For decades, epidemiologists and oncologists have pointed out that alarge majority of lung cancer victims were inveterate cigarettesmokers. They and anti-smoking activists used this statistical fact as asmoking gun to support their contention that cigarettes cause cancer.

The tobacco companies have countered their claim by arguing _successfully in court cases to date _ that statistical correlation is notproof.

Now, a team of molecular biologists report, in effect, that they havefound the bullet that caused the statistical smoking gun. Indeed, theyfound it in cigarette smoke.

Their paper in the current issue of Science, dated Oct. 18, 1996,bears the title: "Preferential formation of benzo[a]pyrene adducts atlung cancer mutational hot spots in P53."

That same issue of Science implicates that same gene's tumor-suppressing protein, P53, in a quite separate research article thatsuggests a putative cure for many cancers. Its title: "An adenovirusmutant that replicates selectively in p53-deficient human tumorcells."

Molecular biologist Gerd Pfeiffer, at the City of Hope's BeckmanResearch Institute in Duarte, Calif., is senior author of the Sciencearticle that pins down cigarette smoke as biological culprit rather thanmerely statistical suspect in the etiology of pulmonary carcinoma.

His starting point was the highly potent mutagen and carcinogen,benzo[a]pyrene. It is widely used to generate experimental tumors inlaboratory animals, and is a prime ingredient in cigarette smoke. Asingle cigarette contains 20 to 40 nanograms of the tumorigenicpyrene chemical.

Using a PCR-amplification variant developed at City of Hope,Pfeiffer and his co-authors tested the damage caused by this pyrene tospecific sites along the p53 gene sequence _ sites known as tumorhot spots specific to lung cancer DNA.

That gene's protein product, P53, is nearly ubiquitous in human cells,and repairs or expunges those that threaten to divide out of control.This tumor-suppressing function fails when the p53 gene suffersmutation, such that its protein doesn't function, and cancer ensues.(See BioWorld Today, May 28, 1996, p. 1.)

Its most frequent mutations in lung cancer involve a genomic swapbetween two nucleotide bases, guanine and tyrosine. Pfeiffer'sresearch singled out three hot spots in particular with this mutation,one occurring exclusively in pulmonary carcinoma, the other two inmany other tumor types as well.

Among these was cervical carcinoma, from which a woman inBaltimore named Henrietta Lacks died in 1951. Her tumor cells havesince been kept in constant culture, perpetuating her name amongbiologists worldwide as the HeLa cell line.

In HeLa cells, the City of Hope team confirmed the pyrene'spreferential attack on p53's guanine mutational sites. Then theyturned to human cells more typical of lung tissue _ normal bronchialepithelium. The hot spots were the same as in HeLa cells.

City of Hope's executive vice president of medical and scientificaffairs, John Kovach, acclaimed Pfeiffer's "elegant study, part of theemerging field of molecular epidemiology [for permitting] thepositive identification of the specific molecule associated with thedevelopment of lung cancer."

From Proven Cause To Putative Cure

In this week's Science, p53 is the common denominator betweenconfirming the prime cause of lung cancer and finding possibletherapies for many cancers.

The senior author of the adenovirus paper is Frank McCormick, chiefscientific officer at Onyx Pharmaceuticals Inc., a four-year-old spin-off of Chiron Corp. in Richmond, Calif.

His points of departure were the two facts: in more than half of allhuman cancers _ including breast, colon and lung _ the p53 gene isdeleted or mutated; and a mutated adenovirus can infect and killtumor cells containing non-functioning P53 protein.

Adenovirus is a relatively benign pathogen, responsible for thecommon cold. It gets in its licks just like the big-boy lethal viruses,by invading respiratory tract cells, and replicating. In this process, itencodes a protein of its own, E1B, which binds to and disarms P53.(Besides preventing cancer, P53 helps cells resist viral infection.)

McCormick envisioned an anti-cancer strategy in which mutantadenovirus, unable to make E1B, would invade a normal cell, inwhich the P53 would prevent the virus from replicating. On the otherhand, a tumor cell _ lacking P53 _ would be wide open to infectionand lysing by the E1B-expressing mutant virus.

He and his co-authors engineered an 827-base-pair deletion in theE1B region of the adenovirus genome. This new mutant, theyreasoned, should be unable to replicate in normal cells protected byP53 surveillance, but could do so in P53-lacking tumor cells.

This viral mutant flourished in HeLa carcinoma cells, which lackfunctional P53, and killed them with high efficiency. They did so tooin other tumor cell lines, including colon carcinoma, glioblastomaand pancreatic adenocarcinoma.

Encouraged by these in vitro results, the Onyx team turned to in vivoexperiments with nude mice, into whose flanks they injected HeLacarcinoma and glioblastoma cells, together with either wild-type orE1B-deleted mutant adenoviruses. After six weeks, the mutant-treated carcinomas had shrunk 84 percent compared with wild-typevirus. In one mouse, the tumor disappeared completely, as haveseveral more in repeat experiments.

This preclinical success has led to a Phase I clinical trial, begun lastApril in 12 patients with head and neck cancer. Such tumors areaccessible to the neutered adenovirus treatment, which must beinjected directly into the malignant lesions. The trial, in San Antonio,Texas and Glasgow, Scotland, reports no adverse side effectswhatever, but therapeutic effect remains to be seen.

"What I like is how clever it is," said Richard Klausner, director ofthe National Cancer Institute. Quoted in a Science editorial, he said:"It's been a long-held fantasy to find an [anti-cancer] virus." n

-- David N. Leff Science Editor

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