in advanced high-grade serous carcinoma

Article Open access 30 July 2022

Introduction

Neoadjuvant chemotherapy (NACT) has become a standard treatment for patients with various cancers undergoing surgery, particularly in lung cancer, breast cancer, ovarian cancer, and esophageal cancer^1,2,3,4. It aims to reduce tumor size, improve surgical resectability, and lower the risk of postoperative recurrence. Among the commonly used NACT regimens, paclitaxel plus cisplatin is a primary first-line treatment for several malignancies^5,6,7. Although this regimen is considered highly cytotoxic to tumors, its response rate in esophageal cancer remains limited at approximately 50–60%^8,9. Notably, a subset of patients exhibits resistance to NACT, leading to tumor progression and significantly compromising both survival outcomes and quality of life^10,11. Therefore, identifying reliable predictive biomarkers for NACT sensitivity is essential for guiding personalized treatment.

The mechanisms underlying resistance to NACT are complex, including tumor microenvironment influences and molecular heterogeneity of cancers^12,13. Among these, dysregulation of the cell cycle has emerged as a key contributor. Mitosis is a fundamental biological process that regulates cell division and serves as a crucial therapeutic target in overcoming paclitaxel resistance^14,15. Overexpression of mitotic regulators, such as cyclins and checkpoint proteins, can effectively halt or repair damage in mitosis, thereby diminishing the efficacy of paclitaxel^16,17. Conversely, targeting these regulators, which leads to chromosomal instability or maintains multipolar spindles throughout mitosis, can improve the clinical utility of paclitaxel^18,19. Additionally, cisplatin resistance has been linked to dysregulation of cell cycle-associated proteins^20,21. In this study, we focus on the chromosomal passenger complex (CPC), a key mitotic regulator composed of Aurora B kinase, inner centromere protein (INCENP), survivin, and cell division cycle-associated 8 (CDCA8). The CPC plays a central role in mitosis by orchestrating chromatin-induced microtubule stabilization, chromosome bi-orientation, and spindle assembly checkpoint regulation^22,23,24. However, the role of CPC in regulating NACT resistance remains unexplored.

Emerging evidence suggests that epigenetic modifications, particularly abnormal N⁶-methyladenosine (m⁶A) methylation, contribute to malignancy progression²⁵. m⁶A modifications regulate gene expression through RNA methyltransferases (writers), demethylases (erasers), and m⁶A-binding proteins (readers), all of which are frequently dysregulated in various tumor types^26,27. Recent studies indicate that m⁶A modification plays a key role in drug resistance via multiple molecular pathways^28,29 ; however, the role of m⁶A in mitotic regulation and its impact on NACT response remain poorly understood. Therefore, further investigation is required to elucidate the role of m⁶A methylation in NACT resistance.

In this study, we identify INCENP and CDCA8 as key predictors of poor response to NACT across multiple cancer types. We demonstrate that m⁶A methylation of INCENP and CDCA8 mRNAs is catalyzed by METTL3 and recognized by YTHDF3, which promotes their translation by interacting with eIF3A, thereby contributing to NACT resistance. Importantly, the efficacy of these chemotherapeutic agents is dependent on the m⁶A modification status of INCENP and CDCA8. Inhibition of INCENP and CDCA8 leads to the formation of a multipolar spindle, thereby enhancing sensitivity to NACT both in vitro and in vivo. These results provide a strong rationale for exploring combination therapies that target INCENP and CDCA8 to improve NACT responsiveness across multiple cancers.

Results

Upregulation of INCENP and CDCA8 contributes to poor NACT response

To identify potential drivers of NACT response, we performed transcriptomic profiling of clinical esophageal squamous cell carcinoma (ESCC) tissue samples from NACT recipients (Cohort 1, Supplementary Data 1) following the study design (Fig. 1a). Patients with evaluable responses were classified as responders (TRG0, 1) and non-responders (TRG2, 3) based on the National Comprehensive Cancer Network (NCCN) TRG criteria³⁰. To explore the mechanisms regulating NACT response, we began with gene ontology (GO) analysis, which revealed that differentially expressed genes (DEGs) converge on key regulators of cell division (Fig. 1b). Consistent with these findings, Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis revealed significant alterations in cell cycle pathways (Supplementary Fig. 1a), highlighting the crucial role of cell mitotic regulation in NACT response. Additionally, we observed that the N⁶-methyladenosine-containing RNA binding function ranked as the top-enriched molecular function (Fig. 1b), and m⁶A modification plays a critical role in RNA metabolism by regulating gene expression through modulation of mRNA stability, translation efficiency, and splicing³¹. More importantly, emerging evidence suggests that dynamic m⁶A modifications play a significant role in chemotherapy resistance^32,33. Therefore, we hypothesize that m⁶A modification modulates the progression of the mitotic pathway by regulating the RNA metabolism of mitosis-related genes, thereby driving a poor response to NACT. To address this, we analyzed the differential expression of 20 m⁶A regulators in Cohort 1 (Fig. 1c) and found a markedly elevated YTHDF3 expression in non-responder samples (Supplementary Fig. 1b). These findings suggest that YTHDF3 upregulation plays a significant role in the poor response to NACT in ESCC.

We further investigated whether YTHDF3, as an upstream m⁶A regulator, modulates the mitotic pathway and identified candidate mitosis-related genes that may contribute to poor NACT responsiveness. Therefore, we employed a functional proteomics analysis using YTHDF3 knockdown (KD) ESCC cells. Initially, we analyzed the variation in KEGG pathways, and surprisingly, the results indicated enrichment of numerous proteins involved in the cell cycle (Supplementary Fig. 1c). More importantly, GO enrichment analysis highlighted the chromosomal passenger complex in the cellular component section as the top altered pathway in shYTHDF3 ESCC cells (Fig. 1d). Moreover, Gene Set Enrichment Analysis (GSEA) revealed a negative association between genes involved in mitotic nuclear division (Fig. 1e, Supplementary Fig. 1d) in YTHDF3 KD cells. Additionally, the core components of the CPC, including INCENP, CDCA8, and AURKB, were significantly downregulated in shYTHDF3 cells. Notably, expression levels of INCENP and CDCA8 showed more pronounced changes in both shYTHDF3 and YTHDF3 overexpressing (OE) ESCC cell lines (Supplementary Fig. 1e–h); therefore, we focused on INCENP and CDCA8 as potential targets for further investigation. We subsequently verified that the expression of INCENP, CDCA8 and YTHDF3 proteins was upregulated in non-responders of cohort 1 clinical samples by immunohistochemical staining (IHC) (Fig. 1f, g), and that the expression of INCENP and CDCA8 was positively correlated with YTHDF3 levels (Fig. 1h). Collectively, by integrating transcriptomic profiling of Cohort 1, proteomics analysis of shYTHDF3 cells, and pathological examination of protein expression, we demonstrate that INCENP and CDCA8, as cell mitotic regulators modulated by YTHDF3, may contribute to poor NACT response in ESCC patients.

To further confirm that elevated expression of INCENP and CDCA8 is associated with poor response to NACT in ESCC patients, we analyzed another clinical cohort (Cohort 2, Supplementary Data 2). Radiological images indicated that tumor size was significantly reduced in responders, and IHC results revealed markedly higher INCENP and CDCA8 expression in tissue samples from non-responders (Fig. 1i, j) and in cancer tissues from ESCC tissue arrays (Supplementary Fig. 1i, j, Supplementary Data 3). These findings suggest that the upregulation of INCENP and CDCA8 contributes to poor response to NACT and is associated with ESCC progression. Furthermore, elevated expression levels of INCENP and CDCA8 were associated with poor clinical outcomes, including disease progression (Fig. 1k) and shorter survival times in ESCC patients receiving NACT (Supplementary Fig. 1k). Collectively, INCENP and CDCA8 may serve as potential drivers of NACT non-responsiveness in ESCC.

Knockdown of INCENP and CDCA8 potentiates the sensitivity of ESCC to NACT

Given the elevated expression of INCENP and CDCA8 in non-responders, we explored the impact of these proteins on the chemosensitivity of ESCC cells to paclitaxel and cisplatin. To investigate the correlation between basal expression levels of INCENP and CDCA8 and the sensitivity of ESCC cell lines to paclitaxel and cisplatin, we treated a panel of ESCC cell lines with these drugs. These findings indicate that elevated basal expression levels of INCENP and CDCA8 are associated with increased IC50 values, suggesting a negative correlation between their expression and the sensitivity to paclitaxel and cisplatin across ESCC cell lines. (Fig. 2a, b and Supplementary Fig. 2a).

To further validate the sensitizing effect of INCENP and CDCA8 KD on paclitaxel and cisplatin responses in ESCC cells, an MTT assay was utilized, and the results showed that INCENP and CDCA8 KD decreased the IC50 values of paclitaxel and cisplatin compared with mock controls (Fig. 2c, d and Supplementary Fig. 2b, c). Consistently, plate colony formation assays demonstrated that INCENP and CDCA8 KD increased the sensitivity of ESCC cells to paclitaxel and cisplatin compared with the mock control (Fig. 2e, f and Supplementary Fig. 2d, e). Furthermore, overexpression (OE) of INCENP and CDCA8 increased the IC50 values of paclitaxel and cisplatin in KYSE450 ESCC cells compared with the vector control (Fig. 2g and Supplementary Fig. 2f), with similar results observed in the plate colony formation assay (Fig. 2h). To further validate these findings, in vivo assays were conducted to examine the impact of targeting INCENP and CDCA8 on the therapeutic efficacy of paclitaxel and cisplatin. Consistent with our in vitro findings, treatment of shINCENP and shCDCA8 ESCC cells with paclitaxel or cisplatin resulted in a reduced ESCC tumor burden compared with paclitaxel- or cisplatin-treated control cells. (Fig. 2i–k). These findings support the idea that targeting INCENP and CDCA8 enhances response to NACT in ESCC.

