2023 Cancer Innovation Scholars
Project Title: Porous Hydrogels as a Transducer Material in Microfluidic Electrochemical Cells
REU Scholar: Amina Anowara (Princeton University) Advisor: Sagnik Basuray, Mentor: Zhenglong Li
Abstract: Electrochemical microfluidic sensors would allow for rapid detection of various diseases in a cost-effective, resource-efficient manner. However, the current design of the microfluidic devices includes a carbon nanotube (CNT) transducer that is sandwiched between non-planar microelectrodes. Development of these devices requires the CNT to be packed by hand, which introduces human error into the production of the devices. This decreases the reliability of the results from these sensors, which limits the use of electrochemical sensors in fields where sensor reliability is critical. To improve the current microfluidic sensor design, an alternative transducer material is proposed. Use of hydrogels instead of CNT increases the degree of consistency with which each device is produced because it is possible to 3D-print the hydrogel into the device. To validate the use of hydrogels as a transducer, 2 criteria must be fulfilled: 1) it must have a porous microarchitecture that allows for the flow of biomolecules, and 2) it must be functionalizable. The methods used to develop the porous hydrogel and to functionalize the hydrogels to detect a desired target molecule are shown. The porous hydrogel was developed by experimentally determining the concentration of porogen (calcium carbonate particles) that needed to be added in order to achieve a highly permeable structure. Porosity was qualitatively assessed using scanning electron microscopy, and a mass of 1g of CaCO3 yielded desirable results. However, further work will be completed to achieve higher porosity by increasing the concentration of porogen. Ultraviolet radiation at a wavelength of 365 nm for 2 minutes, along with acid treatment using 2M hydrochloric acid to dissolve the porogen were used to arrive at the final transducer material. Incorporation of either metal-organic frameworks (MOFs) or CNTs into the prepolymerized hydrogel solution are the next steps to functionalizing the hydrogel. It is expected that electrochemical impedance spectroscopy will confirm the device's selectivity and sensitivity. In conclusion, 3D-printed hydrogels are a viable transducer material, as it can be packed in an automated manner and is able to be functionalized. By attaching a probe onto the functionalized hydrogel, many diseases - from COVID-19 to breast cancer - can be detected using this device.
Project Title: Use of Machine Learning Models to Predict Breast Cancer
REU Scholar: Edem Ammamoo (Alcorn State University), Advisor: Dr. Joshua Young, Mentors: Daniel Mottern and Mo Li
Abstract: The need for accurate and fast medical predictions to diagnose medical ailments is on the rise. Timely diagnosis of illnesses is proven to be important in the treatment of these diseases. Technology, and specifically, machine learning models, is effective at analyzing and predicting very good outcomes. This begs the question; can accurate deep learning models be built to help ensure the timely diagnoses of patients? To answer this question, machine learning models were built to diagnose breast cancer. Machine learning and deep learning models, such as convolutional neural networks (CNN), are programmed to detect certain features that can help diagnose malignant and benign tumors. The research is divided into two because of the data used. The first dataset is a numerical dataset obtained from previous Fine Needle Aspirate (FNA) research. This data is used to train and test models such as Logistic Regression, Random Forest and XGBoost Classifier. The aim of this aspect of the project is to perform Recursive Feature Elimination (RFE) to identify the important features that determine if a tumor is malignant or benign. The best model, Logistic Regression model, had a 98% accuracy and was used for RFE. The most important features were the worst measurements, specifically, radius, perimeter, and concave points. The effect of the different balancing techniques on the accuracy of the model is also tested. Different techniques affect the accuracy of the model prediction; the best technique for the Random Forest model was Borderline SMOTE 1 and SMOTE ENN was the best technique for the XGBoost Classifier model. The second part of this project uses a dataset of ultrasound images. The aim of this second project is to build a convolutional neural network (CNN) model that can accurately predict breast cancer. The expected accuracy of the model is about 80% accuracy.
