DishTrial4OncoEfficacy™

Anticancer Efficacy Validation Platform

Currently available for Hepatocellular Carcinoma (HCC) and Pancreatic Ductal Adenocarcinoma (PDAC).

STEP 1

STEP 2

STEP 3

STEP 4

Design Your Patient Cohort

Build a representative cohort tailored to your therapeutic program

Define the patient population you want to evaluate according to your scientific and clinical objectives.

Your cohort can be customized based on:

Number of patients
Sex and age distribution
Clinical characteristics
Disease-specific characteristics
Other relevant population criteria

This enables efficacy to be evaluated across biologically diverse patient profiles, rather than against a single standardized model.

Define the
TICK Threshold

Separate non-specific cell killing from true anticancer activity

Before evaluating therapeutic efficacy, we determine the compound’s Target-Independent Cell Killing (TICK) profile using a cohort of healthy individualized spheroids.

This critical step identifies the concentration range at which cell killing becomes non-specific and could otherwise be misinterpreted as anticancer efficacy.

By establishing this reference first, PredictCan helps prevent overestimation of a drug candidate’s therapeutic potential.

Evaluate
Patient-Specific Efficacy

Measure true therapeutic activity across patient diversity

Using our hSES™ platform, individualized tumor models are exposed to your therapeutic candidate under standardized experimental conditions.

Patient responses are interpreted relative to the previously established TICK profile, allowing us to distinguish true tumor-specific activity from non-specific cytotoxicity.

This Patient Trial-in-a-Dish approach reveals differences in therapeutic response across individuals and enables the identification of responders and non-responders.

Obtain
Decision-Ready Efficacy Data

Turn individual responses into meaningful efficacy insights

Receive a comprehensive dataset characterizing both overall and patient-specific therapeutic responses.

Depending on your study design, deliverables may include:

Dose-response curves
TICK profile and threshold
Individual patient-specific responses
Responder vs. non-responder profiles
Therapeutic window assessment
Raw experimental data

The result is a more stringent assessment of whether your drug truly works, and for whom.

Turning Scientific Evidence into Better Efficacy Prediction

DishTrial4OncoEfficacy™ builds on years of research and experimental validation of PredictCan’s individual-centric approach to oncology. Our scientific studies demonstrate the ability of our technology to capture interpatient variability in drug response, identify responder and non-responder profiles, and distinguish true therapeutic efficacy from target-independent cell killing (TICK).

By combining patient-specific tumor models with TICK-based efficacy assessment, our platform provides a more clinically relevant evaluation of anticancer activity, helping you identify meaningful therapeutic responses, define effective exposure ranges, and de-risk efficacy decisions before clinical development.

Interindividual variability in CYP induction and activity

Capturing Patient-Specific Molecular Heterogeneity in Hepatocellular Carcinoma (HCC).

Our HCC hSES™ model captures the molecular heterogeneity associated with disease progression and cancer stage.

Using serum from patients across BCLC-A, BCLC-B, and BCLC-C stages, we generated personalized phenotypic tumor spheroids and assessed their ability to reproduce clinically relevant molecular differences.

Analysis of 13 HCC-associated hub genes revealed distinct expression profiles across patient-derived models and disease stages. Importantly, our models recapitulated molecular signatures associated with BCLC staging previously described by Xu and colleagues, further supporting their clinical relevance. These genes are also associated with overall survival in HCC.

By reproducing both patient-specific heterogeneity and stage-associated molecular phenotypes, our platform provides a more clinically relevant representation of HCC and offers a personalized framework for evaluating anticancer therapies across diverse tumor phenotypes.

Cherradi et al., 2025

Controlled growth and long-term stability of liver spheroids

Modeling the Molecular Heterogeneity within HCC Patients.

An advanced HCC model should capture not only differences in disease stage, but also patient-specific molecular features that can influence therapeutic response.