YTHDF3 drives mitosis and NACT resistance via INCENP and CDCA8

To further investigate whether YTHDF3 acts as an upstream regulator that modulates the cell mitotic pathway and suppresses NACT responsiveness through the regulation of INCENP and CDCA8 expression, we examined cell cycle and mitotic alterations in shYTHDF3 ESCC cells. The results showed that YTHDF3 KD led to G2/M phase arrest (Fig. 3a) and decreased protein expression of G2 phase regulators (Supplementary Fig. 3a, b). Consistently, immunofluorescence (IF) results demonstrated that YTHDF3 KD induced a defective cell mitosis phenotype characterized by multipolar spindle formation in ESCC cells (Fig. 3b and Supplementary Fig. 3c, d).

To further investigate the function of YTHDF3, we used Ebselen as a selective inhibitor. Its target specificity was systematically validated by assessing the viability of YTHDF3 KD cells following Ebselen treatment (Supplementary Fig. 3e, f). Consistently, Ebselen treatment also induced G2/M phase arrest (Supplementary Fig. 3g–i) and promoted the formation of multipolar spindles in ESCC cells (Supplementary Fig. 3j, k). These findings further confirm that YTHDF3 plays a crucial role in regulating mitosis in ESCC cells.

Given the elevated expression of YTHDF3 in non-responders, we investigated its role in modulating chemosensitivity in ESCC cells. Treatment of a panel of ESCC cell lines with paclitaxel and cisplatin revealed that higher basal expression of YTHDF3 was associated with increased IC50 values, indicating reduced drug sensitivity (Supplementary Fig. 3l, m). Furthermore, YTHDF3 KD and Ebselen treatment enhanced the sensitivity of ESCC cells to paclitaxel and cisplatin compared with the control (Fig. 3c–f and Supplementary Fig. 3n, o). In contrast, YTHDF3 OE decreased the sensitivity of ESCC cells to paclitaxel and cisplatin compared with the vector control (Fig. 3g, h and Supplementary Fig. 3p). In vivo studies revealed that Ebselen sensitized ESCC cells to paclitaxel (Fig. 3i–k) and cisplatin treatment (Fig. 3l–n), as evidenced by reduced tumor size and weight. Importantly, no significant differences in total body weight were associated with any treatment regimens (Supplementary Fig. 3q, r). These findings support the idea that inhibiting YTHDF3 potentiates NACT responsiveness both in vitro and in vivo.

To determine whether the effect of YTHDF3 on NACT response is functionally dependent on INCENP and CDCA8, we examined the impact of their overexpression in YTHDF3 KD cells. We found that INCENP and CDCA8 OE reversed the enhanced sensitivity to paclitaxel and cisplatin caused by YTHDF3 KD, resulting in decreased drug sensitivity (Fig. 3o, p and Supplementary Fig. 3s, t). These findings support the notion that both INCENP and CDCA8 are essential for proper mitotic progression and functionally act as downstream effectors of YTHDF3. Collectively, our results suggest that the effect of YTHDF3 on NACT responsiveness depends on INCENP and CDCA8.

INCENP and CDCA8 mRNAs are m⁶A-methylated and recognized by YTHDF3 to modulate NACT response

To elucidate whether YTHDF3 regulates INCENP and CDCA8 expression in an m⁶A-dependent manner, we assessed the requirement of its m⁶A-binding ability in this regulatory process. YTHDF3 contains two hydrophobic residues W438 and W492, which are integral to the specific recognition of m⁶A^31,34. To examine this, we constructed a YTHDF3 mutant (YTHDF3-mut) plasmid (Supplementary Fig. 4a) and transfected both WT and YTHDF3-mut plasmids into YTHDF3 KD cells. As expected, the decreased expression of CDCA8 and INCENP in YTHDF3 KD cells was rescued by ectopic expression of WT YTHDF3, but not by the YTHDF3-mut (Fig. 4a, b and Supplementary Fig. 4b, c). Furthermore, G2/M phase arrest (Fig. 4c) and reduced expression of G2 phase regulator (Fig. 4d and Supplementary Fig. 4d) observed in YTHDF3 KD cells were reversed upon WT YTHDF3 transfection. Moreover, the formation of multipolar spindles in mitotic YTHDF3 KD ESCC cells was rescued by WT YTHDF3 overexpression, but not by the mutant YTHDF3 (Fig. 4e, f). Collectively, these findings suggest that YTHDF3 regulates the expression of INCENP and CDCA8 in an m⁶A-dependent manner and plays a pivotal role in cell mitosis.

Fig. 4: INCENP and CDCA8 are N6-methylated and recognized by YTHDF3 to modulate NACT response. — **Fig. 4: *INCENP* and *CDCA8* are N⁶-methylated and recognized by YTHDF3 to modulate NACT response.**

To further investigate how CDCA8 and INCENP transcripts undergo m⁶A modification, we analyzed the GEPIA database and observed a positive correlation between the expression of the RNA methyltransferase METTL3 and the levels of both INCENP and CDCA8 (Supplementary Fig. 4e), and METTL3 expression was examined to identify its role in the m⁶A modification of these transcripts. Subsequently, we generated stable cell lines with low levels of m⁶A decoration by METTL3 KD (Supplementary Fig. 4f). We observed reduced protein levels of CDCA8 and INCENP in both METTL3 KD cells (Fig. 4g and Supplementary Fig. 4g, h) and ESCC patient-derived xenograft organoids (PDXO) (Fig. 4h–k and Supplementary Fig. 4i). Moreover, multipolar spindle formation was observed in mitotic METTL3 KD cells (Fig. 4l, m and Supplementary Fig. 4j). These data collectively demonstrate that YTHDF3 regulates METTL3-mediated N⁶-methylation of CDCA8 and INCENP transcripts in an m⁶A-dependent manner. More importantly, METTL3 expression was significantly higher in non-responder tissues compared to responder tissues (Fig. 4n, o). Further analysis revealed that METTL3 expression was positively correlated with INCENP and CDCA8 expression (Fig. 4p), thereby identifying N⁶-methylation of INCENP and CDCA8 transcripts as a key factor associated with poor NACT responsiveness in ESCC cells.

YTHDF3 promotes the translation of INCENP and CDCA8 mRNAs

Given that the expression of INCENP and CDCA8 proteins was decreased in shYTHDF3 ESCC cells (Supplementary Fig. 1e, f) and PDXO (Fig. 5a, b and Supplementary Fig. 5a, b), we sought to elucidate the molecular mechanisms by which YTHDF3 regulates INCENP and CDCA8. RNA immunoprecipitation sequencing (RIP-seq) analysis revealed direct binding of YTHDF3 protein to both INCENP and CDCA8 transcripts (Fig. 5c, d), which was subsequently confirmed by RIP-qPCR assays (Fig. 5e). Given that YTHDF3 facilitates translation and decay of N⁶-methyladenosine-modified RNA³¹, we next examined whether YTHDF3 knockdown affected target mRNA abundance. qRT-PCR results showed that the mRNA levels of INCENP and CDCA8 were unaffected by YTHDF3 KD (Supplementary Fig. 5c), suggesting that the observed protein downregulation likely occurs at the translational level. Indeed, the nascently synthesized proteins on ESCC cell lines overexpressing Flag-YTHDF3 were increased compared to the vector control group as determined by a protein synthesis assay (Fig. 5f, g) and the Surface Sensing of Translation (SUnSET) assay (Supplementary Fig. 5d). To further investigate, we constructed pmirGLO-CDCA8 and pmirGLO-INCENP luciferase reporters. The dual-luciferase assay showed that the translation efficiency of CDCA8 and INCENP was significantly lower in YTHDF3 KD ESCC cells compared with mock cells (Fig. 5h, i). Collectively, these findings indicate that YTHDF3 regulates the translation of CDCA8 and INCENP mRNAs.

To elucidate the underlying mechanism by which YTHDF3 regulates the translation of INCENP and CDCA8, we performed co-immunoprecipitation (co-IP) to identify proteins associated with translation regulation, including eEF1A2, RPS3A, eIF4A1, and eIF3A (Supplementary Fig. 5e). Our results demonstrated that YTHDF3 interacts with eIF3A, a component of the eukaryotic translation initiation factor (eIF) complex, as confirmed by both co-IP and IF assays (Fig. 5j, k and Supplementary Fig. 5f, g). To further confirm the direct interaction between YTHDF3 and eIF3A, we performed a proximity ligation assay (PLA), which illustrated the distribution of in situ PLA spots of YTHDF3-eIF3A in the cell cytoplasm (Fig. 5l, m). To investigate whether YTHDF3 regulates the translation of INCENP and CDCA8 through eIF3A, we performed eIF3A RIP-qPCR and found that both YTHDF3 KD and Ebselen treatment significantly reduced the binding of eIF3A to CDCA8 and INCENP mRNAs (Fig. 5n, o and Supplementary Fig. 5h). Additionally, ESCC cells treated with Ebselen showed reduced binding of YTHDF3 to CDCA8 and INCENP mRNAs (Supplementary Fig. 5i), leading to decreased protein expression levels of INCENP and CDCA8 (Supplementary Fig. 5j, k). Collectively, these findings highlight that YTHDF3 enhances the translation of CDCA8 and INCENP by recruiting eIF3A to its target mRNAs.

