Introduction

Tumor biology studies of pancreatic ductal adenocarcinoma (PDAC) used surgical specimens, even though most PDACs are unresectable.

Aims & Methods

The aims of the study were to 1) establish PDAC organoids platform from endoscopic ultrasonography (EUS)-guided fine needle biopsy (FNB), 2) investigate clinicopathologic and genomic characterization, and 3) find out the clinical applicability of the patient-derived PDAC organoids model from PDAC diagnosis to treatment timeline The patients with suspected PDACs were enrolled prospectively, and PDAC minimal specimens were acquired from EUS-guided FNB at Samsung Medical Center.

Results

Ninety-four PDAC organoids were successfully established (94/113, 83.2%) via EUS-guided FNB within 8.2±2.6 days. Clinical factors were also analyzed associated with the success rate in establishing PDAC organoids; specimens with high cellularity (>10 clusters) were an independent factor in multivariate analysis. Whole exome sequencing showed the high concordance between primary PDAC and its organoid and more enriched genetic alterations in organoids. In addition, the high-throughput screening (HTS) drug test revealed that the clinical correlation with real-world data was 12 out of 15 between PDAC organoids and the chemotherapeutic response of the study patients. Whole-transcriptome sequencing identified genes and pathways associated with the drug responses.

Conclusion

We have achieved the PDAC organoid platform in a high success rate via EUS-guided FNB, and the concordance between primary PDAC and its organoid seems very high. More importantly, the HTS drug test showed that PDAC organoids may well reflect clinical chemotherapeutic response and may also be used as the “Patient Avatar Model” in clinical practice on time.

Introduction

Recent developments in diagnostic imaging have increased the detection rate of lymphoma, for which a clear, histopathological diagnosis is critical for the subsequent choice of therapeutic regimen. Over the past decade, endoscopic ultrasound-guided fine needle aspiration/biopsy (EUS-FNA/B) has become widely-used means of tissue collection. However the open lymph node biopsy has conventionally been used to diagnose lymphoma histologically, and the evidence backing the use of EUS-FNA/B for this purpose is still limited.

Aims & Methods

The present study aimed to compare the performance of EUS-FNA/B and the open biopsy for diagnosing lymphoma by retrospectively analyzing 1113 patients with 1224 lesions who underwent either of the two methods at the study center between 2012 and 2024. The patients’ background, pathological diagnosis, procedural factors, and complications were evaluated. The performance of the open biopsy was compared with that of EUS-FNAB in diagnosing lymphoma in accordance with the 5^th edition of the World Health Organization classification of hematolymphoid tumours ¹⁾ (WHO classification).

Results

An open biopsy was performed for 943 lesions, 621 of which were malignant lymphomas, with the face, head, neck, axilla or groin, limb, and other areas being the site of 46, 556, 96, 225, 4, and 16 lesions, respectively. EUS-FNAB was performed for 281 lesions, 82 of which were malignant lymphomas, with the intra-abdominal area, mediastinum, and rectum being the site of 270, 10, and 1 lesion, respectively. The needle gauge used for FNA was 19G, 22G, and 25G in 1, 143, and 36 lesions, respectively, while and the needle gauge for FNB was 20G, 22G, and 25G in 3, 74, and 3 lesions, respectively. The mean number of punctures was 2.3 (range 1-9). EUS-FNAB involved significantly fewer, procedural complications than the open biopsy (14/943 vs 0/281; p=0.04). According to the WHO classification, the diagnostic performance (95% confidence interval) of the open biopsy has sensitivity 93.4 % (91.1-95.2), specificity 100.0 % (98.3-100.0), positive predictive value 100.0 % (99.0-100.0), and negative predictive value 88.7 % (85.0-91.8). The diagnostic performance (95% confidence interval) of EUS-FNAB according to the same classification has sensitivity 79.3 % (68.9-87.4), specificity 100.0 % (97.2-100.0), positive predictive value 100 % (91.8-100.0), and negative predictive value 92.1 % (87.7-95.3 %). Between the groups, there was no significant difference in the proportion of lesions for which immunostaining was able to be performed (615/621 vs 80/82; p = 0.24). The positive diagnosis rate was significantly higher for the open biopsy (580/621 vs 27/40; p < 0.001) but did not differ significantly for the EUS-FNB (580/621 vs 38/42; p = 0.52). Univariate and multivariate analyses found that needle type (FNB needle) was a procedural factor in EUS (needle type, gauge, number of punctures, stroke length) contributing to the diagnosis of lymphoma (odds ratio: 3.83; p = 0.037).

Conclusion

EUS was less complicated than the open biopsy as means of collecting tissue for histologically diagnosing lymphoma, and EUS-FNB had a positive diagnosis rate comparable to that of the open biopsy.

References

1. The 5th edition of the World Health Organization Classification of Haematolymphoid Tumours: Myeloid and Histiocytic/Dendritic Neoplasms. Leukemia. 2022 Jul;36(7):1703-1719.