by Azra Raza
WHAT CAUSES CANCER?
In his poem, “Miss Gee,” W. H. Auden offered a scathing criticism of the prevailing view of cancer in the 1930s where society associated the disease with a failing of the individual.
Doctor Thomas sat over his dinner,
Though his wife was waiting to ring,
Rolling his bread into pellets;
Said, ‘Cancer’s a funny thing.
‘Nobody knows what the cause is,
Though some pretend they do;
It’s like some hidden assassin
Waiting to strike at you.
‘Childless women get it.
And men when they retire;
It’s as if there had to be some outlet
For their foiled creative fire.’
It is not exactly childless women and retired men who get cancer; today, one in two men and one in three women will get it. Many of my patients look puzzled by their diagnosis, as did Doctor Thomas, not because of the state of their creative fire but because of how they lived; these people never smoked or drank, and they exercised regularly. Take, for example, Suketu Mehta, the author of the fantastic book Maximum City. He became my friend not long after I moved to New York City. One evening in 2009, I received an unexpected call from Suketu. He sounded shaken. “Azra, I have just been diagnosed with lung cancer. How is this happening? I am forty-five years old. I have never even smoked.” After a night of chili with his partner and their families, Suketu had woken with a fluttering in his chest. Worried by memories of his uncle, who died at thirty-four of heart disease, he went to see his doctor. She gave him an EKG. “Your heart is fine,” she told him. “The fluttering is probably nothing more than heartburn. But let’s get you a chest x-ray, just in case.”
And there it was: a two-inch spot over my lung, the earliest stage of a malignant tumor. I’ve never smoked, so I never would have been checked for this. By the time I developed symptoms, it would have been too late: 85 percent of people diagnosed with lung cancer die within six months.
Cancer is what happens when some part of ourselves wants to live forever. The body is more a confederation of cells agreeing to act in concert than a single organism. When a cell refuses to die and transmits that obdurate life force to its neighbors, we get cancer—death brought on by the striving for immortality.
Where does such agreement, the pursuit of immortality, come from? Is cancer related to our lifestyle, exposure to toxins, what we eat, or where we live, or is it a random event? Is it a consequence of aging? In the memorable phrasing of the science writer Wayt Gibbs, anyone seeking “a workable theory of cancer has to explain both why it is predominantly a disease of old age and why we do not all die from it. A 70 year old is roughly 100 times as likely to be diagnosed with a malignancy as a 19 year old is. Yet most people make it to old age without getting cancer.”
Cancer begins with genes. Genes, made up of DNA, coiled and packed into chromosomes during mitosis, carry the code for proteins. DNA is first copied into RNA, which serves as a template for protein synthesis by the cell. Proteins carry out cellular functions. Each time a cell divides, it must faithfully double its DNA, to parcel it out equally to the two daughter cells. Because three billion base pairs need rapid replication, errors or mutations occur. Mutations are continuously edited, repaired, and corrected by built-in cellular mechanisms. If repair is not possible, and the mutation is in a vital gene, the cell is forced to commit suicide. If the mutation is in a gene not vital for the cell, it can persist and be passed on to the next generation. Most DNA mutations are inconsequential—their resulting proteins are either insubstantially changed or not changed at all. If, however, the error affects genes whose function is to either promote or arrest growth, a cell can be driven into wildly irregular paths of unstoppable proliferation: cancer.
Essentially, cancer-initiating events can be triggered by a factor internal to the individual, such as increasing age or a genetic predisposition, or by something external to the individual, such as DNA-damaging environmental toxins, tobacco, alcohol, ultraviolet radiation, or pathogens. Pathogens as etiologic agents for malignancy might seem surprising, but roughly 20 percent of cancers worldwide are caused by viruses or bacteria. For example, in 1977, adult T cell lymphoma was described among the Japanese population and later, human T cell lymphotropic virus-1 (HTLV-1), discovered in the laboratory of Robert Gallo, was shown to be the cause. HTLV-1 can cause many nonmalignant, morbid and fatal diseases like uveitis or myelopathy, but its cancer-causing or oncogenic potential is dramatic. Other viruses considered as causative agents in cancer include papilloma viruses (associated with several types of cancers, most notably, cancer of the cervix), Epstein-Barr virus (Burkitt’s lymphoma, some forms of nasopharyngeal and stomach cancers), the hepatitis B and C viruses (liver cancer), and human herpes virus-8 (HHV-8 associated with Kaposi’s sarcoma). Helicobacter pylori is the first and only bacterium directly associated with cancer (gastric cancer and gastric lymphoma).