To further investigate whether YTHDF3 regulates CDCA8 and INCENP translation in an m⁶A-dependent manner. RIP-qPCR was performed, and the results showed that YTHDF3 and eIF3A bind more strongly to CDCA8 and INCENP mRNAs in ESCC cells overexpressing WT YTHDF3, but not in cells expressing mutant YTHDF3 (Supplementary Fig. 5l, m). Additionally, dual luciferase assays demonstrated a significant increase in the translation efficiency of CDCA8 and INCENP mRNAs in YTHDF3 KD ESCC cells ectopically expressing WT YTHDF3, compared with those expressing mutant YTHDF3 (Supplementary Fig. 5n). Moreover, RIP-qPCR assays showed that the binding of YTHDF3 and eIF3A to m⁶A-modified CDCA8 and INCENP mRNAs was significantly reduced in METTL3 KD cells (Fig. 5p, q). These findings collectively suggest that YTHDF3 regulates the translation of CDCA8 and INCENP in an m⁶A-dependent manner.

INCENP and CDCA8 contribute to poor NACT response in an m⁶A-dependent manner

Based on our findings that m⁶A modification of INCENP and CDCA8 mRNAs is indispensable for YTHDF3 recognition and regulation, it is essential to explore the m⁶A modification sites of INCENP and CDCA8 mRNAs for the regulation of NACT response in ESCC cells. Very high-confidence m⁶A modification sites were predicted using the SRAMP web tool (www.cuilab.cn/sramp) (Fig. 6a, b). Methylated RNA immunoprecipitation followed by qPCR (meRIP-qPCR) results showed that, among the predicted m⁶A regions of INCENP, the 5′UTR m⁶A site (site 1: GGACT at position 113) exhibited the most pronounced decrease in methylation upon METTL3 KD in ESCC cells (Fig. 6c, Supplementary Fig. 6a, b). For CDCA8, the CDS region (site 2: TGACT at position 909) and the 3′UTR region (site 4: GGACT at position 1041) showed the greatest reduction in m⁶A levels following METTL3 KD (Fig. 6d, Supplementary Fig. 6c, d). Dual-luciferase reporter assays further demonstrated that mutation of the 5′UTR m⁶A site in INCENP, as well as mutations of the 3′UTR and dual-site regions in CDCA8, significantly reduced translation efficiency in ESCC cells, whereas mutation of the CDS m⁶A site alone in CDCA8 had no detectable effect (Supplementary Fig. 6e, f). Collectively, these data indicate that the 5′UTR (GGACT at position 113) of INCENP and the 3′UTR (GGACT at position 1041) of CDCA8 undergo METTL3-mediated m⁶A modification, which contributes to their translation regulation in ESCC.

Fig. 6: INCENP and CDCA8 drive poor NACT response in an m6A-dependent manner. — **Fig. 6: INCENP and CDCA8 drive poor NACT response in an m⁶A-dependent manner.**

To elucidate whether CDCA8 and INCENP contribute to the poor response to NACT through these m⁶A modification sites, we overexpressed either WT or double-mutant CDCA8 (mut-CDCA8) and 5’UTR-mutant INCENP (mut-INCENP) in YTHDF3 KD ESCC cell lines. We found that the frequency of multipolar spindle formation observed in YTHDF3 KD ESCC cells was decreased upon expression of WT INCENP or CDCA8 (Fig. 6e, f). To determine whether m⁶A modification of INCENP and CDCA8 plays a pivotal role in NACT responsiveness, we assessed the sensitivity of YTHDF3 KD ESCC cells to paclitaxel and cisplatin following transfection with either WT or m⁶A site–mutant INCENP and CDCA8 constructs. Our results showed that paclitaxel and cisplatin sensitivity was reduced in YTHDF3 KD cells transfected with WT INCENP or CDCA8 plasmids, but not in those transfected with the m⁶A site–mutant constructs (Fig. 6g, h and Supplementary Fig. 6g, h). These findings indicate that the ability of INCENP and CDCA8 to modulate the NACT response is dependent on their m⁶A methylation. Collectively, INCENP and CDCA8 transcripts are N⁶-methylated by METTL3 and subsequently recognized by YTHDF3, which promotes their translation via interaction with eIF3A, ultimately leading to paclitaxel and cisplatin resistance in ESCC (Fig. 6i).

INCENP and CDCA8 serve as predictors of poor NACT response across multiple cancer types

Given the contribution of INCENP, CDCA8, and YTHDF3 to poor NACT response, we retrospectively analyzed their expression levels in relation to clinical outcomes using a cohort of 65 pre-treatment biopsy samples from ESCC patients (Cohort 3, Supplementary Data 4). Based on annotated clinical and molecular attributes of individual patient data, we observed that non-responders were more likely to present with advanced TNM stages and medullary histological types (Fig. 7a and Supplementary Fig. 7a), suggesting a correlation between these clinical parameters and NACT response. IHC staining revealed that responders exhibited significantly lower expression levels of INCENP, CDCA8, and YTHDF3 compared with non-responders in biopsy tissues from ESCC (Fig. 7b, c). Given the widespread use of the paclitaxel plus cisplatin regimen as an NACT strategy across multiple cancer types^35,36,37. We extended our analysis to assess the protein expression levels of INCENP, CDCA8, and YTHDF3 in response to this regimen in breast and lung cancer patients from Cohort 3. IHC results revealed that patients who responded favorably to NACT exhibited significantly lower expression levels of INCENP, CDCA8, and YTHDF3 compared with non-responders (Fig. 7b, d, e). Furthermore, expression levels of INCENP and CDCA8 showed a positive correlation with YTHDF3 expression (Fig. 7f and Supplementary Fig. 7b, c). Collectively, these findings demonstrate that INCENP, CDCA8, and YTHDF3 are upregulated in multiple cancer types, supporting the pivotal role of the YTHDF3-m⁶A-CDCA8/INCENP axis in driving poor NACT responsiveness across multiple cancers.

To further investigate the prognostic relevance of INCENP, CDCA8, and YTHDF3 expression in the context of NACT, we performed survival analysis. Results revealed that responders exhibited significantly longer progression-free survival (PFS) and overall survival (OS) compared with non-responders (Supplementary Fig. 7d, e). Notably, among non-responders, high expression of INCENP, CDCA8, and YTHDF3 proteins was associated with significantly shorter PFS in ESCC patients (Fig. 7g–i), suggesting that elevated levels of these proteins are correlated with poor clinical outcomes following NACT.

Receiver operating characteristic (ROC) curves were generated for each biomarker across multiple cancer types to evaluate their predictive performance, reflecting the sensitivity and specificity of these predictors. The results revealed area under the curve (AUC) values of 0.785, 0.736, and 0.707 for INCENP, CDCA8, and YTHDF3, respectively. Importantly, a combined analysis of these biomarkers demonstrated superior predictive accuracy, with an AUC of 0.817 in ESCC (Fig. 7j). These findings indicate that a composite biomarker model provides more accurate prediction of poor NACT response in ESCC patients. Additionally, in both lung and breast cancers, YTHDF3, INCENP, and CDCA8 also exhibited significant potential as predictors of NACT response (Fig. 7k, l). Collectively, these results highlight the critical role of YTHDF3, INCENP, and CDCA8 as robust and broadly applicable predictors of NACT responsiveness across multiple cancer types, with enhanced accuracy achieved through combined assessment.

Discussion

To improve the prognosis of advanced cancer, NACT, followed by surgery, is a promising treatment strategy. Among various NACT regimens, paclitaxel plus cisplatin is routinely used as a first-line chemotherapeutic combination³⁸. While NACT has been shown to improve prognosis in responders, its overall efficacy is compromised by the poor NACT response observed in some patients. Non-responders not only endure unnecessary side effects but also lose valuable time that could be used for alternative treatments. Although recent studies have found that targeted molecular therapies complement chemotherapy, there are limited options available in clinical practice^39,40,41. Therefore, it is crucial to identify reliable biomarkers that can predict a patient’s response to NACT prior to treatment, ultimately improving clinical outcomes. Moreover, the development of an effective combination therapeutic approach to improve NACT responsiveness remains to be investigated. Our study identifies elevated expression of INCENP and CDCA8 across multiple cancer types as critical predictive biomarkers of poor response to NACT. Notably, we found the YTHDF3-m⁶A-INCENP/CDCA8 axis as a key regulator of NACT responsiveness, targeting INCENP and CDCA8, thus serving as a promising strategy to improve NACT efficacy across multiple cancer types (Supplementary Fig. 8).

