In breast cancer, there are few measurable disease monitors to helpguide diagnosis, treatment and prognosis. But within the last twodecades, a single mutant protein has stood out as signaling the extentof many mammary carcinomas, and the outlook for their clinicaloutcome.

This signpost is protein 53 (p53), expressed by a gene, p53, on theshort arm of human chromosome 17. Mutations of that gene'ssequence express altered protein products. These turn up in mosthuman malignancies, especially in breast cancers _ which strikeover 180,000 women a year in the U.S. alone.

Occurrence of these abnormal proteins enables clinical oncologists totailor their treatment regimens more cogently. Current wisdom holdsthat the more p53 mutants pile up in a breast cancer cell nucleus, theworse the prognosis.

Current clinical practice in finding these ominous proteinaccumulations relies on immunohistochemistry (IHC) _ testingtumor tissue with monoclonal antibodies. This method is cheap, fast_ and frequently wrong. IHC misses many p53 mutants, and often"sees" some that aren't there.

Last week's Journal of the National Cancer Institute, reports the firstclinical trial of a new p53 test based on sequencing the cDNA's totalcoding region. Its title: "The p53 gene in breast cancer: prognosticvalue of complementary DNA sequencing vs.immunohistochemistry."

Pharmacia Biotech AB, of Uppsala, Sweden, developed thisproprietary technique, which it calls Sequence-Based Diagnosis(SBD). Oncologists and epidemiologists at the University of Uppsalameasured the efficiency of SBD against that of IHC in the tumors of316 Swedish women, operated on for breast cancer from January1987 through December 1989.

Of the 316 tumor samples tested, the journal paper reported, SBDdetected p53 gene mutations in 69, (22 percent). IHC found "elevatedlevels of p53 protein in 64 tumors (20 percent), suggesting thepresence of mutations."

This apparent similarity is misleading, explained oncological surgeonand clinical epidemiologist Lars Holmberg, a co-author of the report.

"Yes," he told BioWorld Today, "about the same frequency ofmutations in the trial was indicated by IHC. But it looks as if theantibodies pick up some as mutations that we, with full sequencing,do not. On the other hand," Holmberg continued, "we pick up someas mutated that IHC does not."

Specifically, the paper reported that SBD detected 23 tumors withp53 mutations that IHC missed, "suggesting that IHC failed to detect33 percent of the [69] mutations." The antibody technique, the reportadded, also signaled 19 mutations found negative by the gene-sequence technique, a false-positive rate of 30 percent.

Holmberg explained: "There are some specific types of stop codons[gene mutations] that you will not detect anyway withimmunohistochemistry. It takes sequencing of the whole gene todetect them.

"Our method has a higher specificity and sensitivity," he said. Thispoint touches the purpose of p53 detection altogether; namely,clinical decision-making in the treatment of post-surgery, primarybreast-cancer patients.

"If we could detect the patients with the p53 mutations in a betterway, maybe we could also relate p53 mutations in a better way toprognosis, or prediction of response to therapy. And it seems thatwith this technique we have a better way to discriminate amongpatients with bad or better prognosis."

Pharmacia's SBD detected only 69 mutated p53 gene mutations in309 testable tumor tissue samples. Asked what about the other 240,Holmberg replied, "That's a good question. No one knows."

He added: "It's not that we think a mutation in p53 is a necessarycomponent of getting a cancer. There are other combinations ofmutations that can make cells behave in a malignant way."

This year Pharmacia Biotech is further testing its breast-cancerdetection approach in Europe and the U.S.

In San Juan Capistrano, Calif., the Corning Nichols Institute, aleading pathology reference laboratory, is now engaged in acollaborative study with the Swedish company, to evaluate theclinical relevance of Pharmacia's sequence-based technology.

New Tests Tooling Up in U.S., Italy

The Institute's chief science officer, Delbert Fisher, told BioWorldToday: "We have an ongoing arrangement with the University ofTexas at San Antonio, where we maintain a breast-cancer tissue bank.We provide the samples that come through, many thousands a year."

Corning Nichols' research and development center, Fisher added, is"now pulling some of those samples, and looking again atimmunochemistry analysis of p53, looking at protein, and sequencingthe p53 gene in those patients."

He continued: "We're now in the process of correlating thatinformation with prognosis of the clinical course in those women, todetermine whether p53 gene sequencing is more efficacious as aprognostic marker than the p53 IHC."

If and when Pharmacia's SBD system proves efficacious, Fisherallowed, his Institute intends to offer it as a routine test. "It will be aquestion of price, of course," he observed. "We can run about 20 aday with one technician." As for what that price might be, he said,"I'd better not guess." But Fisher noted that "Generally speaking,molecular testing now in a reference laboratory runs in the $100- to$200-a-sample range. We've never cranked up an on-line runningsequencing test."

This year too, the Swedish company is beginning a joint project withPharmacia-Upjohn's oncology center in Milan, Italy. Thisretrospective study will gauge "the predictive value of the SBDtechnique in breast cancer," the company's diagnostics marketingmanager, Carina Schmidt, told BioWorld Today. "The center willundertake at least two projects in 1996," she said, "looking in fresh-frozen tumor specimens at p53 mutation status in relation to specifictreatments." n

-- David N. Leff Science Editor

(c) 1997 American Health Consultants. All rights reserved.