The pathogens listed above cause cancer in a manner similar to the way smoking causes lung cancer. Both initiate changes in a cell that free it to launch into endless cycles of division, unchecked by normal growth-inhibiting impulses, acquiring a life of its own, evolving and metamorphosing into a killer machine with a lawless, mutinous, riotous independence. Lung cancer does not disappear when the patient stops smoking because the harm done by smoking was only the initiating event. No matter what the triggering event—smoking, a virus, or toxic exposure—ultimately, the prevalent view is that a genetic change must happen within the cell if there is to be cancer.
What type of genetic change? In cancerous cells, one can find mutations that seem to shut down genes protective against cancer and to trigger those that seem to cause it. One can also find a condition known as aneuploidy, or a change in the chromosomal makeup of cells. There could be extra copies of chromosomes, or chromosomes could be missing, or they could be broken. In short, the causes seem to involve either genetics—what genes are there—or cytogenetics—what the chromosomes are like, or both. And then there’s the questions I was discussing with Per Bak: whether each change could be like a grain of sand, and cancer the point when the pile suddenly collapses. Could the anarchic insurrection staged by cancer, the nihilistic mobocracy, be the result of external factors forcing the cells into mutiny?
BORN IN 1879 in Baltimore, Peyton Rous was keenly interested in biology from an early age. He graduated from Johns Hopkins University in 1905 with an MD, and from 1909 to his death in 1970 at ninety years of age, he was attached to the Rockefeller Institute in New York. Rous was working as a pathologist in 1910 when a farmer brought him a Barred Plymouth Rock fowl with a lump in its breast. Rous diagnosed this as a sarcoma and proceeded to study it further in his lab. He transplanted the malignant cells from the primary tumor into other animals. When he transplanted them into unrelated animals, nothing happened. When he transplanted them into related animals, however, fresh tumors not only appeared, they became increasingly more aggressive, more invasive, with subsequent passages. “It is a spindle-celled sarcoma of a hen,” Rous wrote in his report, “which thus far has been propagated to the fourth generation. This was accomplished by the use of fowls of pure blood from the small, intimately related stock in which the growth occurred. Market bought fowls of the same variety have shown themselves insusceptible, as have fowls of mixed breed, pigeons and guinea-pigs.”
Cancer could be transmitted from one animal to another, but the question about the causative agent remained. Rous began by mincing the tumor in saline and passing it through a filter so fine that it trapped cells and any other particles as small as bacteria. He injected the filtered extract into related healthy fowl. New tumors appeared. Because both cancer cells and bacteria had been filtered from the extract, Rous concluded that something smaller than a bacterium, a virus, was the cause of sarcoma. With this observation began the field of tumor virology. The Rous sarcoma virus (RSV) was later classified as an RNA virus because of
its RNA genome and subsequently as a retrovirus after the discovery of how RNA could be reverse-transcribed into DNA. RSV became the first known cancer-causing virus.
Initially, Rous’s discovery, which would eventually earn Rous a Nobel Prize half a century later, went unacknowledged, unstudied, ignored. At the time when Rous reported his findings, cancer was not a widely examined, popularly studied subject, and neither were viruses. What’s more, it was hard for scientists of the time to imagine how a tumor in birds could have any relevance for humans. Rous himself doubted the significance of his findings and abandoned cancer research. But then in 1930, a second cancer-causing virus surfaced, when Richard Thorpe showed the papilloma virus to be the cause of warts in rabbits. It was now hard to ignore Rous’s work, and the discovery of a second cancer-related virus rekindled interest in RSV. The newfound attention restored confidence in Peyton Rous, who returned to studying cancer. Subsequently, cancer-causing viruses were discovered in many other animals, including mice, cats, and primates. In 1964, Epstein-Barr virus was shown to be the causative agent of a type of lymphoma in humans. The race was on to find new oncogenic viruses and the mechanism by which they induced cancerous behavior in cells.