Project Title: Electrospun PVDF Nanofibers for Early Cancer Detection via Acoustic Wave Sensing
REU Scholar: Elizabeth Hervias (NJIT), Advisor: Lin Dong, Mentor: Sun Kwong
Abstract: For 2023 alone, the American Cancer Society estimated over 600,000 projected cancer deaths and nearly two million new cancer cases. Early cancer detection is an important factor of successful treatment, helping cancer deaths enter a continuous decline of 32% since 1991 as of 2019. However, there are barriers to early diagnosis: accessibility to and affordability of resources. Conventional screening devices for cancer are bulky, expensive, and require trained professionals to administer the tests and interpret results. Even cell culture growth methods used to observe the rate of cell growth can take as long as a week in a clinical lab to show results for a single patient. Improvements and alternatives to these issues include enabling portability or utilizing implantable or wearable devices. Wearables such as biosensors are generally smaller, rapid, and cheap, and are therefore well suited to be adapted to an affordable alternative for early cancer detection. In this lab, we electrospun polyvinylidene fluoride (PVDF) nanofibers and encapsulated them in a biocompatible elastomer polydimethylsiloxane (PDMS) with carbon nanotubes (CNT) as electrodes to fabricate a flexible and wearable biosensor. The PVDF nanofibers serve as the functional layer to convert acoustic waves into electrical signals for measurement. By utilizing the piezoelectric effect, the flexible and wearable biosensor can characterize and quantify the carcinogenic particles located underneath the epidermis as the acoustic wave vibrations travel through the cells and change by specific mass densities, therefore altering frequency responses. Two types of fiber arrangements have been fabricated and tested: highly aligned nanofibers and randomly aligned nanofibers. Voltage characterization and acoustic wave testing were also achieved to evaluate the biosensor's electrical performance and sensitivity. PVDF has yet to be used as a flexible and wearable acoustic wave sensor for early cancer detection applications, and its highly desirable properties such as biocompatibility, high flexibility, and great processability make such devices capable of transforming point of care for cancer patients and increasing accessibility to cancer detection technology at low cost.
Project Title: Targeted Drug Delivery: Investigating Protein Corona Behavior
REU Scholar: Kaylie Greenis (Washington State University), Advisor: Kathleen McEnnis, Mentor: Atharva Markale
Abstract: Conventional chemotherapy is often nonspecific and toxic to both targeted and non-targeted cells. To overcome this limitation, targeted drug delivery through nanoparticles is studied to reduce toxicity and improve selectivity. When a nanoparticle enters the bloodstream, a protein corona coats the nanoparticle. The protein corona can determine the nanoparticle's fate with cells in a way that could be harmful and affect its targeting abilities. The objective of this project was to investigate the impact of the protein corona on nanoparticle behavior by using a nanoparticle tracking analysis (NTA) system to monitor the behavior of polystyrene particles (PS) in plasma. The experimental conditions included various ratios of saline mixed with bovine plasma with Alsever's solution, bovine plasma with sodium citrate, goat plasma with Alsever's solution, and goat plasma with sodium citrate. The particle size was measured under these conditions using 1) different syringe pump speeds (0, 10, 20, and 30) and 2) over a duration of 24 hours. Two separate experiments were conducted: (1) assessing the effect of syringe pump speed on nanoparticle size, targeting a speed at which particles would travel across the screen within approximately 10 seconds, followed by a reduction in speed, and (2) conducting a 24-hour study with samples of each plasma type under physiological conditions, while recording particle size measurements after specific intervals of time. Our results uncovered that (1) higher syringe pump speeds generally result in a slight reduction in particle size, and (2) particle size reflects the Vroman effect of proteins with varying ratios of plasma types and saline. With this information, we can infer that the protein corona comprises of diverse proteins. Ongoing and future research will focus on the characterization of the protein corona using techniques such as ultraviolet-visible spectroscopy, Fourier transform infrared spectroscopy, and protein assays.