Given that NACT responsiveness is influenced by multiple independent molecular pathways, it is critical to systematically investigate the mechanisms underlying poor NACT responses and assess their functional relevance in patient-derived samples. Here, transcriptomics profiling of clinical ESCC samples identified cell division processes as key determinants of NACT responsiveness. In addition to mitotic dysregulation, alterations in m⁶A RNA modifications have also been observed. m⁶A modification plays a critical role in RNA metabolism, regulating gene expression through modulation of mRNA stability, translation efficiency, and splicing³¹. More importantly, dynamic m⁶A modifications have been implicated in key oncogenic processes, including chemotherapy resistance^42,43,44,45. Therefore, a deeper understanding of how m⁶A modification modulates the mitotic pathway by regulating the RNA metabolism of mitosis-related genes and contributes to NACT resistance is urgently needed. Next, we focused on identifying differentially expressed m⁶A regulators that modify mitotic genes and contribute to NACT resistance. Analysis of surgically resected ESCC samples revealed a significant upregulation of the m⁶A reader YTHDF3 in NACT non-responders. To further investigate the role of YTHDF3, we performed functional proteomics analysis using YTHDF3 KD ESCC cells. KEGG pathway analysis showed enrichment of proteins involved in multiple cellular processes, including cell cycle regulation and cell migration. Although previous reports have linked YTHDF3 to pathways such as epithelial-to-mesenchymal transition (EMT) and mesenchymal-to-epithelial transition (MET)^46,47, these pathways were ranked lower in our analysis. Notably, cell cycle-related pathways, particularly those governing mitotic progression, consistently emerged among the top five enriched pathways. This finding was further supported by transcriptomic profiling of clinical ESCC samples, which highlighted cell division as a central process associated with YTHDF3 function.

Furthermore, our findings revealed that YTHDF3 directly promotes the m⁶A-dependent expression of key mitotic regulators, CDCA8 and INCENP, thereby reducing sensitivity to paclitaxel and cisplatin. This study uncovers a mechanism by which YTHDF3 KD suppresses the expression of INCENP and CDCA8, leading to multipolar spindle formation and consequently enhancing sensitivity to NACT. These findings highlight that m⁶A epigenetic modification of mitotic genes leads to mitotic dysregulation, which jointly contributes to NACT resistance. Collectively, INCENP and CDCA8 were identified as critical mitotic regulators associated with poor NACT response in ESCC patients, and targeting INCENP and CDCA8 emerges as a promising therapeutic strategy to enhance NACT efficacy.

Given that the combination of paclitaxel and cisplatin is a widely applied NACT regimen across multiple cancer types^35,36,37, we further evaluated the predictive value of INCENP and CDCA8 in other cancers. By analyzing pre-treatment biopsy samples from patients with ESCC, breast cancer, and lung cancer who exhibited differential responses to NACT, we found that high expression of INCENP and CDCA8 was associated with poor therapeutic outcomes. These markers effectively predicted NACT responsiveness across multiple cancer types. Further considering that INCENP and CDCA8 are both essential subunits of the chromosomal passenger complex, the loss of either INCENP or CDCA8 severely compromises CPC function. Therefore, INCENP and CDCA8 are likely to be equally important in predicting clinical responses to NACT. More importantly, inhibition of INCENP and CDCA8 to disrupt mitotic processes provides a therapeutic strategy to potentiate NACT responses, with the potential to improve survival outcomes across multiple cancer types.

A comprehensive understanding of the molecular pathways driving NACT resistance, coupled with the identification of robust predictive biomarkers and personalized therapeutic strategies, is essential for optimizing treatment regimens⁴⁸. Future research should focus on integrating combination therapies to overcome resistance mechanisms and enhance NACT efficacy, ultimately improving clinical outcomes for cancer patients.

Methods

Ethical compliance statement

The study complied with all relevant ethical regulations and was approved by the Medical Ethics Committee of Affiliated Cancer Hospital of Zhengzhou University & Henan Cancer Hospital (Approval No. 2021-KY-0239-001). Written informed consent was obtained from all patients. Sex and/or gender were not considered in the study design, and the sex of participants was determined based on information recorded in their medical records; gender identity was not separately collected in this study. Clinical information about the patients can be found in Supplementary Data 1–4.

Animal studies

Female athymic nude mice (nu/nu, Mus musculus, genetically immunodeficient, 6–8 weeks old, from Beijing HFK Bioscience) were used for the subcutaneous xenograft models. The mice were maintained under specific pathogen-free conditions in 12 h light/dark cycles at a temperature of 20 ± 2°C and humidity of 55 ± 10%. Mice had free access to water and rodent diet (Beijing keao xieli feed co., LTD, 1016706476803973120). This study was approved by the Ethical Committee of China-US (Henan) Hormel Cancer Institute (CUHCI2018063). All animal experiments complied with the ethics committee, and the maximum allowable tumor volume was set at 1000 mm³. Notably, the tumor volumes observed in this study did not exceed the specified limit.

Antibodies, key commercial kit, Plasmids, and software

Detailed information on the antibodies, key commercial kits, plasmids, and software used in this study is provided in Supplementary Table 1.

Cell lines and cell culture

Human ESCC cell lines (KYSE30, KYSE70, KYSE150, KYSE410, KYSE450, and KYSE510) and HEK293T were preserved and donated by Professor Ziming Dong (the Department of Pathophysiology, School of Basic Medical Sciences of Zhengzhou University). The immortalized normal esophageal epithelial cell line (SHEE) was generously provided by Dr. Enmin Li (Shantou University, Guangdong, China). Authentication of all cell lines was performed using STR analysis. ESCC cell lines were cultured in RPMI-1640 medium (VivaCell, Cat# C3010-0500), and HEK293T cells were cultured in DMEM high glucose medium (VivaCell, Cat# C3110-0500). Both RPMI-1640 medium and DMEM medium were supplemented with 10% FBS (VivaCell, Cat #2302008) and 1% penicillin/streptomycin. All the cell lines were maintained in a humidified incubator at 37 °C under 5% CO₂.

PDX-derived organoid (PDXO)

PDXO were established from fresh PDX tumor fragments and cultured according to methods described in a previous report⁴⁹. Briefly, the PDX tumor tissue was minced into small fragments and incubated in HBSS with dispase, collagenase Ⅳ, Y-27632, and Amphotericin B to create a single-cell suspension. Next, the tissue fragments were filtered, trypsinized, and further filtered through a 70 μm strainer. The cells were then centrifuged (200 × g) at 4 °C, and the cell pellet was washed and resuspended in basal medium prior to quantifying cell numbers using the Trypan Blue exclusion test. To initiate PDXO in a 24-well plate, 1 × 10⁴ live cells were seeded per well in Matrigel and cultured in organoid medium. The medium was replaced every 2–3 days, and organoid growth was monitored under a microscope. After 4-6 days, we used Mock, shYTHDF3, and shMETTL3 viruses for infection. On day 14, the organoids were photographed. Organoid numbers and size were quantified using the ImageJ software.

Western blotting

Cells were harvested with a cell scraper on ice and lysed with RIPA lysis buffer for 30 min (50 mM Tris-HCl, 0.5 mM NaCl, 10 mM imidazole, 5% glycerol, 1% NP-40, and 0.25% TritonX-100; pH adjusted to 8.0). Next, the cell lysates were centrifuged (14,000 × g, 10 min) at 4 °C, and the supernatant was transferred to a fresh tube. A BCA kit (Cat# PC0020, Solarbio) was used to quantify the protein concentration of each harvested lysate. 10–30 µg of protein was denatured in loading buffer and resolved via SDS-PAGE. The gel was subsequently transferred to a 0.45-μm PVDF membrane. The membranes were then blocked with 5% non-fat milk and incubated on a rotator with specific antibodies at 4 °C overnight. The next day, the membranes were washed with TBST and incubated with the secondary antibody for 1 h. The proteins were detected by using chemiluminescence (ECL) reagent. The membranes were exposed using an Amersham Imager 800.

Co-immunoprecipitation

Cell lysates were pre-cleared with IgG antibody and protein A/G agarose for 3 h at 4 °C. After centrifugation (600 × g, 5 min) at 4 °C, the cell lysates were incubated on a rotator overnight (4 °C) with IgG and specific antibodies. Protein A/G agarose was added to the samples, and they were rotated for another 2 h at 4 °C. The beads were then washed three times with lysis buffer. The beads were then eluted with 2×SDS loading buffer. The eluted proteins were heated at 95 °C for 10 min and resolved using Western blotting.

Cell cycle

In preparation for cell cycle analysis, cells were seeded in 60 mm dishes for 48 h with Ebselen treatment. The cells were fixed in 70% pre-cooled ethanol for 24 h and washed with 1 mL PBS three times. The cells were resuspended with 250 µL 0.5% Triton X-100 and incubated with 5 µL 10 µg/mL RNase A at room temperature for 1 h. The cells were then incubated with 5 µL 1 mg/mL propidium iodide at room temperature for 20 min. After staining, the cells were analyzed using a flow cytometer (BD Biosciences, San Jose, CA).

Cell proliferation and colony formation assays

For the cell proliferation assay, 1500 cells were seeded into 96-well plates, and cell viability was assessed for 72 h with or without paclitaxel and cisplatin treatment using an MTT assay. For the colony formation assay, 1000 cells were seeded into 6-well plates for approximately 10 days with or without paclitaxel and cisplatin treatment. Afterward, the medium was aspirated, and the plate was gently washed with distilled water. Finally, the cells were stained with crystal violet and counted at the endpoint. All studies were conducted at least in triplicate.

Plasmid construction

Candidate genes ORF expression plasmid were constructed by YouBio Biology Company (Changsha, China). Short hairpin RNAs (shRNA) were designed and subcloned into the pLKO.1 vector. All plasmids were used after being confirmed by sequencing. The primers are listed in Supplementary Table 2, and the plasmid information is listed on Supplementary Table 1.

