Prediction of TKI response in EGFR-mutant lung cancer patients-derived organoids using malignant pleural effusion

Abstract

Patient-derived organoids (PDOs) are valuable in predicting response to cancer therapy. PDOs are ideal models for precision oncologists. However, their practical application in guiding timely clinical decisions remains challenging. This study focused on patients with advanced EGFR-mutated non-small cell lung cancer and employed a cancer organoid-based diagnosis reactivity prediction (CODRP)-based precision oncology platform to assess the efficacy of EGFR inhibitor treatments. CODRP was employed to evaluate EGFR-tyrosine kinase inhibitors (TKI) drug sensitivity. The results were compared to those obtained using area under the curve index. This study validated this index by testing lung cancer-derived organoids in 14 patients with lung cancer. The CODRP index-based drug sensitivity test reliably classified patient responses to EGFR-TKI treatment within a clinically suitable 10-day timeline, which aligned with clinical drug treatment responses. This approach is promising for predicting and analyzing the efficacy of anticancer, ultimately contributing to the development of a precision medicine platform.

Fig. 1. In the PE of lung cancer patients, various types of cells are present.

a A schematic representation of cell separation from pleural effusion, followed by organoid culture and drug screening process. b Identification of cell types present in the PE through FACS analysis.

Fig. 2. Establishing a precision medicine platform for treating cancer patients.

a Schematic representation of the method for isolating CD3-negative and CD3-positive cells from PE and their utilization in anticancer drug-sensitive testing. b Introduction of CODRP flat form for precision medicine. c Cells were loaded using the ASFA® spotter and cultured for indicated days. The Mean area of LCOs was quantified using the ASFA Ez Cell analyzer (MBD). *p < 0.05 compared with Day 0 or 1000 cells. Scale bars: 100 μm.

Fig. 3. Establishment of LCOs and drug testing platform by CODRP.

a Overview of the patient selection process. b Schematic workflow of CODRP platform using LCOs. c Representative brightfield images of LCOs after 7–10 days of culturing in LCOs growth media. H&E and IHC for the indicated antibodies of NSCLC patients PE samples or tissue specimens, as well as on the derived LCOs. Scale bars: 100 μm.

Fig. 4. LCOs-based high-throughput screening (HTS) and CODRP index analysis.

a A total of 5000 cells of LCOs, mixed with Matrigel at an 80% concentration, were spotted on the 384-micropillar surface. LCOs were exposed to Afatinib for 72 h. Subsequently, the mean area was determined by fluorescence intensity, and intracellular ATP levels were assessed. b The Z-score values for drug response were calculated based on different means and standard deviations. c Comparative analysis of drug response based on the AUC and CODRP indices for Afatinib; the CODRP index takes into account the LCOs growth rate, and it is calculated as a Z-score value based on the different mean and standard deviation (SD) values. Cut-off values for classifying drug responses into sensitive and resistant groups were identified for Afatinib. Cut-off: −0.17.

Fig. 5. Clinical relevance of LCOs-based HTS analysis and CODRP index analysis.

Patient #316 was diagnosed as a stage IVB lung cancer patient with an EGFR exon19 deletion. After treatment with Afatinib, a noticeable trend of size reduction was observed. Consequently, the drug sensitivity test using #316 LCOs derived from PE showed a sensitive response to Afatinib. Patient #334, diagnosed with stage IVA lung cancer and bearing an EGFR exon19 deletion, exhibited a significant reduction in lesions after Afatinib treatment. The drug sensitivity test confirmed a strong positive response to Afatinib.

Fig. 6. LCOs-based HTS and CODRP index analysis.

Comparative analysis of drug response based on the AUC and CODRP indices for Osimetinib; the CODRP index accounts for the LCOs growth rate and is calculated as a Z-score value based on the different mean and standard deviation (SD) values. Cut-off values for classifying drug responses into sensitive and resistant groups were identified for Osimertinib. Cut-off: −0.17.

Fig. 7. Clinical relevance of LCOs-based HTS analysis and CODRP index analysis.

Patients #246 and #278 showed disease progression accompanied by malignant PE during afatinib treatment. Upon disease recurrence, both tissue and malignant PE samples showed the presence of an EGFR exon 20 T790M mutation; consequently, Osimertinib or Lazertinib treatment was initiated. These patients have shown no signs of disease progression while receiving Osimertinib or Lazertinib. Clinically, their response to third-generation TKIs aligns with the findings of the LCO drug screening, suggesting a favorable outcome. Patient #282 experienced disease progression along with the development of malignant PE during gefitinib and afatinib treatment. An EGFR exon 20, T790M mutation was confirmed in PE, leading to the initiation of Lazertinib; however, progressive disease (PD) was observed. Clinically, this case was categorized as resistant to 3rd generation EGFR TKI. Additionally, resistance to the third-generation EGFR TKI, osimertinib, was also noted in LCO drug screening.

Fig. 8

The CODRP platform distinguishes between responders and non-responders to EGFR-TKIs.