Project Title: Manufacturing a State-of-the-Art Selector Valve for a Miniature Peptide Synthesizer
REU Scholar: Maryom Rahman (NJIT), Advisors: Sagnik Basuray, Vivek Kumar, Mentors: Yu Hsuan Cheng, Alexandra Griffith
Abstract: Recent innovations in cancer research show that using peptide therapeutics can mitigate a variety of ailments. However, peptide synthesis is costly and produces large amounts of hazardous waste. Utilizing the core functions of a peptide synthesizer, a miniature model can be innovated to produce smaller peptides with less cost and waste. Additionally, a miniature peptide synthesizer would have point-of-use capabilities, increasing the efficiency and readiness to treat patients. Synthesis of peptides relies on five main steps: protection, deprotection, coupling, cleaving, and peptide formation. Synthesizers complete the first three steps. The amino acids are selected and attached to a solid polymer (usually a resin). From there, the amino acids are deprotected (unbonded) and then coupled (bonded) with the next amino acid in the sequence. Afterward, the chain is cleaved from the resin and undergoes dialysis, and then lyophilization. In the proposed novel design, the main functions of a peptide synthesizer can be scaled down to a miniature model to create smaller-scale peptides. The miniature peptide synthesizer consists of 4 main parts: the outer shell, two multi-selector valves, the reactor, and an electrical chamber. Innovations in the selector valves were made to improve upon this synthesizer design. Previous designs for the selector valve in the miniature peptide reactor have design issues, leading to leaks and poor flow control. In this iteration, a new state-of-the-art selector valve will be fabricated utilizing Fusion 360 computer-aided design (CAD) software and stereolithography (SLA) 3D printing. This design utilizes a ball piston and a spring to open and close access to the outlet chamber (Figure 1). A simple version of the selector valve has been implemented with one inlet and one outlet and is currently being tested for leaks over various solvents. A 3D model of the multi-selector valve, with 8 inlets and one outlet is currently being designed and implemented.
Project Title: Manipulation of Burst Pressure within FRESH vascularization
REU Scholar: Resty Mercado (Rowan University), Advisor: Amir Miri, Mentor: Swaprakash Yogeshwaran
Abstract: Tumors create negative pressure gradients within vascularized tissue, leading to adverse effects such as low concentrations of oxygen, acidity changes and low perfusion in the blood. The goal of this project was to recreate the tumor microenvironment using hydrogels and explore burst pressure, which is the maximum pressure before a vessel ruptures due to the negative pressure gradients. To achieve this, a multifaceted burst pressure design was created. Vascularization within the hydrogels was achieved through the Freeform reverse embedding of Soft hydrogels (FRESH) technique and the subtractive needle method. The gel bath consisted of 4.5% Gelatin, while 5% Alginate served as the protrusion material. Calcium Chloride was used as the crosslinker, and needle gauges ranging from 14G to 27G were employed. The burst design approach involved conducting a consumer's needs analysis and literature review to fabricate several designs, aiming to find the most optimal approach. Notable implementations included the use of Piezoelectric film, which exhibits transduction properties between mechanical and electrical inputs, enabling it to function as both a sensor based on vessel bulge strain. Additionally, a 3D-printed channel adapter was fabricated, and a camera setup was used. A syringe pump was employed for controlled fluid delivery. Burst pressure was measured in mmHg using an Integrated Development Environment (IDE) for data acquisition and control. The results consist of a max burst pressure, a burst pressure range based on the data we find, a max bulge pressure and a max bulge strain.
Project Title: Exposure Guidelines for Dermal Diffusion of Chemical Warfare Agents
REU Scholar: Ricardo Inoa (NJIT), Advisor: Laurent Simon
Chemical and biological warfare agents have been commonly used to promote health hazards on account of their mass destruction capabilities. With the new threats of skin penetrating agents, further assessments to scrutinize long-term effects became imperative. Researchers developed skin permeation models to determine the level of chemical exposure required to penetrate the dermal barrier and diffuse into the bloodstream. These evaluations help trace chemical effects on the body and provide an estimated timeframe for public health agencies to respond to a threat. Furthermore, previously found data on Acute Exposure Guidelines Limits (AEGL) will serve as a tool to measure the impact of chemical warfare agents absorbed through the respiratory tract. The AEGL values will be adapted to simulate skin permeation data and determine Permissible Exposure Limits (PELs) through the stratum corneum. The PEL values will be estimated using Fick's first law of diffusion and databases such as CompTox and Pubchem.
