Analysis finds new differences between kidney cancers

9 minutes ago

Above, renal cell carcinoma under microscope. Single-cell sequencing can now differentiate subtypes of cells, improving treatment options for kidney cancer. (Shutterstock)

Renal cell carcinoma, the most common form of kidney cancer, manifests and reacts to treatment differently based on the subtype of kidney cell that is cancerous. According to an unprecedented analysis published Monday, these subtypes, their behavior and their differences could facilitate a more informed and precise development of treatments.

University of Michigan researchers analyzed the gene expression of tens of thousands of tumorous and healthy kidney cells to create a "cell atlas" of data, which was used to identify differences between healthy and cancerous genes and track down which subtype of kidney cell started each cancerous tumor.

Among the findings, published Monday in Proceedings of the National Academy of Sciences, was that most tumors of a common form of renal cell carcinoma originated in a rare kind of proximal tubular cell — part of the nephron, the structure that filters blood and produces urine and numbers at least 1 million in each kidney.

Kidney cancer is the eighth most diagnosed cancer in the U.S., and about 76,000 new cases and nearly 14,000 deaths are expected in 2021. It is a category that includes many different kinds of cancer and is dominated by renal cell carcinoma. 

There are also many kinds of renal cell carcinoma, and how a given cancer develops and responds to treatment depends on the subtype of kidney cell from which it originates. But technology has only recently allowed for exploration of these differences with the development of single-cell sequencing, said Saravana Dhanasekaran, a senior author of the study and a pathology researcher at the University of Michigan.

Single-cell sequencing, which has "fueled many important discoveries" in biomedical research, allows for detailed analysis of samples at the scale of the individual cell, rather than the more blunt results given from tissue samples.

To Dhanasekaran, tissue sampling is like a "smoothie," in how the intricate differences between cells are blended together in aggregate measurements, while single-cell sampling allows them to be distinguished much like the ingredients in a "salad."

"That was really a great opportunity for us to … give attention each cell type deserves, and then know what they have to offer to us," he said.

In the study, Dhanasekaran and his co-authors employed single-cell RNA sequencing to determine which genes a given cell expresses. The process provided great detail about the behavior of each cell subtype, allowing for the creation of a gene-expression profile of each subtype, which could then be used to determine the type of cell each tumor began as.

"Single-cell RNA sequencing was key to allowing us to monitor gene-expression patterns in each individual cell, revealing the mechanisms at play within the tumor microenvironment that can predict overall survival," said cancer biologist Arul Chinnaiyan, another senior author and a professor of pathology and urology at the University of Michigan.

Using 14 samples from nine patients, the researchers analyzed the gene expression of nearly 30,000, including both benign and tumorous kidney cells. They identified 12 subtypes of healthy kidney cells, some well recognized and others poorly understood, and analyzed the cells and their interactions present in the tumor microenvironment in clear cell renal cell carcinoma, the most common kind of kidney cancer.

The benign and the tumorous cells were compared to identify which subtype each cancer originated from, as well as the differences between the two categories, to determine what gene-expression changes were associated with cancerous growth.

"We can systematically go after these differences and see which are the new targets that can be therapeutically developed," Dhanasekaran said.

The researchers' predictions based on the results were then tested on a large preexisting database of RNA-sequencing data.

One key finding was that clear cell renal cell carcinoma, which makes up about 80% of all renal cell carinoma cases, originated from a poorly understood subtype of kidney cell. For all seven patients in the study with clear cell renal cell carcinoma, the tumors had begun from rare proximal tubule cells. The cells, named PT-B by the researchers, had their gene expression characterized for the first time.

Treatments were also found to vary in effectiveness, depending on the tumors' contents. Patients with more immune cells in their tumor generally responded better to immunotherapy, while patients with more endothelial cells — which line blood vessels — did better with treatments that inhibited the growth of new blood vessels.

Dhanasekaran said the findings can improve mouse models for kidney cancer, which would be a "huge advantage" in studying the creation and progression of tumors as well as in testing drug treatments. Understanding how cancer develops from each cell subtype would allow researchers to more accurately recreate specific kinds of kidney cancer in mice, he said, rather than bluntly blocking several tumor-suppressing genes and creating a cancer that may not resemble a "natural" tumor. 

"Having an accurate mouse model that truly reflects the human disease will get us that much closer to matching therapy with a particular cancer type," Dhanasekaran said.

According to Dhanasekaran, the vast amount of data generated by single-cell RNA sequencing will be a continuing resource for future hypotheses and testing for kidney cancer. Other single-cell sequencing methods could also be used to complement the atlas, the pathology researcher said.

"It's a totally exciting stage right now," Dhanasekaran said.

The study, "Single-cell analyses of renal cell cancers reveal insights into tumor microenvironment, cell of origin, and therapy response," published June 7 in PNAS, was authored by Yuping Zhang, Sathiya P. Narayanan, Rahul Mannan, Gregory Raskind, Xiaoming Wang, Pankaj Vats, Fengyun Su, Noshad Hosseini, Xuhong Cao, Chandan Kumar-Sinha, Stephanie J. Ellison, Thomas J. Giordano, Todd M. Morgan, Sethuramasundaram Pitchiaya, Ajjai Alva, Rohit Mehra, Marcin Cieslik, Saravana M. Dhanasekaran, and Arul M. Chinnaiyan, University of Michigan.

We use cookies to improve your experience on our site and to show you relevant advertising.