Starr Lab In The News

Naturally kill “stressed cells”, i.e. virally infected, transformed cells, MHC low, identify target cells through binding membrane bound receptors or antibodies (antibody dependent cellular cytotoxicity). Target cells are killed by membrane disruption or engagement of death receptors. However, there are major shortcomings of NK cells:

1. Persistence (< 3 weeks in mice and humans),

2. In vivo expansion,

3. Intransigence to genome editing

4. Immunosuppression by TME,

5. Require IL15 for survival

The Starr Lab's approach and previous work overcome many of these shortcomings. Their current aims with these NK cells are to:

1. Determine the functional effects of CISH KO combined with feeder cell activation/expansion on NK function, cytokine response and persistencem,

2. Assess NK targeting and activation potential of MSLN-CAR alone or in combination with CISH KO

3. Perform a clinical trial to test the hypothesis that removal of inhibitory signaling via CISH KO, enhancing targeting via MSLN-CAR and improved activation via feeder cells will produce NK cells that can be safely administered and are capable of controlling advanced EOC.

Pancreatic cancer, a particularly deadly cancer, grows so rapidly that it outpaces growth of blood vessels and subsequently becomes starved for oxygen. HIF1A is a gene that is turned on when cells are starved for oxygen. The Bagchi lab performed experiments in mice and cell lines showing that loss of the HIF1A gene, somewhat surprisingly, caused an increase in the aggressiveness of pancreatic cancer. In these models of pancreatic cancer they identified an unexpected role for the HIF1A protein, which is to inhibit a second protein (PPP1R1B) that functions to destabilize a universal tumor suppressor (p53). The study concluded that pancreatic cancers are more aggressive when they lose expression of HIF1A because the loss of HIF1A leads to an increase in PP1R1B, which then destabilizes p53, resulting in a more aggressive tumor. The Starr lab contributed to this study by assisting with the analysis that applied these findings to human tumors to show that patients with pancreatic cancer have a worse prognosis when PPP1R1B levels rise and HIF1A levels decrease. These findings were published in the journal Gastroenterology in August of 2020.

The APOBEC3 family of proteins normally function to control viruses attempting to infect cells. The APOBEC3 proteins are capable of mutating viral DNA in a very precise manner to block the ability of the virus to replicate. Dr. Rueben Harris is a leading expert in the field and has studied this family of proteins for decades. The Harris lab demonstrated that, in addition to its anti-viral function, the protein is also responsible for creating many of the mutations found in cancers. It was unknown, however, if these mutations were driving the cancer or were merely a byproduct. The Starr lab assisted the Harris lab with developing a mouse model of GI tract cancer that over-expressed one of the APOBEC3 proteins in the intestines. This model clearly demonstrated that APOBEC3 contributes to increased aggressiveness of the tumors.

These findings were published in the Journal of Experimental Medicine in September of 2020.

Understanding the molecular drivers of ovarian cancer could help in the discovery of new treatment modalities. Many groups have proposed that there are ~ 4 molecular subtypes of ovarian cancer. This hypothesis is based on a comparison of gene expression patterns found in ovarian cancer samples taken from many women. However, the ability to assign ovarian cancer patients into one of these four groups has not translated into improvements in therapy. Using sophisticated single cell sequencing techniques to explore gene expression patterns within individual cells, along with an analysis of copy number changes in ovarian cancer samples, the Waldron and Starr labs propose a new way of thinking about molecular subtypes. Instead of ovarian cancers being static, the molecular subtypes might represent stages in the evolution of the cancer. This could explain why assigning a molecular subtype based on a single snapshot of gene expression taken from the patient at time of surgery might not be directly useful for changing the therapeutic approach. The image on the left shows a 3-D graphic of 4,000 cells from a single patient.

These findings were published in Cancer Research in August 2020.

Dr. Starr and Dr. Prahba teamed up to test a novel cell therapy in an ovarian cancer mouse model. The novel cell therapy was developed in the Prabha lab and consists of mesenchymal stem cells engineered to grab onto nanoparticles loaded with chemotherapy. Using advanced mouse models of ovarian cancer developed in the Starr lab, the team tested this novel therapy for its ability to control ovarian cancer. They found that simultaneous treatment with the mesenchymal stem cells and the chemotherapy-loaded nanoparticles could prolong survival in a patient-derived xenograft mouse model. The treatment works because the mesenchymal stem cells naturally traffic to the tumor. When the nanoparticles loaded with chemotherapy pass through the tumor in blood vessels, they attach to the mesenchymal cells and then release their payload, killing the cancer cells. The hope is that this therapy can create a high concentration of chemotherapy right where it is needed in the tumor, and will be better than traditional system administration of chemotherapy.

These findings were published in the journal Cancer in April of 2020.

Dr. Starr received funding from the Randy Shaver Cancer Research and Community Fund on his grant titled "Gene-edited natural killer cells for ovarian cancer therapy." The Goal of this grant is to develop novel natural killer (NK) cell therapy to treat ovarian cancer by targeting NK cells to ovarian tumors using NK specific chimeric antigen receptors and combining CAR-NK with loss of PD1 to enhance NK cell cytotoxicity.

New technologies have given us the unprecedented ability to analyze gene expression at the level of the single cell. Dr. Starr co-leads a group at the University of Minnesota that is using these new technologies to understand how ovarian cancer develops and becomes resistant to chemotherapy. The data produced by these new technologies, however, can be overwhelming and hard to interpret. Dr. Starr has teamed up with Dr. Jinhua Wang's group, which specializes in high level bioinformatic analyses, to develop a new method that can identify what cell types are present in an ovarian cancer based on single cell gene expression data. The method they developed, ccFindR, combines nonnegative matrix factorization, which takes advantage of the sparse and nonnegative nature of single-cell RNA count data, with Bayesian model comparison enabling de novo prediction of the depth of heterogeneity. Using this new approach, the team is analyzing ovarian cancer samples from women who graciously have agreed to participate in this Ovarian Cancer Precision Medicine Initiative.

The method was published in the journal Life Science Alliance in July 2020.

The Starr lab staffed a booth at the Minnesota Science Museum on Saturday as part of the Cancer and the Human Body exhibit designed by the Masonic Cancer Center (https://www.smm.org/cancer-and-human-body ) Their booth provided fun educational activities for anyone visiting their table to test their cancer knowledge and learn more about cancer research.