Cell transfection and lentiviral infection

The lentiviral shRNA/ORF plasmid was co-transfected with packaging vectors (pMD2G and psPAX2) into HEK293T cells using Lipo2000 transfection reagent. After transfection, virus particles were harvested at 48 h and filtered using a 0.45 µm filter. To establish stable knockdown cell lines, cells were infected with the designated virus particles along with 8 µg/ml polybrene. After 24 h of infection, selection of stable cell lines was performed by incubation with 2–5 μg/mL puromycin for 48 h. Stable knockdown cell lines were generated through 2–3 passages. To overexpress YTHDF3, INCENP, and CDCA8, a pcDNA3.1-3×Flag-YTHDF3, pcDNA3.1-myc-INCENP, or pcDNA3.1-myc-CDCA8 plasmid was transfected into ESCC cells using the Lipo2000 reagent. After 4 h, the culture medium was replaced with fresh complete growth medium, and the cells were incubated for an additional 24 h. The knockdown and overexpression efficiency were evaluated by a Western blotting.

In vivo study

KYSE150 cells stably transfected with pLKO.1 were used as a mock control, and KYSE150 shINCENP or shCDCA8 cells were resuspended in cold-PBS with 1‰ penicillin-streptomycin. A total of 5 × 10⁶ cells were subcutaneously injected into the flank of nude female mice. Mice were randomly divided into Mock (n = 15), shINCENP (n = 15), and shCDCA8 (n = 15) groups. Each group was further divided into three subgroups, with five mice per subgroup: (1) vehicle, (2) 10 mg/kg paclitaxel per week (i.p.), (3) 10 mg/kg cisplatin per week (i.p.). For the Ebselen combination treatment experiment, mice were injected with KYSE150 cells and divided into six groups, with six mice per group: (1) vehicle, (2) 20 mg/kg Ebselen (i.p.), five times per week, (3) 10 mg/kg paclitaxel/week (i.p.), (4) 10 mg/kg cisplatin/week (i.p.), and (5) 10 mg/kg paclitaxel combined with 20 mg/kg Ebselen, and (6) 10 mg/kg cisplatin combined with 20 mg/kg Ebselen. Tumor volume was measured twice per week throughout the experiment. At the end of the study, mice were euthanized, and tumors were extracted for further analysis. Tumor volume (T) was calculated as T = (length × width²)/2. The maximum tumor size allowed was 1000 mm³ according to the Animal Ethics Committee. Notably, the maximal tumor burden didn’t exceed the limitation in this study. Nude mice aged 6-8 weeks were used to evaluate the tumorigenesis of the cells, with no specific gender requirements. Female mice were chosen for their more moderate character.

Immunofluorescence

KYSE30 and KYSE150 cells were seeded into a 24-well plate at 60% density with a slide. After 24 h, the cell culture medium was aspirated, and the cells were rinsed three times with cold PBS. The cells were then fixed in 400 μL of cold methanol for 25 min at room temperature, washed three times with PBS, and incubated overnight at 4 °C with YTHDF3 and eIF3A antibody (1:100, in solution 2% BSA, 0.3% Triton-100, 1% normal goat serum, 0.1% NaN₃). The next day, the cells were washed with PBS and then incubated with the appropriate secondary antibody (Goat anti-rabbit IgG H&L (Alexa Fluor 488), Goat anti-mouse IgG(H + L) (Alexa Fluor 568)) and DAPI. Then, the slides were washed with PBS, air-dried, and mounted. Finally, images were captured by a Nikon confocal microscope. For cell mitosis visualization, cells were seeded into a 24-well plate at 60% density onto slides. 16 h post-seeding, the cells were fixed, washed, and incubated 2 h at RT with α-Tubulin antibody (1:300, in solution 2% BSA, 0.3% Triton-100, 1% Normal goat serum, 0.1% NaN₃) and DAPI. Afterwards, the slides were washed with PBS, air dried, and mounted. Finally, images were captured by a Nikon confocal microscope. For organoid slides, the organoids were embedded in paraffin and cut into slices before IF detection. Slices were subjected to antigen retrieval and blocking with 5%BSA. The tissue slices were incubated with 30 µL primary antibodies, including YTHDF3, METTL3, INCENP, and CDCA8 antibodies (1:100 in 2% BSA solution) at 4 °C overnight. The slices were then washed with 1×PBS and incubated with the appropriate secondary antibody (Goat anti-rabbit IgG H&L (Alexa Fluor 488), Goat anti-mouse IgG(H + L) (Alexa Fluor 568), 1:500) and DAPI. Then, the slides were washed with PBS, air-dried, and mounted. Finally, images were captured by a Nikon confocal microscope. The relative protein expression of IF images was quantified using ImageJ software. Fluorescence intensity was measured and normalized by calculating the average intensity per cell or per unit area, effectively controlling for variations in cell numbers. Comparisons between experimental groups were made to assess relative changes in protein expression levels.

Protein synthesis assay

After KYSE30 and KYSE150 cells were transfected with 3×flag-YTHDF3 for 48 h, the nascently synthesized proteins were labeled with 10 µM Click-iT® OPP for another 30 min according to the manufacturer’s specifications of the Protein Synthesis Assay Kit. After fixation with 3.7% formaldehyde for 15 min, the cells were subsequently stained using a 100 µL OPP reaction cocktail and NuclearMask™ Blue. The stained cells were imaged using a Nikon confocal microscope and analyzed by ImageJ.

RNA Isolation and Quantitative Real-Time PCR (qPCR)

Cells were harvested and lysed using Trizol reagent according to the manufacturer’s instructions. Total RNA was then reverse-transcribed into cDNA using the Prime RT Reagent Kit (Vazyme, Cat#R323). Next, 20 ng of cDNA was then combined with gene-specific primers and SYBR qPCR Master Mix (Vazyme, Cat#Q711). Relative mRNA levels were then measured using a quantitative real-time PCR assay. The primers were listed in Supplementary Table 2.

Dual-luciferase reporter assay

ESCC cell lines were seeded in 48-well plates to allow 70% confluence the following day. The reporter plasmids pmirGLO-INCENP and pmirGLO-CDCA8 were then transfected into mock and shYTHDF3 cells at 200 ng per well using Lipofactamine 2000 (Invitrogen) according to the manufacturer’s protocol. 48 h later, firefly luciferase and renilla luciferase activities were measured using the Dual-Luciferase Reporter Assay System (Promega) according to the supplier’s instructions.

SUnSET assay

The SUnSET assay was performed to monitor protein synthesis rates using a western blot. Briefly, pcDNA3.1-3×flag vector and Flag-YTHDF3 transfected cells were seeded in a 12-well plate at a density of 5 × 10⁴ cells per well and incubated in culture medium supplemented with puromycin (2 μg/mL) for the indicated time periods. Proteins were then extracted from the cells, and Western blotting was conducted using an anti-puromycin antibody.

Proximity ligation assay (PLA)

PLA was performed according to the Duolink® In Situ Red Starter Kit (Mouse/Rabbit). In detail, cells were seeded onto sterile coverslips in a 24-well tissue-culture plate supplemented with 1 ml complete culture media and incubated at 37 °C under 5% CO₂. After incubation for 24 h, the supernatant was removed and the cells were rinsed three times with 1 mL PBS, followed by fixation with 4% paraformaldehyde for 15 min. After three washes with PBS, the cells were incubated with 0.6% TritonX-100 Buffer for 15 min. The cells were then blocked with Duolink block solution for 1 h at 37 °C, and incubated overnight at 4 °C with the following antibodies: mouse anti-eIF3A and rabbit anti-YTHDF3. Subsequently, the cells were washed twice with 40 μL of 1×Duolink in situ wash buffer A and incubated in a mixture of PLA probes (anti-rabbit PLUS and anti-mouse MINUS, respectively) for 1 h at 37 °C. The cells were washed twice with 1×Duolink in situ wash buffer A and incubated with ligation mix at 37 °C for 30 min. The cells were then washed twice with 1×Duolink in situ wash buffer A and incubated with amplification mix at 37 °C for 100 min. Finally, the cells were washed twice with 1×Duolink in situ wash buffer B, washed once with 0.01×Buffer B, and mounted with Duolink in situ mounting medium with DAPI. Images were captured using an A1R Nikon confocal microscope.

m⁶A RNA methylation assay

For quantification of methylated m⁶A RNA, total RNA was extracted from human cells, and 200 ng RNA was used as input for the EpiQuik™m⁶A RNA Methylation Quantification Kit (Colorimetric). The kit was used according to the manufacturer’s instructions. The absolute m⁶A abundance of the sample was proportional to the absorbance at a wavelength of 450 nm as measured by a microplate reader (Thermo Scientific Multiskan GO).

