A paired comparison found medium-dependent responses to 33 compounds in three pancreatic and colorectal tumor organoid models. Switching medium partly reversed sensitivity, but the study did not test prediction of patient outcomes.
Researchers at the University of Antwerp and Antwerp University Hospital compared conventional Wnt/R-spondin/noggin medium with OncoPro (OP) medium across tumor material from 36 patients. The study examined establishment, molecular identity and cell state before testing drug responses in a smaller pancreatic and colorectal cancer subset.
The design separated transfer of established cultures from growth directly out of fresh tissue. Fifteen of 18 existing Wnt-grown lines adapted to OP. Among 18 fresh specimens, long-term establishment succeeded in six with Wnt medium and two with OP; none succeeded exclusively in OP. Larger early organoids in some OP cultures did not translate into better long-term establishment. Two non-small-cell lung cancer cultures showed nonmalignant epithelial overgrowth in both media and were excluded from subsequent comparisons.
Transfer history also mattered. Direct replacement with OP arrested growth in some lines, prompting a gradual transition through 25%, 50%, 75% and 100% OP across passages. This restored expansion in several cultures, although some still failed at full OP. Adaptation was thus an observed process rather than an assumed property of an established line.
The reported transcriptomic comparison included 11 matched organoid lines. Samples clustered mainly by patient of origin, and exome analysis in nine models generally retained major tumor-associated alterations across media. Wnt-grown cultures nevertheless showed stronger proliferation and stemness-related expression, whereas OP cultures showed adhesion-associated and inflammatory programs, including changes related to TGF-β signaling.
Comparisons with public pancreatic and colorectal single-cell atlases placed OP-associated signatures more prominently in malignant epithelial compartments. That provides a reference for the transcriptional states represented in culture. It does not establish which medium better predicts treatment response in patients.
An exploratory cytokine analysis identified total TGF-β and IL-15 in the OP supplement before contact with organoids. Active TGF-β emerged during OP culture. The conditioned-medium experiment used one colorectal model and was not normalized to organoid cellularity, so it cannot establish a causal explanation for the drug-response shifts.
Five paired cultures were initially selected for testing, but two OP-grown lines expanded too poorly for the screen. The final experiment tested 33 compounds in three pancreatic or colorectal models, using four concentrations and imaging over five days. Responses were expressed through normalized organoid growth rate and a normalized area-over-the-curve measure, with controls matched to each medium. These measures distinguish growth inhibition from killing relative to the assay controls.
For the normalized growth-rate metric, zero denotes complete growth inhibition and −1 the killing level of the positive control. The normalized curve-area measure ranges from zero to one, with larger values indicating a stronger effect. Medium-matched controls help interpret responses when the culture condition itself changes baseline growth.
Within this subset, Wnt-grown organoids were generally more sensitive. Differences were prominent for the MAPK-pathway inhibitors ASTX029 and trametinib and the inhibitor-of-apoptosis protein antagonist birinapant. Taxanes, 5-fluorouracil and SN-38 showed smaller medium effects. Cisplatin and oxaliplatin produced largely flat dose-response curves over the tested range, indicating limited activity in both conditions. Sensitivity predictions derived from cell-line transcriptomic databases did not fully match the measured organoid responses.
The team also switched one pancreatic model from Wnt to OP and back to Wnt, and transferred an independently established colorectal OP culture into Wnt medium. Sensitivity increased for some drugs after transfer, but the response profiles did not simply converge on the original Wnt state. Individual responses could resemble that reference, exceed it or change little. These findings support a partly reversible, medium-dependent state while leaving room for smaller selection effects.
Two independent Wnt-to-OP experiments in the same pancreatic model, conducted after cryopreservation and at different passages, yielded highly concordant response profiles (Spearman ρ = 0.92). This supports repeatability within that model and transfer workflow.
The comparison shows why medium belongs in the interpretation of an organoid drug screen. Tumor identity, successful expansion and a reproducible response each provide different information. The pharmacology sample was small, cultures selected for downstream testing had maintained adequate growth, and the proprietary OP formulation limited mechanistic analysis. No matched patient treatment outcomes were assessed. The study therefore supports controlled, paired evaluation of culture conditions without identifying a clinically superior medium.
A study connects aloperine binding and VPS34 lactylation to disrupted endosomal trafficking in lung cancer cells. Patient-derived organoids provided additional evidence of a stronger response to aloperine plus cisplatin than to either agent alone.
Single-cell analysis identified conserved epithelial programs associated with survival and resistance. Patient-derived organoid experiments then shifted platinum responses through SERCA2 activation or CCDC6 inhibition.
