By David N. Leff

Almost half a century before Sherlock Holmes launched his crime-cracking career in 1887, a fictional French sleuth, Auguste Dupin by name, gave birth to the modern detective story. His creator was Edgar Allan Poe, who in 1845 wrote "The Purloined Letter."

To find this internationally earth-shaking document, police ransacked a stately ambassadorial residence from top to bottom and inside out. When they came up dry, detective Dupin discovered the lost letter — in plain view on a mantelpiece. His deductive reasoning: The best concealment is no concealment.

A quarter-century later, in 1869, an Australian physician, T. W. Ashworth, described cells he had discerned in the blood of a patient who had died of cancer. They resembled cells found in the tumor at autopsy.

This finding prefigures a simple blood test for cancer, described in the current Proceedings of the National Academy of Sciences (PNAS), dated April 14, 1998. Its title: "Detection and characterization of carcinoma cells in the blood." The paper's senior author, oncological immunologist Jonathan Uhr, pointed out that 80 percent of all malignant neoplasms are carcinomas. Uhr is a professor of microbiology at the University of Texas Southwestern Medical Center, in Dallas.

These solid tumors arise from epithelial cells, the most common constituents of skin, but also comprising the bulk of tissues in the breast, colon, lung and prostate.

When any of these sites turns malignant, physicians ransack the suspected organ to discover and diagnose the cancerous tumors. Mammograms seek out breast cancers; colonoscopy goes after tumors of the large intestine; elevated prostate-specific antigen (PSA) may signal malignancy in that gland.

Biopsy of the hidden but now suspect tumor, by needle or knife, often follows up on these preliminary diagnoses. Meanwhile, the malignant epithelial cells, like Poe's purloined letter, are hanging out in virtual plain view: They populate the patient's blood, just a simple syringe away from diagnostic sampling.

A Path-Lab-Friendly Diagnostic

Updating that 1869 discovery of tumor cells in the blood, oncologists today are finding evidence that primary tumors are constantly shedding cancerous cells into the circulation, and probably fairly early in the life of the tumor.

"Cancer is a group of growing cells that break down basement membrane," Uhr told BioWorld Today. "It always seemed incomprehensible to me that one could keep all the cells within the primary tumor, once it had, let's say, a million cells. They would have to be spilling out into the lymph and blood. And as it takes between 100 million and a billion cells to detect a primary tumor," he went on, "one would have predicted that there would be shedding."

But Uhr made the added point, "Whether that shedding could be detected or not was another matter."

He envisioned a simple blood test to harvest those shed epithelial carcinoma cells, then count and analyze them, as an eventual substitute for diagnostic biopsies. The system that he and his co-authors conceived, constructed and confirmed consists of three components: A ferromagnetic step that removes the tumor cells from most of the other elements in the blood; flow cytometry to physically separate and tag the shed cells; and microscopy to verify their malignant contours.

"The immunomagnetic procedure enriches for the epithelial cells in the blood 10,000-fold," Uhr said. "It depends on using submicroscopic particles which are made by the Immunicon Corp., in Huntingdon Valley, Pa.

"Those iron particles," he explained, "don't even start out as particles, but as a colloidal suspension, almost a solution, which coalesces into tiny spherules, maybe 100 nanometers in diameter."

The Dallas co-authors coated these magnetic microspheres with an antibody that detects the cancerous epithelial cells. After separation in a magnetic field, these then transited a flow cytometer, one cell at a time, at which a laser peered. Whenever its light rays hit the tumor-specific antibody, the machine sorted those critical cells from the rest of the blood content.

"We then confirmed under the microscope that these were in fact neoplastic cells from the blood of patients bearing cancer," Uhr recounted, "by looking at their morphology and particular molecules associated with tumor cells."

Cell Count Tracks Cancer Progression

"This cellular assay," he added, "was able to detect one epithelial cell or less in one milliliter of blood — which makes it more sensitive than PCR [polymerase chain reaction] analysis."

The team drew blood from 30 women with carcinoma of the breast, three men with prostate cancer, and from 13 controls.

In 14 breast-cancer patients with primary tumors confined to their initial organ sites, the test detected an average of 16 cells per 20-milliliter blood sample. But in 11 others with distant metastases, the cell count averaged about 130. In control subjects, the number of cells ranged from zero to five.

Moreover, the level of those carcinoma cells went down in concert with chemotherapy that induced remissions. Thus, in one breast cancer patient on high-dose chemotherapy who experienced a partial remission, the epithelial cells disappeared, then rose to previous levels when a relapse occurred.

In recent weeks, Uhr has repeated these experiments on three more prostate cancer patients, with similar results, and on an additional 15 to 20 breast cancer patients. Results in this cohort are still coded to ensure against bias.

He is planning "right now, to test every woman who comes into the Dallas medical center, with a lump in a breast or a shadow on a mammogram, and every man with a high PSA or other suspect prostate condition.

"All of these tests will be blinded," Uhr pointed out, "and they will determine definitively where there's a cut-off point at which we won't get a false positive reading. Then we can say: 'This patient has cancer.'"

He continued: "This is the only way we will convince a properly conservative oncology physician to change the way he thinks and does things, and possibly substitute this blood test for the conventional biopsy."

Uhr gives himself another year or so in which to optimize the procedure by eliminating the need for confirmation under the microscope and by adding channels to the flow cytometer that will recognize the breast-cancer-specific mucin glycoprotein. He also aims to add colon cancer and melanoma to the panel of blood-testable carcinomas.

"We're going to work very hard over the next year or so," Uhr concluded, "to continue pursuing this blood test with ferocity." *

No Comments