Immunohistochemistry

The use of paraffin-embedded human ESCC (cohort 1, n = 14; cohort 2, n = 68; cohort 3, n = 65), lung cancer (n = 26), and breast cancer (n = 22) tissue samples was approved by the Affiliated Cancer Hospital of Zhengzhou University & Henan Cancer Hospital. All tissue samples were collected in compliance with the informed consent policy. For the detailed method, the tissues were subjected to antigen retrieval and blocking according to the rabbit or mouse SP Detection Kit (Cat#SP9001, ZSGB-BIO). The tissue slices were incubated with 50 µL YTHDF3, INCENP, and CDCA8 antibodies at 4 °C overnight. The slices were then washed with 1×PBS and incubated with secondary antibodies at room temperature for 30 min. The slices were then stained with 3,3’-diaminobenzidine (DAB) working buffer. After re-staining with hematoxylin, the slices were photographed and analyzed using the ImageJ software. The staining of the tissue sections was quantitatively scored according to the percentage of positive cells. The following proportion scores were assigned to the sections: 0 for 0% of the tumor cell exhibited positive staining, 1 for 0–1% of cells stained, 2 for 2–10% of cells stained, 3 for 11–30% of cells stained, 4 for 31–70% of cells stained and 5 for 71–100% of cells stained. In addition, the staining intensity was rated on a scale of 0–3: 0, no staining; 1, weak; 2, moderate; and 3, strong. The proportion and intensity scores were then combined to obtain a total score (range, 0–8). The scores were compared with progression-free survival (PFS), which is defined as the time from the start of treatment to the time of disease progression.

meRIP assay

EpiQuik™ CUT&RUN m⁶A RNA Enrichment Kit was used to investigate the m⁶A modification of specific mRNA transcripts. 50 μg total RNA was isolated using Trizol reagent from KYSE30 and KYSE150 cells. Next, RNA sequence fragments containing the m⁶A target were cleaved, bound to magnetic beads containing an m⁶A antibody, and pulled down. Finally, the enriched RNA was released, purified, and eluted. According to the information provided on the SRAMP platform (http://www.cuilab.cn/sramp), we designed primers for specific mRNA transcripts. qPCR was subsequently conducted to detect differences in m⁶A modification of certain transcripts.

RIP-qPCR analysis

Tumor cells were grown in a 10-cm dish until 80% confluence was achieved. The cells were then harvested, and RNA immunoprecipitation was performed using the RNA Immunoprecipitation Kit according to the manufacturer’s instructions. Briefly, magnetic beads coated with normal antibodies against YTHDF3, eIF3A, mouse normal IgG, or rabbit normal IgG were individually incubated with cell lysates overnight at 4 °C. Collected RNA-protein complexes were washed five times and then subjected to RNA extraction for further analysis by quantitative real-time PCR. The primers were listed in Supplementary Table 2.

RIP-seq analysis

The RIP assay was performed on KYSE30 cells by SeqHealth (Wuhan, China). Cells were collected and lysed with cell lysis buffer. Ten percent (10%) of the lysate was stored as “input”, 80% was used in immunoprecipitation reactions with anti-YTHDF3 antibody (Abcam, ab220161), and the remaining 10% was incubated with rabbit IgG (Cell Signaling Technology) as a negative control and named “IgG”, respectively. The RNA of input and IP was extracted using TRIzol reagent (Invitrogen, cat. NO 15596026). The stranded RNA sequencing library was constructed by using KC-DigitalTM Stranded mRNA Library Prep Kit for Illumina® (Catalog NO. DR08502, Wuhan Seqhealth Co., Ltd, China) following the manufacturer’s instructions. The library products corresponding to 200–500 bp were enriched, quantified, and finally sequenced on a DNBSEQ-T7 sequencer (MGI Tech Co., Ltd, China) with the PE150 model. Raw sequencing data were initially filtered by Trimmomatic (version 0.36). Low-quality reads were discarded, and reads contaminated with adaptor sequences were trimmed. Clean reads were further treated with in-house scripts to eliminate duplication bias introduced in library preparation and sequencing. In brief, clean reads were first clustered by UMI sequence, and reads sharing the same UMI sequence were grouped into the same cluster. Reads in the same cluster were compared to each other by pairwise alignment, and then reads with sequence identity over 95% were extracted to a new sub-cluster. After all sub-clusters were generated, multiple sequence alignment was performed to get one consensus sequence for each sub-cluster. After these steps, any errors and biases introduced by PCR amplification or sequencing were eliminated. The de-duplicated consensus sequences were used for protein binding site analysis. Sequences were mapped to the reference Homo sapiens (human) genome from NCBI using STAR software (version 2.5.3a) with default parameters. The RSeQC (version 2.6) was used for read distribution analysis. The exomePeak (Version 3.8) software was used for peak calling. Peaks were annotated using bedtools (Version 2.25.0). The deepTools (version 2.4.1) was used for peak distribution analysis. The differentially binding peaks were identified by a Python script using the Fisher test. Sequence motifs enriched in peak regions were identified using Homer (version 4.10). Gene ontology (GO) analysis and Kyoto encyclopedia of genes and genomes (KEGG) enrichment analysis for annotated genes were both implemented by KOBAS software (version: 2.1.1) with a corrected P-value cutoff of 0.05 to judge statistically significant enrichment.

Transcriptomic analysis

ESCC tumor tissues (cohort 1, n = 14) from patients were collected during surgery in the Affiliated Cancer Hospital of Zhengzhou University & Henan Cancer Hospital in China. Appropriate tissue samples were transferred into the corresponding tissue grinding tube for sample lysis using TRIzol reagent, then RNA purification, quality testing, mRNA isolation, fragmentation, cDNA synthesis, end repair, Add A and Adaptor Ligation, PCR, corresponding library quality control, circularization, and sequencing process were performed. The sequencing data were filtered with SOAPnuke; afterwards, clean reads were obtained. Bowtie2 was applied to align the clean reads to the gene set, in which known coding and noncoding transcripts were included. The expression levels of genes were calculated by RSEM (v1.3.1). Essentially, differential expression analysis was performed using the DESeq2 (v1.4.5) (or DEGseq or PoissonDis) with Q value ≤ 0.05 (or FDR ≤ 0.001). To gain insight into the change of phenotype, GO (http://www.geneontology.org/) and KEGG enrichment analysis of annotated differentially expressed genes was performed by Phyper based on the Hypergeometric test. The significant levels of terms and pathways were corrected by Q value with a rigorous threshold (Q value ≤ 0.05).

Label-free proteomic analysis

The proteomic analysis was performed on mock and shYTHDF3 KYSE30 cells (n = 3 technical replicates per group). In total, six samples were analyzed: three for mock cells and three for shYTHDF3 cells. As controls, mock-treated cells were included. Initially, total cellular protein was extracted using SDT (4% SDS, 100 mM Tris-HCl, pH 7.6) buffer. The protein concentration was quantified using a BCA Protein Assay Kit (Bio-Rad, USA). 20 µg of protein for each sample was individually mixed with 5×loading buffer and boiled for 5 min. The proteins were then resolved on a 4–20% SDS-PAGE gel (constant voltage 180 V, 45 min). Protein bands were visualized by Coomassie Blue R-250 staining. Protein digestion by trypsin was performed according to the filter-aided sample preparation (FASP) procedure described by Wisniewski⁵⁰. LC-MS/MS was performed on an Orbitrap Astral mass spectrometer coupled with a Vanquish Neo UHPLC system (Thermo Fisher Scientific). Peptides from each sample were loaded into a column (50 cm Low-Load µPAC™ Neo HPLC Column, Thermo Scientific) at a flow rate of 2.2 μL/min. Peptides were eluted with a gradient of 0.1% formic acid in water (A) and 0.1% formic acid in 80% acetonitrile (B) at 2.2 μL/min. The eluted peptides were analyzed on an Orbitrap Astral mass spectrometer. DIA MS/MS scans were acquired from 150 to 2000 m/z with a 2 m/z isolation window, 500% AGC target, and 3 ms injection time. Data were recorded in profile mode for MS scans and centroid mode for DIA scans. Data analysis was performed using DIA-NN 1.8.1⁵¹ for protein identification, with the MS data searched against the UniProtKB human protein database⁵². Trypsin was used for digestion, with one missed cleavage allowed and a mass tolerance of 10 ppm. Carbamidomethylation of cysteines was set as a fixed modification, and acetylation of the N-terminus and oxidation of methionine were variable modifications. Results were filtered with a false discovery rate (FDR) < 1%. Bioinformatics analysis was performed in R, including hierarchical clustering, volcano plots, GO, and KEGG enrichment analysis using Fisher’s exact test p < 0.01 with FDR correction. Protein–protein interaction (PPI) networks were constructed using STRING and Cytoscape software⁵³.

Statistical analysis

Statistical analyses were conducted using the GraphPad Prism 9 software. Only biological replicates were used for statistical analyses. A two-tailed unpaired Student’s t-test was conducted for comparison of means of data between two groups. One-way ANOVA was used to compare multiple independent groups, whereas two-way ANOVA was employed to analyze the interaction between two independent variables and their effects on the dependent variable. All graphs are presented as mean ± SD and exact P values; P < 0.05 was recognized as statistically significant.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

The transcriptomic raw sequence data generated in this study have been deposited in the GSA-Human database under the accession code HRA011110. The raw sequence data are available under restricted access for research purposes only, access can be obtained by the DAC (Data Access Committees) of the GSA-human database. According to the guide lines of GSA-human, all non-profit researchers are allowed access to the data, and the Principle Investigator of any research group is allowed to apply for Controlled-access of the data. The user can register the GSA database [https://ngdc.cncb.ac.cn/gsa-human/] and request the data. The approximate response time for accession requests is about 3 days. The access authority can be obtained for Research Use Only. The user can also contact the corresponding author directly. Once access has been approved, the data will be available to download for 2 months. The mass spectrometry proteomics publicly available data for Mock and shYTHDF3 cells generated in this study have been deposited in the ProteomeXchange Consortium database under accession code PXD062909 [https://www.iprox.cn//page/project.html?id=IPX0011698000]. The original RIP-Seq publicly available data for KYSE30 cells have been deposited in the NCBI GEO database under the accession code GSE294561. The remaining data are available in the Article, Supplementary Information or Source Data file. Source data are provided with this paper.