We characterized key epigenetic regulators - including writers, readers, and erasers - in our HCC hSES™ model, revealing distinct patient-dependent epigenetic profiles. Importantly, reproducing this epigenetic heterogeneity is highly relevant for therapeutic evaluation, as epigenetic dysregulation can influence tumor phenotype, treatment sensitivity, and resistance.

By recapitulating patient-specific epigenetic landscapes, our platform provides a more clinically relevant model to assess therapeutic efficacy and identify treatment responses across molecularly diverse HCC patients.

Cherradi et al., 2025

Reproducibility of the PredictCan liver model

HCC hSES™ Model Predicts Patient-Specific Clinical Treatment Response.

A major challenge in HCC is the substantial interpatient variability in therapeutic response, which limits the ability of population-level models to predict how an individual patient will respond to treatment.

Using serum collected from HCC patients before treatment initiation, we generated patient-specific HCC hSES™ models and exposed them in vitro to the same tyrosine kinase inhibitors (TKIs) subsequently administered to the corresponding patients. This patient-matched strategy enabled us to evaluate therapeutic response in vitro using the biological characteristics of each individual patient.

Importantly, in vitro drug-response profiles showed strong concordance with clinical outcomes, with personalized models distinguishing responders from non-responders. These findings demonstrate that the HCC hSES™ model can capture clinically relevant determinants of therapeutic sensitivity and reproduce patient-specific differences in treatment response.

By directly linking a patient's biological profile to their subsequent clinical outcome, our platform moves drug testing beyond conventional population-based models toward personalized prediction of treatment efficacy. This creates a functional bridge between patient-derived biology and real-world therapeutic response, with potential to support treatment selection and precision oncology in HCC.

Cherradi et al., 2025

Reproducing the Tumor Microenvironment of Pacreatic Ductal Adenocarcinoma (PDAC).

Our PDAC hSES™ model goes beyond tumor cells to recapitulate key features of the extracellular matrix (ECM) characteristic of the PDAC tumor microenvironment.

Histological characterization demonstrated preserved tissue organization and collagen-rich ECM deposition. Sirius Red analysis under polarized light further revealed predominantly type I collagen, a major component of the dense desmoplastic stroma that characterizes PDAC.

Importantly, reproducing this collagen-rich ECM is highly relevant for therapeutic evaluation. The desmoplastic matrix can profoundly influence tumor architecture, drug penetration and distribution, and ultimately treatment sensitivity or resistance. Incorporating this critical component therefore provides a more physiologically relevant context for assessing anticancer efficacy than tumor-cell models alone.

By recapitulating the dense, collagen-rich ECM of PDAC, our platform offers a clinically relevant system to investigate how the tumor microenvironment contributes to therapeutic response and to better evaluate drug efficacy in a physiologically relevant tissue context.

Cherradi et al., 2026

Patient-Specific Chemograms for Preclinical Anticancer Efficacy Assessment.

The PDAC hSES™ model enables the generation of patient-specific chemograms, providing a functional framework to compare anticancer efficacy across clinically relevant and investigational therapies.

While clinical guidelines define evidence-based treatment options for PDAC, they cannot fully capture the biological heterogeneity underlying individual treatment responses. Using serum from patients with metastatic PDAC, we generated patient-specific models and evaluated multiple clinically relevant regimens, including gemcitabine, Gem-Pac, and FOLFIRINOX. Treatment responses were assessed using our TICK exclusion strategy to distinguish specific therapeutic efficacy from non-specific cell killing.

Importantly, testing multiple therapies within the same patient-derived model enables a direct, patient-matched comparison of therapeutic efficacy, reducing the biological variability that can confound conventional preclinical comparisons. The resulting chemograms reveal differential drug sensitivity across patients and provide a functional basis for identifying which compounds are most effective in which patient populations.

For drug developers, this approach can support earlier candidate prioritization, more informed go/no-go decisions, and patient-population selection, helping focus development efforts on the therapies and populations with the greatest likelihood of clinical benefit.

Cherradi et al., 2026