ADVERTISEMENT
Precision oncology without biomarkers: Assessing drug sensitivity in patient-derived tumoroids to guide mCRC 3rd line therapy
Background
“Precision Oncology” refers to strategies and tools to find the best treatment for a specific patient. This concept is often reduced to the combination of targeted therapies with molecular biomarkers. Nevertheless, most options indicated in guidelines for 3rd line treatment of mCRC patients do not have associated biomarkers. In vitro assessment of tumor drug sensitivity based on patient-derived 3D tumoroids can be used to guide therapy decision-making. Furthermore, it can inform the “off label” use of drugs susceptible of improving patient’s PFS. We have checked the feasibility of such approach.
Methods
Core needle biopsies from mCRC patients having failed at least two treatment lines were collected and shipped fresh to 2cureX labs. Following the IndiTreat® [2cureX, Copenhagen] protocol, the biopsies were mechanically disrupted, and the resulting fragments were cultured to form 3D tumoroids. These tumoroids were added to pre-loaded gel arrays containing the drugs of interest (FOLFOX, FOLFIRI, FOLFOXIRI, regorafenib, trifluridine + tipiracil, mitomycin C + 5FU, gemcitabine + 5FU, temozolomide + irinotecan). Tumoroid growth was assessed after seven days by capturing the images of day 0 and day 7 and comparing them to those of untreated tumoroids, used as negative controls. A proprietary convolutional neural network (IndiNet) translated the images into a growth inhibition (G.I.) score that was used to assign each tumor to a category of sensitivity.
Results
The G.I. results obtained with each of the treatments covered a wide range and were assessed for normality by four different methods: Anderson-Darling, Shapiro-Wilk, Kolmogorov-Smirnov and D’Agostino – Pearson. The results allow us to assume a normal distribution, and define four categories based on the mean and standard deviation of each of these cohorts. Categories have been labelled as “L” (low sensitivity: growth inhibition below mean – 1SD), “ML” (mid-low sensitivity: growth inhibition between mean and mean – 1SD), “MH” (mid-high sensitivity: growth inhibition between mean and mean + 1SD) and “H” (high sensitivity: growth inhibition above mean + 1SD). The percentage of cases in each category was (Low / Mid-Low / Mid-High / High): FOLFOX : 19% / 35% / 19%/ 27% FOLFIRI: 17% / 31% / 28% / 24% FOLFOXIRI: 14% / 39% / 25% / 21% Regorafenib: 15% / 33% / 37% / 15% trifluridine / tipiracil: 17% / 24% / 41% / 17% mitomycin C + 5FU: 25% / 13% / 50% / 13% gemcitabine + 5FU: 17% / 40% / 27% / 17% temozolomide + Irinotecan: 5% / 57% / 14% / 24%.
Conclusions
IndiTreat® provides an individual drug sensitivity profile that clearly differentiated between “low sensitivity” and “high sensitivity” tumoroids for each drug in the panel. This information can be used by oncologists to determine if any of the treatments indicated in guidelines can be an option for that patient, or if she might benefit from the off-label use of certain drugs, thus expanding their treatment options and achieving the benefits of personalization even with standard chemotherapy regimens.
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosures
J. Thastrup: Shareholder / Stockholder / Stock options: 2cureX; Full / Part-time employment: 2cureX. O. Thastrup: Officer / Board of Directors: 2cureX. G. Hagel: Shareholder / Stockholder / Stock options: 2curex; Full / Part-time employment: 2curex. H. Harling: Officer / Board of Directors: 2curex.