References

Varshney, V. K. et al. Neoadjuvant chemotherapy vs chemoradiotherapy for malignancy of esophagus: a prospective comparative study. Cureus 16, e73525 (2024).
PubMed PubMed Central Google Scholar
Patel, A. et al. Emerging Trends in neoadjuvant chemotherapy for ovarian cancer. Cancers 13, https://doi.org/10.3390/cancers13040626 (2021).
Zarotti, C. et al. Biomarker dynamics and prognosis in breast cancer after neoadjuvant chemotherapy. Sci. Rep. 12, 91 (2022).
Article ADS CAS PubMed PubMed Central Google Scholar
Passaro, A., Attili, I. & de Marinis, F. Neoadjuvant chemotherapy plus immunotherapy in early-stage resectable non-small-cell lung cancer. J. Clin. Oncol. 40, 2871–2877 (2022).
Article CAS PubMed Google Scholar
Stathopoulos, G. P. et al. Liposomal cisplatin combined with paclitaxel versus cisplatin and paclitaxel in non-small-cell lung cancer: a randomized phase III multicenter trial. Ann. Oncol. 21, 2227–2232 (2010).
Article CAS PubMed PubMed Central Google Scholar
Ezzat, A. A. et al. Phase II study of neoadjuvant paclitaxel and cisplatin for operable and locally advanced breast cancer: analysis of 126 patients. Br. J. Cancer 90, 968–974 (2004).
Article CAS PubMed PubMed Central Google Scholar
Polee, M. B. et al. Phase II study of neo-adjuvant chemotherapy with paclitaxel and cisplatin given every 2 weeks for patients with a resectable squamous cell carcinoma of the esophagus. Ann. Oncol. 14, 1253–1257 (2003).
Article CAS PubMed Google Scholar
Ai, D. et al. Comparison of 3 Paclitaxel-based chemoradiotherapy regimens for patients with locally advanced esophageal squamous cell cancer: a randomized clinical trial. JAMA Netw. Open 5, e220120 (2022).
Article PubMed PubMed Central Google Scholar
Zhang, W. et al. Encouraging pathological complete response rate from neoadjuvant chemotherapy with albumin-bound paclitaxel plus cisplatin and capecitabine for locally advanced esophageal squamous carcinoma: preliminary outcome of a retrospective study. Cancer Manag. Res. 13, 2163–2170 (2021).
Article ADS CAS PubMed PubMed Central Google Scholar
Eyck, B. M. et al. Pathological response to neoadjuvant chemoradiotherapy for oesophageal squamous cell carcinoma: multicentre East Asian and Dutch database comparison. Br. J. Surg. 109, 1312–1318 (2022).
Article PubMed PubMed Central Google Scholar
An, J., Peng, C., Tang, H., Liu, X. & Peng, F. New advances in the research of resistance to neoadjuvant chemotherapy in breast cancer. Int. J. Mol. Sci. 22, https://doi.org/10.3390/ijms22179644 (2021).
Yaniz-Galende, E. et al. Spatial profiling of ovarian carcinoma and tumor microenvironment evolution under neoadjuvant chemotherapy. Clin. Cancer Res. 30, 2790–2800 (2024).
Article CAS PubMed PubMed Central Google Scholar
Yang, Y. et al. Comprehensive landscape of resistance mechanisms for neoadjuvant therapy in esophageal squamous cell carcinoma by single-cell transcriptomics. Signal Transduct. Target Ther. 8, 298 (2023).
Article ADS PubMed PubMed Central Google Scholar
Scribano, C. M. et al. Chromosomal instability sensitizes patient breast tumors to multipolar divisions induced by paclitaxel. Sci. Transl. Med. 13, eabd4811 (2021).
Article CAS PubMed PubMed Central Google Scholar
Rodrigues-Ferreira, S. et al. Improving breast cancer sensitivity to paclitaxel by increasing aneuploidy. Proc. Natl. Acad. Sci. USA 116, 23691–23697 (2019).
Article ADS CAS PubMed PubMed Central Google Scholar
Liao, L. et al. Protein Phosphatase 1 Subunit PPP1R14B Stabilizes STMN1 to promote progression and paclitaxel resistance in triple-negative breast cancer. Cancer Res. 83, 471–484 (2023).
Article CAS PubMed PubMed Central Google Scholar
Huang, K. C. et al. Cyclin A1 expression and paclitaxel resistance in human ovarian cancer cells. Eur. J. Cancer 67, 152–163 (2016).
Article PubMed PubMed Central Google Scholar
Roberts, M. S. et al. LIN9 and NEK2 are core regulators of mitotic fidelity that can be therapeutically targeted to overcome taxane resistance. Cancer Res. 80, 1693–1706 (2020).
Article CAS PubMed PubMed Central Google Scholar
Agnese, V. et al. The role of Aurora-A inhibitors in cancer therapy. Ann. Oncol. 18, vi47–vi52 (2007).
Article PubMed Google Scholar
Horibe, S. et al. Cisplatin resistance in human lung cancer cells is linked with dysregulation of cell cycle associated proteins. Life Sci. 124, 31–40 (2015).
Article CAS PubMed Google Scholar
Hsu, W. H. et al. Checkpoint Kinase 1 inhibition enhances cisplatin cytotoxicity and overcomes cisplatin resistance in SCLC by promoting mitotic cell death. J. Thorac. Oncol. 14, 1032–1045 (2019).
Article CAS PubMed PubMed Central Google Scholar
Carmena, M., Wheelock, M., Funabiki, H. & Earnshaw, W. C. The chromosomal passenger complex (CPC): from easy rider to the godfather of mitosis. Nat. Rev. Mol. Cell Biol. 13, 789–803 (2012).
Article CAS PubMed PubMed Central Google Scholar
Hindriksen, S., Meppelink, A. & Lens, S. M. Functionality of the chromosomal passenger complex in cancer. Biochem Soc. Trans. 43, 23–32 (2015).
Article CAS PubMed Google Scholar
Sampath, S. C. et al. The chromosomal passenger complex is required for chromatin-induced microtubule stabilization and spindle assembly. Cell 118, 187–202 (2004).
Article CAS PubMed Google Scholar
Frye, M., Harada, B. T., Behm, M. & He, C. RNA modifications modulate gene expression during development. Science 361, 1346–1349 (2018).
Article ADS CAS PubMed PubMed Central Google Scholar
Wang, T., Kong, S., Tao, M. & Ju, S. The potential role of RNA N6-methyladenosine in Cancer progression. Mol. Cancer 19, 88 (2020).
Article CAS PubMed PubMed Central Google Scholar
Shi, H., Wei, J. & He, C. Where, when, and how: context-dependent functions of RNA methylation writers, readers, and erasers. Mol. Cell 74, 640–650 (2019).
Article CAS PubMed PubMed Central Google Scholar
Lan, Q. et al. The emerging roles of RNA m(6)A methylation and demethylation as critical regulators of tumorigenesis, drug sensitivity, and resistance. Cancer Res. 81, 3431–3440 (2021).
Article CAS PubMed Google Scholar
Fan, X. et al. YTHDF2-mediated m(6)A modification of ONECUT2 promotes stemness and oxaliplatin resistance in gastric cancer through transcriptionally activating TFPI. Drug Resist Updat 79, 101200 (2025).
Article CAS PubMed Google Scholar
Ajani, J. A. et al. Esophageal and esophagogastric junction cancers, Version 2.2023, NCCN clinical practice guidelines in oncology. J. Natl. Compr. Cancer Netw. 21, 393–422 (2023).
Article Google Scholar
Shi, H. et al. YTHDF3 facilitates translation and decay of N(6)-methyladenosine-modified RNA. Cell Res. 27, 315–328 (2017).
Article CAS PubMed PubMed Central Google Scholar
Liu, Z. et al. Biological and pharmacological roles of m(6)A modifications in cancer drug resistance. Mol. Cancer 21, 220 (2022).
Article CAS PubMed PubMed Central Google Scholar
Zhou, Y. et al. Targeting treatment resistance: unveiling the potential of RNA methylation regulators and TG-101,209 in pan-cancer neoadjuvant therapy. J. Exp. Clin. Cancer Res. 43, 232 (2024).
Article CAS PubMed PubMed Central Google Scholar
Chang, G. et al. YTHDF3 induces the translation of m(6)A-enriched gene transcripts to promote breast cancer brain metastasis. Cancer Cell 38, 857–871.e857 (2020).
Article CAS PubMed PubMed Central Google Scholar
Wang, B. et al. A randomized phase 3 trial of Gemcitabine or Nab-paclitaxel combined with cisplatin as first-line treatment in patients with metastatic triple-negative breast cancer. Nat. Commun. 13, 4025 (2022).
Article ADS CAS PubMed PubMed Central Google Scholar
Ni, M. et al. Comparative efficacy of Chinese herbal injections combined with paclitaxel plus cisplatin for non-small-cell lung cancer: a multidimensional bayesian network meta-analysis. Evid. Based Complement Altern. Med. 2020, 1824536 (2020).
Article Google Scholar
Wu, M. F. et al. Determining the maximum tolerated dose of paclitaxel combined with fixed dose of cisplatin for hyperthermic intraperitoneal chemotherapy in ovarian cancer: a multicenter phase I trial. Gynecol. Oncol. 181, 125–132 (2024).
Article CAS PubMed Google Scholar
Zhang, X. et al. A phase II trial of paclitaxel and cisplatin in patients with advanced squamous-cell carcinoma of the esophagus. Am. J. Clin. Oncol. 31, 29–33 (2008).
Article PubMed Google Scholar
Li, F., Zhao, C. & Wang, L. Molecular-targeted agents combination therapy for cancer: developments and potentials. Int. J. Cancer 134, 1257–1269 (2014).
Article CAS PubMed Google Scholar
Lawrence, M. S. et al. Mutational heterogeneity in cancer and the search for new cancer-associated genes. Nature 499, 214–218 (2013).
Article ADS CAS PubMed PubMed Central Google Scholar
Lee, Y. T., Tan, Y. J. & Oon, C. E. Molecular targeted therapy: treating cancer with specificity. Eur. J. Pharm. 834, 188–196 (2018).
Article CAS Google Scholar
Deng, X., Qing, Y., Horne, D., Huang, H. & Chen, J. The roles and implications of RNA m(6)A modification in cancer. Nat. Rev. Clin. Oncol. 20, 507–526 (2023).
Article CAS PubMed PubMed Central Google Scholar
Zhang, Z. et al. m(6)A regulators as predictive biomarkers for chemotherapy benefit and potential therapeutic targets for overcoming chemotherapy resistance in small-cell lung cancer. J. Hematol. Oncol. 14, 190 (2021).
Article CAS PubMed PubMed Central Google Scholar
Visvanathan, A. et al. Essential role of METTL3-mediated m(6)A modification in glioma stem-like cells maintenance and radioresistance. Oncogene 37, 522–533 (2018).
Article CAS PubMed Google Scholar
Zhou, S. et al. FTO regulates the chemo-radiotherapy resistance of cervical squamous cell carcinoma (CSCC) by targeting beta-catenin through mRNA demethylation. Mol. Carcinog. 57, 590–597 (2018).
Article CAS PubMed Google Scholar
Chen, H. et al. YTHDF3 drives tumor growth and metastasis by recruiting eIF4B to promote Notch2 translation in breast cancer. Cancer Lett. 614, 217534 (2025).
Article CAS PubMed Google Scholar
Li, X. et al. N6-methyladenosine reader YTHDF3-mediated zinc finger protein 41 inhibits hepatocellular carcinoma progression by transcriptional repression of Snail. MedComm 5, e763 (2024).
Article ADS CAS PubMed PubMed Central Google Scholar
La Thangue, N. B. & Kerr, D. J. Predictive biomarkers: a paradigm shift towards personalized cancer medicine. Nat. Rev. Clin. Oncol. 8, 587–596 (2011).
Article PubMed Google Scholar
Karakasheva, T. A. et al. Generation and characterization of patient-derived head and neck, oral, and esophageal cancer organoids. Curr. Protoc. Stem Cell Biol. 53, e109 (2020).
Article CAS PubMed PubMed Central Google Scholar
Wisniewski, J. R., Zougman, A., Nagaraj, N. & Mann, M. Universal sample preparation method for proteome analysis. Nat. Methods 6, 359–362 (2009).
Article CAS PubMed Google Scholar
Demichev, V., Messner, C. B., Vernardis, S. I., Lilley, K. S. & Ralser, M. DIA-NN: neural networks and interference correction enable deep proteome coverage in high throughput. Nat. Methods 17, 41–44 (2020).
Article CAS PubMed Google Scholar
Bekker-Jensen, D. B. et al. Rapid and site-specific deep phosphoproteome profiling by data-independent acquisition without the need for spectral libraries. Nat. Commun. 11, 787 (2020).
Article ADS CAS PubMed PubMed Central Google Scholar
Kohl, M., Wiese, S. & Warscheid, B. Cytoscape: software for visualization and analysis of biological networks. Methods Mol. Biol. 696, 291–303 (2011).
Article CAS PubMed Google Scholar