RSV reliably transmitted sarcoma in inbred animal models. Once molecular techniques became available, the study of RSV began in earnest. Mutations were artificially induced in its genome, and a strain was developed that continued to replicate but failed to cause cancer. When Peter Duesberg and Peter Vogt compared the two strains of cancer-causing and non-cancer-causing RSV, they found that the former had two subunits of RNA, a large and a small one, while the latter contained only the smaller one. The larger piece of RNA was the ultimate driver of the malignant phenotype. The first cancer-causing gene, or oncogene, had been identified. It was named src because it caused a sarcoma. Once the transforming activity of a virus was shown to depend upon the oncogene it was carrying, additional oncogenes were discovered in rapid succession in cancers affecting birds and mammals. The joke in the 1980s was to name an A-list cancer researcher who had not yet discovered an oncogene.
A wise person once said that an important discovery in science should not be followed by an exclamation mark but by a semicolon, as science is always a continuous process. Certainly the story of oncogenes became more exciting when two scientists, Mike Bishop and Harold Varmus, showed that the src oncogene, with minor variations, was also present in human cells. The gene was likely picked up from human cells by the RSV retrovirus during its natural life cycle. Now there were two oncogenes with minor differences—the RSV viral version called v-src and the human cellular version called c-src. The proteins made by v-src and c-src control fundamental functions of cell proliferation and death. Because the c-src in human cells was not directly associated with existing cancer, it is considered a proto-oncogene. Proto-oncogenes acting normally serve to promote cell division. They can become dysfunctional in one of two ways—by a mutation that changes the behavior of the gene, causing it to drive cell division in the absence of normal growth signals, or because the regulation of the proto-oncogene becomes abnormal, leading to excessive copies of the gene—and so its own regulatory proteins—being made. Either way, the result is the runaway tissue growth characteristic of cancer.
Cancer can also result when growth-arresting signals are lacking. The genes responsible for arresting the growth of tissues are known as tumor suppressor genes (TSG). The TSG p53 is the most important member of this class. Its function is to constantly survey the cell for any sign of DNA damage. Upon detecting an unrepaired piece of DNA or abnormal growth signal, p53 forces the cell to either repair itself urgently or commit suicide, thereby preventing cancerous behavior of the cell. TSG p53 is known as the guardian of the genome. It activates proteins that put brakes on cell division. It is our most prominent intracellular defender against cancer. In order to make it past this policeman of the cell cycle, cancer cells need to subdue the normal surveillance function of p53. Mutations in the gene lead to production of an abnormal p53 protein incapable of performing the vital cell-wide supervision and induction of programmed cell death. This failure results in unchecked growth of the cell. Indeed, p53 is the most commonly mutated gene in many types of cancers.
Germ line mutations in tumor suppressor genes also lead to cancer susceptibility. Li-Fraumeni syndrome (LFS) is a hereditary disease in which 100 percent of affected individuals end up with cancer. Half of them develop a malignancy before thirty years of age and all by seventy years. Cancers of the blood, brain, breast, bones, gonads, adrenals, and GI tract are the most common. Mutations in p53 are present in 70 percent of LFS cases, while the remaining 30 percent show mutations in another tumor suppressor gene called CHEK2.
Aruna and Sam Gambhir found out about Li-Fraumeni syndrome through an unspeakable personal tragedy. Their brilliant fourteen-year-old son, Milan, was lake-tubing when he struck his head and suffered a concussion. The treating doctor ordered a CT scan of his head to rule out intracranial bleeding, but no one could have imagined that this simple act of imaging could damage a cell sufficiently to cause brain cancer. Milan died at the age of sixteen from one of the most aggressive, ruthless killers known to mankind: glioblastoma multiforme, which has a five-year survival of less than 5 percent. Sam Gambhir’s entire professional life had been spent finding ways of detecting cancer early. The previous year, in fact, he had successfully competed for a $10 million grant to detect early signs of cancer. Milan himself had worked with researchers in the Canary Center at Stanford University to develop a wearable ultrasonic wristband for early detection of recurrent cancer using a sophisticated microbubbles technology. In a crushing, ironic twist, Sam—who chairs the Department of Radiology at Stanford University—watched as the first films revealing the large intracranial mass in his son’s brain emerged from the CT machine after Milan presented to the emergency room with a seizure.
