6 The Somatic Mutation Theory

THE GENETIC REVOLUTION

In 1866, Gregor Mendel launched the field of genetics with the publication of his now-legendary paper on plant hybridization, which described, among other things, wrinkly versus round peas. The word genetics was coined in 1906 by biologist William Bateson to denote the burgeoning new “science of heredity and variation.”¹ Traits such as eye color and hair color are passed from generation to generation, encoded by sections of deoxyribonucleic acid (DNA) called genes, which are contained within chromosomes.

The German biologist Theodor Boveri noted in 1902 that some sea urchin eggs with an abnormal number of chromosomes grew rather exuberantly, much like cancer cells. He guessed that certain genes within the chromosomes stimulate growth and that mutations of these genes caused the excessive growth.² Boveri also hypothesized that other genes were responsible for stopping growth. If you cut yourself, your body must activate genes that signal cells to multiply and heal the wound. Once the wound is healed, other genes must tell the cell to stop growing. Boveri laid out his basic hypothesis in his 1914 book, The Origin of Malignant Tumours.³

Boveri’s basic postulates were proven correct with the discovery of those exact genes, now called oncogenes (genes that promote cell growth) and tumor suppressor genes (genes that suppress cell growth). The first human oncogene was identified in the 1970s, when it was discovered that certain strains of the Rous sarcoma virus (RSV) caused cancer in chickens but other strains did not. By comparing the two viral genomes, researchers isolated the src gene responsible for the cancerous transformation, the world’s first oncogene. In 1976, Nobel laureates Harold Varmus and Mike Bishop transformed cancer genetics by discovering the human equivalent of the src gene, immediately transforming src from a viral oddity in chickens to a key player in the genetics of most human (and animal) cancers.

Most cancers contain numerous changes to both oncogenes and/or tumor suppressor genes. Src normally increases cell growth the way the accelerator of a car increases motion. RSV causes a mutation in src, inappropriately activating it, leading to the unregulated growth seen in cancer. By the end of the 1970s, two other highly prevalent human oncogenes were discovered, the myc and the egfr genes.⁴

Tumor suppressor genes normally stop cellular growth like the brakes of a car stop its motion. A mutation that inactivates these genes will therefore promote cell growth, just as releasing the brakes will make a car go faster. The p53 tumor suppressor gene, identified in 1979, is the most frequently mutated gene in human cancer.⁵

These new discoveries seemed to offer a perfect explanation for why cancer cells grow so quickly. Both activating mutations of oncogenes and inactivating mutations of tumor suppressor genes could accelerate cell growth, leading to cancer. This coalesced into the widely accepted somatic mutation theory (SMT), which views cancer primarily as a disease caused by accumulated genetic mutations. Somatic cells include all cells of the body except for germ line cells, the cells responsible for sexual reproduction, such as the sperm and egg. Mutations in these somatic cells (such as breast, lung, or prostate) accumulate, and a random aggregation of these mutations may be enough to cause cancer. This view of cancer, what I call cancer paradigm 2.0 (see Figure 6.1), dominated cancer research in the 1970s and is still championed today by the American Cancer Society, which states plainly that “Cancer is caused by changes in a cell’s DNA—its genetic ‘blueprint.’”⁶

Supporting this view, researchers postulated that specific inherited genetic mutations caused cancer without the need for an external agent. Familial, or inherited, cancers are relatively unusual, accounting for only approximately 5 percent of cancers, leaving the vast majority (95 percent) of cancers as sporadic mutations. Nevertheless, the SMT proved that cancer could be as simple as a disease of genetic mutations.

For example, a single inherited genetic mutation in the retinoblastoma tumor suppressor gene causes rare eye cancers in children. An inherited mutation of the von Hippel–Lindau tumor suppressor gene leads to an increased risk of kidney cancers. In breast cancer, the BRCA1 and 2 genes are the best-known susceptibility genes that confer a high risk of breast cancer, but these account for only an estimated 5 percent of breast cancer cases. Overall, the contribution of inherited genetic defects to cancer is small, but those rare cases confirmed the underlying unifying mechanism of carcinogenesis.