Download references

Acknowledgements

This work was supported by the National Natural Science Foundation of China [No. 82573804 (Z.D.), 32571376 (S.Z.), 82203290 (S.Z.), 82503996 (T.W.)]; Major Science and Technology Project of Henan Province (No. 221100310100, Z.D.); the Key projects jointly built by Henan Province and the Ministry of Finance [No. SBGJ202502038 (S.Z.), 242300420096, (Q.X.)]; the Natural Science Foundation of Henan Province (252300423924, X.W.); the Young and Middle-aged Scientific and Technological Innovation Talent Program (JQRC2025003, S.Z.).

Author information

These authors contributed equally: Xiangyu Wang, Ting Wang.

Authors and Affiliations

Department of Pathophysiology, School of Basic Medical Sciences, Zhengzhou University, Zhengzhou, Henan, China
Xiangyu Wang, Ting Wang, Zhibo Li, Feifei Liu, Xiaodan Shi, Zihan Zhang, Rui Wang, Ludan Jia & Zigang Dong
China-US (Henan) Hormel Cancer Institute, Zhengzhou, Henan, China
Xiangyu Wang, Ting Wang, Keke Wang, Zhibo Li, Feifei Liu, Xueli Tian, Xiaodan Shi, Zihan Zhang, Rui Wang, Ludan Jia, Kyle Vaughn Laster & Zigang Dong
Department of Pathology, Affiliated Cancer Hospital of Zhengzhou University & Henan Cancer Hospital, Zhengzhou, Henan, China
Chengjuan Zhang, Qingxin Xia, Simin Zhao & Zigang Dong
Department of Clinical Research and Translational Medicine, The Third Affiliated Hospital of Zhengzhou University, Zhengzhou, China
Xueli Tian
Tianjian Laboratory of Advanced Biomedical Sciences, Institute of Advanced Bio-medical Sciences, Cancer Prevention and Treatment Lab, Zhengzhou University, Zhengzhou, Henan, China
Simin Zhao & Zigang Dong
The Collaborative Innovation Center of Henan Province for Cancer Chemoprevention, Zhengzhou, Henan, China
Simin Zhao & Zigang Dong

Authors

Xiangyu Wang
View author publications
Search author on:PubMed Google Scholar
Ting Wang
View author publications
Search author on:PubMed Google Scholar
Keke Wang
View author publications
Search author on:PubMed Google Scholar
Chengjuan Zhang
View author publications
Search author on:PubMed Google Scholar
Zhibo Li
View author publications
Search author on:PubMed Google Scholar
Feifei Liu
View author publications
Search author on:PubMed Google Scholar
Xueli Tian
View author publications
Search author on:PubMed Google Scholar
Xiaodan Shi
View author publications
Search author on:PubMed Google Scholar
Zihan Zhang
View author publications
Search author on:PubMed Google Scholar
Rui Wang
View author publications
Search author on:PubMed Google Scholar
Ludan Jia
View author publications
Search author on:PubMed Google Scholar
Kyle Vaughn Laster
View author publications
Search author on:PubMed Google Scholar
Qingxin Xia
View author publications
Search author on:PubMed Google Scholar
Simin Zhao
View author publications
Search author on:PubMed Google Scholar
Zigang Dong
View author publications
Search author on:PubMed Google Scholar

Contributions

Writing original draft: X.W.; Workshop director and co-first authors: T.W.; data analysis and manuscript editing: X.W., K.W., and K.L.; figures editing: K.W.; methodology: F.L., L.J. and X.T.; Clinical samples collection and annotation: C.Z., Z.L., X.S., Z.Z., R.W.; Formal analysis: X.W., T.W., K.W., Z.D., and S.Z.; Investigation: X.W.; project administration and supervision: Q.X., Z.D. and S.Z; Funding acquisition: Z.D., Q.X., S.Z., T.W., and X.W.; Conceptualization, review, editing, co-corresponding author: Z.D. and S.Z. All authors have read and approved the article.

Corresponding authors

Correspondence to Simin Zhao or Zigang Dong.

Ethics declarations

Competing interests

The authors declare no competing interests.

Peer review

Peer review information

Nature Communications thanks Paraskevi Giannakakou, who co-reviewed with Urko del Castillo Rojo, Anutosh Ganguly, Yan-Yi Jiang, and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary information

supplementary information (download PDF )

Description of Additional Supplementary Files (download PDF )

Supplementary Data 1 (download XLSX )

Supplementary Data 2 (download XLSX )

Supplementary Data 3 (download XLSX )

Supplementary Data 4 (download XLSX )

Reporting Summary (download PDF )

Transparent Peer Review file (download PDF )

Source data

source data (download XLSX )

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

Wang, X., Wang, T., Wang, K. et al. INCENP and CDCA8 predict neoadjuvant chemotherapy response and outcomes in esophageal squamous cell carcinoma. Nat Commun 17, 1672 (2026). https://doi.org/10.1038/s41467-026-68371-x

Download citation

Received: 06 August 2024
Accepted: 02 January 2026
Published: 21 January 2026
Version of record: 16 February 2026
DOI: https://doi.org/10.1038/s41467-026-68371-x