Project Title: Effects of EMFs on PME and T47D Cells
REU Scholar: Ricardo Otake (Rowan University), Advisor: Nellone Reid, Mentor: Luis Medina
Abstract: Cancer is a leading cause of death, with breast cancer specifically resulting in the deaths of more than 600,000 people as of the year 2023. Electromagnetic Fields (EMFs) show potential to be a normalized noninvasive treatment, as they have anti-inflammatory properties which correlates to the inflammation caused by cancer. The objectives of this project is to 1) observe effects of EMF exposure to healthy cells (primary mammalian epithelial [PME] are the healthy cells used) and cancer cells (T47D which was the type of breast cancer cells used) using a phase contrast microscope; and 2) to conduct a live cell count using Trypan Blue and a hemocytometer. A Helmholtz coil was used to produce an EMF that was exposed to the cells for 6 to 12 hours by running a current of 0.7 amps to produce approximately 2 to 10 mT of EMF intensity. By testing EMF exposure we expect to observe 1) apoptosis of T47D cells and minimal detriment to PME cells; and 2) a decreasing quantity for live cell count during EMF exposure.
Project Title: Integrated Electronics and Pulsed Electric Generators to Mimic Tumor Cell Response to Electrical Stimulations
REU Scholar: Riya Patel (New York University), Advisor: Amir Miri, Mentor: Aydasadat Pourmostafa
Abstract: Malignant tumors, such as certain types of cancer, present challenges in treatment due to their resistance to conventional therapies and complexity of surgical removal. Pulsed electric fields (PEF) are gaining prominence as an effective approach overcoming the limitations of traditional treatments for cancer, including standard chemotherapies. Our research proposes incorporation of electrodes into a bioreactor for studying the tumor cell response to electrical stimulation. This design will enhance monitoring of electrical signals within the bioreactor system and facilitate the development of innovative therapeutic strategies, such as electrotherapy, that can selectively target and inhibit tumor growth while minimizing damage to healthy tissues. Our focus consists of using electrodes in the bioreactor to deliver PEF to the cells. Experimental design began with using computer-aided design to create molds through stereolithography 3D printing. The molds are filled with polydimethylsiloxane (PDMS), a versatile biomaterial in which cells can be seeded. Stainless steel electrodes are integrated in the PDMS prior to curing, to ensure stable delivery of an electrical pulse. The PDMS bioreactor is surface engineered, using plasma and collagen treatment, to gain adherent properties for cell and glass slide attachment. The electrodes are connected to a function generator to deliver an electrical stimulus to induce effects in the cells, which will be studied and analyzed. Our studies consist of using rat muscle cells and later cancer cells. For muscle cells, we expect to see twitches in response to stimulation while cancer cells should result in eventual cell death, implying a decrease in the size of a tumor. The overall goal of this research is to increase productivity and efficacy in studying the effects of electrotherapy.
Project Title: Project Title: Utilizing Anticancer Peptides to Combat Triple Negative Breast Cancer
REU Scholar: Sofia Ruiz (Lehigh University), Advisor: Vivek Kumar, Mentor: Joseph Dodd-o
Abstract: Breast cancer encompasses about 25% of all cancer diagnoses in women and leads to the deaths of 370,000 women annually. Triple Negative Breast Cancer (TNBC) is a breast cancer subtype that is characterized by its lack of three receptors: estrogen, progesterone, and HER2. This aggressive cancer does not respond to hormone therapy or therapies targeting these receptors, making chemotherapy the primary option. Unfortunately, more than half of women with TNBC relapse or develop resistance to chemotherapy. Peptides have recently emerged as promising therapeutic potential due to their ability to selectively target cancer cells without compromising the other healthy cells surrounding them, their biocompatibility, and their ease of production. Anticancer peptides have promising therapeutic potential due to their ability to induce cell cancer death through various mechanisms. The objective of this research project consists of synthesizing four anticancer peptides: SLAPOP2, SLAPOP3, SLKr5, and K6. First, we will characterize the peptides. We will conduct Mass Spectroscopy to determine the molecular weight of each peptide and High-Performance Liquid Chromatography to test their purity. To evaluate the secondary structure, we will use Fourier-Transform Infrared Spectroscopy and Circular Dichroism. Second, we will analyze the cytocompatibility of the peptides using a fibroblast cell culture. We will use a Live/Dead Assay as well as Cell Counting Kit 8 to determine the viability and test the cells' proliferation in the presence of the peptides. Lastly, we will analyze the cytotoxicity of the peptides on a breast cancer cell line through the TUNNEL Assay.