Date: 24th August 2021
Cancer is the second leading cause of death globally, and with up to 40% of patients being treated with ineffective drugs this represents a huge challenge for the field. In part, it is thought that current drug screening models for cancer are inadequate as they lack the complex relationships such as tumour-stroma interactions. Now, researchers have created microengineered perfusable 3D-bioprinted glioblastoma (GB) models, that mimic the in vivo tumour microenvironment and can be used for personalised therapy screening and drug development.
The conventional preclinical drug development process relies on in vitro evaluation of the drug toxicity and efficacy usually in 2D cell culture, which is then followed by in vivo animal studies. Whilst, 2D models are inexpensive, rapid and easy to use, they do not support tissue-specific architecture. In vivo models face the challenge that human stroma is replaced by host equivalents over time, and the time to results is slow, time which these patients often don’t have.
Now, researchers at Tel Aviv University, Israel, led by Ronit Satchi-Fainaro have recreated tumour heterogenic microenvironment by creating fibrin glioblastoma bioink, containing patient-derived glioblastoma cells, astrocytes and microglia, and perfusable blood vessels coated with brain pericytes and endothelial cells. The resulting 3D bioprinted tumour represented a clinically relevant platform for rapid drug screening and prediction of treatment outcome.
The team started 3D engineering the tissue constructs using two bioinks, a tumour bioink and a vascular bioink. The tumour bioink was composed of natural polymers fibrinogen and gelatin, and contained samples from patients, taken directly from operating rooms at the Neurosurgery department of Tel Aviv Sourasky Medical Center. Cells of the microenvironment were also included with the cancerous cells, such as astrocytes and microglia. The vascular bioink was composed of a thermo-reversible biocompatible synthetic polymer. Here, the gelation process was reversible meaning this sacrificial material was used to embedded the vasculature inside the 3D model and was then liquefied to create the lumen of the vessels. Next, endothelial cells and pericytes were allowed to homogenously attach to the lumen walls. These vessels were connected to a microfluid system which facilitated the delivery of substances such as blood cells and drug to the tumour. The replicas were printed in a bioreactor designed by the lab, and they used a hydrogel sampled and reproduced from the extracellular matric of the patient, again adding to the authenticity of the model.
The team initially tested the system by creating two types of models, using dormant and fast-growing GB cells. The 3D GM bioprinted models both resembled the in vivo tumour settings far better than 2D models, where they lost their growth characteristics and differences. Treatment with a P-selectin inhibitor – P-selectin has a role in GB progression – inhibited cell proliferation of GM tumours in vivo and also resulted in a substantial reduction in GB cell proliferation in the fibrin 3D-bioink compared to a 2D culture, showing the bioprinted models responded in a similar fashion to the in vivo tumours.
To explore the differences further the team conducted genetic sequencing of the cancer cells grown in the 3D-bioprinted model, and compared them to both cancer cells grown on 2D plastic and cancer cells taken from patients. The found higher similarities in the transcriptional profile between in vivo and 3D models compared to 2D models. Through time, the 2D cancer cells had changed considerably, losing resemblance to original samples.
Finally, the team tested the clinical relevance of the bioprinted GM models, using TMZ an alkylating agent used as a standard-of-care chemotherapy for patients with GB. They evaluated the response of 3 patient-derived cells to TMZ either 3D printed or grown in 2D culture. Here, they found the IC50 values in 2D culture were almost identical, but in the bioprinted tumours they found that each cell type exhibited a different IC50 value when grown. The corresponding patients also responded differently to the same TMZ therapy, and survived for different period of time thus, the bioprinted models far better represented patient outcomes, and emphasised the need for a personalised approach.
The team here have developed a 3D-bioprinted tumour model based on biocompatible polymers containing both tumour and stromal cells, and supported by perfusable blood-like vessels. Using GB as a proof-of-concept, they have demonstrated the bioprinted models recapitulate more faithfully in vivo tumours showing similar growth curves, drug responses, and genetic signatures.
This powerful platform should allow for rapid, reproducible, and robust target discovery, accelerating personalised therapy screening and drug development.
As the mechanical properties of the 3D bioink can be controlled, the platform could be easily modified to emulate other tissues, and other cancer types. To show evidence for the robustness of their system, the team also established a metastatic breast cancer model, creating a vascularised 3D model of breast cancer brain metastasis. Other cancers, such as melanoma also frequently metastasise to the brain.
This ground breaking platform could be the basis for potentially replacing cell cultures and animal models, accelerating the development of new drugs, and enabling must faster predictions of best treatment for patients.
It has the potential to be an invaluable tool for drug discovery for aggressive, difficult-to-treat cancer. With an unprecedented plethora of next generation tools now becoming available to target and deliver cancer fighting drugs and therapies, aggressive cancers are starting to become far more treatable. We have seen tools developed for the treatment of the GB, such as synthetic nanoparticles and immunocytokines. Researchers have discovered and ‘off’ switch for aggressive breast cancer, and more recently a new class of drug target for aggressive cancer has been found.
With a typical current time from drug development, to reaching approval and commercialisation of over a decade, and average costs of ~$2.6 billion (US), unfortunately the majority of anticancer compounds fail due to lack of efficacy during Phase I trials. Bioprinted tumour models such as these could be a game changer, mimicking better the clinical scenario, enabling optimal investigation, and bringing more efficient drug to the clinic in much shorter time frames.
For more information please see the press release at the Israel Ministry of Foreign Affairs
Neufeld, L., Yeini, E., Reisman, N., Shtilerman, Y., Ben-Shushan, D., Pozzi, S., Madi, A., Tiram, G., Eldar-Boock, A., Ferber, S., et al. (2021). Microengineered perfusable 3D-bioprinted glioblastoma model for in vivo mimicry of tumor microenvironment. Science Advances 7, eabi9119.