Introduction

Esophageal squamous cell carcinoma (ESCC) has high incidence and mortality rate. Neoadjuvant immunotherapy (nIT) is promising because of its therapeutic efficacy. However, there are still only limited studies on preoperative immune checkpoint inhibitor in combination with chemotherapy followed by surgery for the locally advanced ESCC. Some patients will not respond to nT but will be exposed to side effects. Functional magnetic resonance imaging (fMRI) techniques can reflect tumors biological and microstructural features, so they can quantify therapy-induced changes before anatomic variation in tumor size. Diffusion weighted imaging (DWI) is a functional imaging technique that is widely used in therapeutic efficacy evaluation and prediction. This study was conducted to explore the value of DWI-derived parameters for the prediction of pathologic response to nIT combined with chemotherapy in locally advanced resectable ESCC.

Methods

Thirty-two patients with locally advanced ESCC who were treated with nIT combined with chemotherapy followed by surgery were prospectively enrolled from May 2022 to May 2023. DWI was performed within 1 week (median 2 days) before nT, and 2 to 5 weeks (median 25 days) after completion of nT, prior to surgery. The tumor size was depicted by three-dimensional position. Parameters including apparent diffusion coefficient (ADC), true diffusion coefficient (D), pseudo diffusion coefficient (D*), and pseudo diffusion fraction (f) before and after nT were measured. Pathologic response was evaluated according to tumor regression grade (TRG) system. We grouped patients with TRG 0-2 into responders, while TRG 3 into non-responders; we also grouped patients with TRG 0 into complete response (CR), while TRG 1-2 into minimal residual cancer cell (MRC). The changes in tumor size measurement and DWI-derived values before and after nT in different TRG groups were assessed. Receiver operating characteristic (ROC) curves analysis was used to determine the best cutoff value for predicting the pathologic response or MRC to nT.

Results

Twenty-one patients were identified as TRG 0-2 (responders), and eleven as TRG 3 (non-responders) in pathologic evaluation. While seven patients were identified as TRG 0 (CR), and fourteen as TRG 1-2 (MRC) for minimal residual disease detection. The ADC, D, and f values increased significantly after nT. The Δtumor size, post-ADC, post-D ΔADC and ΔD values of responders were significantly higher than those of non-responders. The Δtumor size, ΔADC and ΔD values of patients achieved CR were significantly higher than those with MRC. The area under curve (AUC) values of the post-ADC, post-D, ΔADC and ΔD for prediction of pathologic response to nT were 0.801-0.857, 0.688-0.758, 0.879-0.913 and 0.805-0.861. The AUC values of ΔADC and ΔD for MRC identification after nT were 0.776-0.796 and 0.786-0.867.

Discusion

Our study investigated the prediction capability for pathologic response to neoadjuvant immunotherapy combined with chemotherapy in locally advanced resectable ESCC by DWI sequences and achieved acceptable results.
All 32 patients received Nivolumab combined with standard chemotherapy. Twenty-one (65.625%) patients achieved TRG 0-2, who were divided into responders, while 11 (34.375%) were divided into non-responders (TRG 3). There were also 7 (21.875%) patients achieved CR (TRG 0), compared to 14 (43.75%) patients who remained MRC (TRG 1-2). Another study aimed to explore the value of IVIM-DWI for the prediction of pathologic response to nCT in locally advanced ESCC¹. We aimed to increase the predictable sensitivity, which meant more accurate to identify people who are responders for nT. The post-ADC, post-D, ΔADC and ΔD values had favorable predictable performance in identifying treatment response in locally advanced resectable ESCC with sensitivity and specificity of 90.5%-95.2% and 54.5%-63.6, 71.4%-81.0% and 54.5%, 95.2% and 81.8%-90.6%, 90.5%-95.2% and 54.4%-63.6%, respectively. The diagnostic performance of ΔADC value (AUC = 0.879-0.913) in our study is superior to ΔD value (AUC = 0.859) reported in prior study, while that of ΔD value was like our study (AUC = 0.805-0.861)¹. There were also other studies investigating the utility of DWI-MRI in predicting nT response of EC concluded that changes in ADC were highly predictive of histopathological response with high specificity^2,3. ΔADC and ΔD in patients without tumor residual (CR/ TRG 0) were significantly higher than those in patients with MRC after nT (TRG 1 and 2). The diagnostic performance of ΔADC and ΔD for identification MRC after nT were favorable (AUC = 0.776-0.796 and 0.786-0.867) with sensitivity and specificity of 85.7% and 50%-57.1%, 85.7% and 57.1%-64.3%, respectively. To our knowledge, there have been no previous studies detecting the MRC in ESCC patients after nT.

Conclusion

DWI may be used as an effective functional imaging technique to predict pathologic response and detect minimal residual disease to nIT combined with chemotherapy in locally advanced resectable ESCC

Acknowledgements

None