National Cancer Institute NCI Cancer Bulletin: A Trusted Source for Cancer Research News
July 10, 2012 • Volume 9 / Number 14

Cancer Research Highlights

Test May Reduce the Need for Surgery to Diagnose Thyroid Cancer

A new test may spare some patients with suspicious thyroid nodules from diagnostic surgery. Researchers analyzed thyroid nodule samples collected via fine-needle aspiration (FNA) for the expression of a panel of 167 genes and found that the test accurately identified whether nodules were cancerous. Their results were published June 25 in the New England Journal of Medicine.

About 15 to 30 percent of patients undergoing FNA for suspicious thyroid nodules have indeterminate findings on standard cytology tests—that is, the tests show cellular changes that indicate a possible cancer but the findings are inconclusive. Although the majority of those with inconclusive cytology results have a benign condition, most have thyroid surgery to determine whether cancer is present.

Several researchers said the test has the potential to change clinical practice by eliminating or delaying the need for such surgery in some patients.

The researchers collected more than 4,800 aspirate samples from nearly 3,800 patients treated at 49 academic centers and community hospitals over a 19-month period. Of these, they analyzed indeterminate FNA samples from 265 nodules for which surgical samples were also available. The FNA samples were analyzed using the 167-gene panel that the researchers had developed based on earlier research.

Overall, when the results from the gene expression test were compared with diagnostic results from thyroid samples removed during surgery, the test correctly identified 92 percent of the malignant samples and 93 percent of the benign samples. But about half of the samples that the gene expression test identified as suspicious—not clearly malignant or clearly benign—were actually benign on surgical analysis.

For patients with indeterminate cytology results, the gene expression test "can be useful in making important [patient] management decisions, such as recommending watchful waiting in lieu of diagnostic surgery," wrote lead author Dr. Erik Alexander of Brigham and Women's Hospital and his colleagues.

Dr. Ann Gramza of NCI's Center for Cancer Research agreed. But she cautioned that "a negative result should not dismiss a patient from further follow-up surveillance of the nodule."

"The risk…is that 5 to 10 percent of nodules classified as benign...are likely to be malignant [false negatives], particularly those that are cytologically indeterminate but suggestive of cancer," wrote Dr. J. Larry Jameson of the University of Pennsylvania in an accompanying editorial. In such patients, he explained, "it might be reasonable" to do another FNA biopsy or perform a diagnostic surgical procedure.

One recent study, Dr. Jameson noted, suggested that the reduction in surgeries that could result from its use—about 25,000 fewer operations per year—"could result in substantial cost savings," even with the added cost of the test.

Skin Cancers Traced to Previously Unknown Effect of UV Radiation

The harmful effects of the sun on the outer layer of skin (the epidermis) are well documented. But ultraviolet (UV) radiation from the sun also may alter cells in the underlying layer of skin (the dermis), setting the stage for the development of cancer in the epidermis, according to findings published June 8 in Cell.

In the study, researchers observed epidermal changes in mice that were similar to those seen in UV-induced human premalignant skin lesions called actinic keratosis, which can progress to squamous cell carcinoma, the most common skin cancer in humans. In the mice, the Notch signaling pathway was lost in stromal cells, which contribute to the dermal compartment of the skin.

The loss of Notch signaling appeared to be sufficient for tumors to emerge in the overlying epidermis, the researchers observed. The increased inflammation that accompanied the loss of Notch signaling also may have played a role in tumor development, they added.

"This study says that changes in the stroma are as important as changes in the epidermis, and we probably need to pay attention to them," said lead investigator Dr. G. Paolo Dotto of Massachusetts General Hospital and the University of Lausanne, Switzerland.

To investigate the clinical relevance of the mouse findings, the researchers analyzed tissue from human patients with actinic keratosis. They found that Notch signaling was reduced in human stromal cells near precancerous lesions. Moreover, similar molecular changes were induced by UVA radiation, which is an environmental cause of skin cancer, the study authors noted.

The findings may provide insights into the phenomenon of field cancerization, in which a patch or field of cells, rather than a single initiating cell, changes when exposed to a carcinogen and has the potential to become premalignant, Dr. Dotto pointed out.

In an accompanying editorial, Drs. Sakari Vanharanta and Joan Massagué of Memorial Sloan-Kettering Cancer Center praised the study for raising the possibility that, in addition to causing mutations, UV radiation may lead to tumor-promoting changes in dermal cells.

The findings add to the extensive list of harmful effects of excessive sun exposure. "This should give [people] one more reason to cover up," wrote Drs. Vanharanta and Massagué.

Experimental Drug Based on Plant Toxin Targets and Kills Tumor Cells

Researchers have engineered a drug that can deliver a potent cell-killing toxin to tumors while largely sparing normal tissues. The drug, known as G202, shrank xenograft tumors of several human cancers in mice, including prostate, breast, kidney, and bladder cancer, and had relatively few toxic effects. On the basis of these findings, reported June 27 in Science Translational Medicine, researchers have initiated an early-phase clinical trial of G202 in patients with advanced cancer.

G202 delivers its toxic payload—a potent analog of the plant substance thapsigargin—to tumors by binding specifically to a protein known as prostate-specific membrane antigen (PSMA). PSMA is found in high levels in most prostate cancers. It is also found in tumor endothelial cells, which line blood vessels in a variety of solid tumors but not in normal endothelial cells. PSMA is an enzyme that spans the cell membrane and can cut proteins in specific places.

Researchers designed G202 so that it not only binds to PSMA but is also a target for PSMA's protein-cutting activity. G202 is an inactive "prodrug," and PSMA causes the release of the active, cell-killing thapsigargin analog from the prodrug form. This release takes place outside the cell, in the tumor microenvironment. Once released, the thapsigargin analog is taken up by nearby tumor cells, where it inhibits a protein known as the SERCA pump. Shutting down the SERCA pump floods the cell with calcium and triggers programmed cell death.

Unlike commonly used chemotherapy drugs, which typically work by killing rapidly dividing cells, "[thapsigargin] and its analogs can kill both rapidly proliferating and nonproliferating cells with equal potency," wrote Drs. Samuel Denmeade and John Isaacs of Johns Hopkins University and their colleagues. This ability, they noted, makes a thapsigargin-based drug particularly suitable for treating prostate cancer because most cancer cells in metastatic prostate cancer are not dividing.

Safety studies of G202 that are required before the drug can be tested in humans showed that it caused transient reversible kidney toxicity in rats and monkeys. G202 did not cause bone marrow toxicity—a common side effect of traditional cell-killing chemotherapy—in mice, rats, or monkeys.

Also in the Journals: Cells Most Vulnerable to HPV Identified

Researchers have identified a distinct population of cervical cells that are targeted by cancer-causing types of human papillomavirus (HPV), according to a study published June 26 in the Proceedings of the National Academy of Sciences.

These cells sit at the junction between the interior and exterior epithelial surfaces of the cervix and have a unique appearance and gene expression profile. The researchers showed that specific biomarkers of these junction cells were expressed in HPV-associated precancers and cancers. Forced expression of proteins coded for HPV (HPV oncoproteins) in other cell types did not cause the expression of junction cell biomarkers. And researchers did not find junction cell biomarkers in HPV-associated vaginal, vulvar, or penile precancers.

By examining tissue from women who had surgery to remove cervical tissue followed several months later by a hysterectomy, the researchers found that cells expressing the junction cell markers did not return once they were removed. These findings raise the possibility that removing junction cells may prevent cervical cancer, although further research is needed to understand how these cells change during a woman's life.