Esophageal Cancer Organoids Predict Treatment Response and Recurrence Risk Before Therapy

A study found that organoids, or “mini-organs,” cultured from tumor tissue of patients with esophageal cancer can be used to predict the effectiveness of preoperative chemoradiotherapy (neoadjuvant chemoradiotherapy) and the risk of recurrence before treatment begins.

According to the study published in the latest issue of the medical journal iScience, the multiparameter model developed by the researchers recorded 87.5% sensitivity and 100% specificity in predicting pathologic response to neoadjuvant chemoradiotherapy. For recurrence prediction, sensitivity was 83.3% and specificity was 81.8%.

The findings were based on organoids cultured and analyzed from diagnostic biopsy tissue from 18 patients with locally advanced esophageal squamous cell carcinoma.

The researchers used an in-house radiation exposure platform to deliver fractionated radiation to the organoids, then measured the area under the growth inhibition curve according to radiation dose and the organoid growth rate. They combined these measures with the patients’ clinical stage and postsurgical pathologic regression grade to create a multiparameter model called CODRP.

For pathologic response prediction, a single measure had limitations. Organoid growth rate performed best, with 87.5% sensitivity and 90.0% specificity, but predictive performance declined substantially when the area under the growth inhibition curve or clinical stage was used alone. In contrast, the CODRP pathologic response index, which combined all three measures, had an area under the curve of 0.9625, with 87.5% sensitivity and 100% specificity.

A specificity of 100% means that none of the patients classified as “resistant” by the model achieved a pathologic complete response. In other words, the finding suggests that patients who are unlikely to benefit from treatment may be identified before therapy begins.

The CODRP model also showed meaningful results in predicting recurrence. When patients were divided according to the CODRP recurrence index, the 2-year recurrence-free survival rate was 71.4% in the sensitive group and 10.0% in the resistant group, a difference of approximately sixfold.

By contrast, pathologic response grade, which is emphasized in clinical practice, did not statistically significantly distinguish recurrence risk in this study population. The 2-year recurrence-free survival rate was 50.0% among patients with a favorable pathologic response and 22.2% among those with an unfavorable response, with a P value of 0.056.

The researchers also highlighted the limitations of relying on a single measure by comparing two patient cases. One patient achieved a clinical complete response after chemoradiotherapy and did not experience recurrence; all CODRP indices classified the patient as “sensitive.” The other patient achieved only a partial response and experienced recurrence 230 days after surgery, leading to a “resistant” classification. The researchers explained that this resistant patient could have been classified as “sensitive” based solely on the area under the growth inhibition curve, with a Z score of -1.019, but the addition of growth rate and clinical stage accurately captured the patient’s actual status.

The researchers said the findings demonstrate the possibility of distinguishing sensitive and resistant groups within approximately 2 weeks before treatment begins. For the sensitive group, nonsurgical management or a reduced radiation dose could be considered, while for the resistant group, adjustment of the timing of surgery or intensification of adjuvant therapy could be considered.

However, the researchers added that the findings were from a feasibility study. They explained that directly testing drug or radiation sensitivity using a patient’s own tumor tissue provides more comprehensive and functional information than conventional tissue or genetic testing, bringing treatment closer to a personalized approach.