Aruna Gambhir had already weathered two bouts with breast cancer. Milan’s wristband idea was a direct inspiration from realizing that it was early detection of breast cancer that saved his mother’s life. Mother and son underwent genetic testing after Milan’s diagnosis, and both showed the presence of an inherited p53 mutation. “It’s possible that he developed this tumor from the CT scan radiation,” says Sam Gambhir. “When you carry this p53 mutation, you are much more susceptible to radiation. In a normal person, a CT scan wouldn’t be a big deal. But in someone with this mutation, it likely increases their chances of cancer. We will never know for sure.”
The functional integrity of p53 is associated with cancer prognosis as well. In MDS, for example, when patients present with damage to multiple chromosomes, the cancer genome has been considered highly unstable with a resultant poor prognosis for the patients. Studies show that if the complexity of cytogenetic damage is accompanied by a p53 mutation, the prognosis is indeed quite poor, but if there is no p53 mutation, then patients can live for many years without disease progression despite having many damaged chromosomes. The primary driver is the mutation in p53 and not the damaged chromosomes. On the other hand, MDS patients who present with an isolated deletion in the long arm of chromosome 5 (deletion 5q) are supposed to have a good prognosis—a stable, slowly progressive disease with a long survival. But one in five such patients with ostensibly low-risk disease shows mutations in p53, and they tend to advance to acute leukemia rapidly. This is why any information relating to a possible bad prognosis in patients with complex cytogenetics or good prognosis in patients with deletion 5 is incomplete until p53 mutational status is known. Genetics trumps cytogenetics.
There is another curious aspect of p53 that has come to light recently. Chances of spontaneous mutations increase each time a cell divides. Because larger animals have more cells, it would appear to stand to reason that they should have more mutations, and so, more cancer. Yet the opposite is true; the incidence of cancer in humans is lower than in mice and higher than in whales. Elephants hardly ever get cancer. This conundrum is known as Peto�
�s paradox, named after the epidemiologist Richard Peto. It poses the question of why the incidence of cancer does not increase with increasing numbers of cells in an organism. Peto speculated that intrinsic biologic mechanisms operating within the cells of an expanding and aging animal protect them from cancer. That seems to be right.
Large body size is important because it improves fitness and assures longer life by avoiding predators. There are eleven placental mammalian orders in the animal kingdom, and ten of them have acquired large body sizes, along with a number of different strategies to avoid cancer. One mechanism discovered recently is that elephants have twenty copies of p53. Just as proto-oncogenes can become oncogenes by increasing their copy number, a higher copy number of p53 can prevent cancer altogether. The discovery prompted excitement; can we become the elephants in the room and begin the ending of the cancer saga by inserting multiple copies of p53 into our genomes? Such redundancy would mean both more gene transcripts and protection against any one copy of the gene being disabled by random mutations. As scientists tinkered with this idea in the lab, they ended up with mice whose cells showed a hyperactive p53. The mice were resistant to developing cancer if exposed to DNA-damaging agents that normally induce malignancy. The discovery was very exciting. Unfortunately, the trade-off was less so. The p53 hyperactive mice aged rapidly, and within months, they looked very old, and their life spans shrank by 30 percent. The mechanism of this rapid aging turned out to be stimulation of the hormone responsible for cell proliferation called insulin-like growth factor 1, or IGF-1, which is controlled in turn by p53. Amplified IGF-1 signaling accelerated the entry of cells into senescence. And senescence, as we have seen before, is closely linked to aging. In short, if there is no p53, the cell becomes cancerous; if there is overactive p53, the cell ages and dies prematurely.