Inherited mutations caused cancer. Chemicals, radiation, and viruses could also cause genetic mutations or other changes in the genetic code leading to the dysregulated growth of cancer. Bingo! The pieces of the puzzle were fitting together perfectly.

Rarely, a single mutation is sufficient to turn normal cells into cancer cells. A normal cell contains various mechanisms to repair damaged DNA, so if the damage is minor, it can often be rectified. But if the DNA repairs cannot keep up with the inflicted damage, then mutations accumulate. When several critical mutations converge, cancer results. Most common cancers require multiple mutations.

But how did the mutations accumulate? Asbestos or tobacco smoke or radiation can cause genetic change, but these were not targeted to any specific gene or chromosome. The implicit answer from the SMT was that these mutations were not planned, but that they accumulated more or less randomly. It is simply bad luck when all the critical mutations occur together.

Figure 6.1

The new genetic tools developed in the 1970s showed that cancer cells were indeed chockablock full of genetic mutations. By the 1980s, animal models confirmed that chemicals, radiation, and viruses, the known causes of cancer, could mutate oncogenes and tumor suppressor genes to cause cancer. When mice were exposed to chemical carcinogens, they developed skin cancer, and those cancers had mutations in their oncogenes.⁷

Chemicals, X-rays, viruses, and inherited genetic disorders all have vastly different physiologic effects, but all caused cancer. The common thread was that they all caused DNA damage and gene mutations. A carcinogen causes cancer because it is mutagenic—that is, it increases the rate of gene mutation. Given that mutations accumulate randomly, more mutations increase the risk of cancer, just as buying more lottery tickets increases the chances of winning the jackpot.

The SMT suggested the following chain of events, represented in Figure 6.2:

Normal somatic cells (e.g., lung, breast, or prostate) sustain DNA damage.
If the rate of DNA injury exceeds the rate of repair, then random genes become mutated.
A chance mutation in a gene controlling growth (oncogenes or tumor suppressor genes) causes exuberant and sustained growth. This is an important first step toward cancerous transformation, but not the only one, because growth represents only one of many hallmarks of cancer.
Other gene mutations accumulate randomly over time. When certain critical abilities (hallmarks) coalesce, the cell fully transforms into a cancer.

Figure 6.2

Most common cancers need multiple mutations. It is like the game of baseball. A big hit, like a home run, scores all by itself. A single horrendous mutation, such as retinoblastoma, may result in cancer. But in baseball, you can also score runs by putting together multiple hits. Multiple gene mutations can also combine to become cancer. Increasing mutation rates—say, with tobacco smoking—increases the risk of mutations. Given enough mutations, cells will eventually, by chance, become cancerous, just as an infinite number of monkeys randomly pounding the keys of an infinite number of typewriters will eventually produce the novel War and Peace.

These random mutations bestow all the “superpowers” needed for cancer to thrive. The ability to constantly grow, to become immortal, to move around, and to use the Warburg effect are all above and beyond what normal cells may do. Having accumulated all the superpowers that define cancer cell behavior, these cells replicate and grow. The resulting mass of cancer cells, the tumor, is a genetic clone of this original cancer cell.

The basic postulates of SMT include:

Cancer is caused by acquiring multiple DNA mutations.
These mutations accumulate randomly.
The cells in the tumor are all derived from one original clone.

Most gene mutations are lethal, but a small percentage are neutral or beneficial. The odds of randomly acquiring all the requisite mutations to transform a cell into cancer is small, but if the rate of mutation is high enough, then it is bound to happen. This small likelihood of success explains why cancer often takes decades to develop, and why cancer risk rises sharply in people over the age of forty-five (see Figure 6.3).⁸

The somatic mutation theory of carcinogenesis patched together all the disparate known causes of cancer into a coherent, unified theory. This paradigm focused research from extrinsic agents (chemicals, radiation, and viruses) onto intrinsic defects (genetic mutations). All these various carcinogenic insults created the seed of cancer by causing gene mutations. While both the seed and the soil are important for growth, it seemed that the seed, according to the SMT, was the most important component. Cancerous cells are similar to normal cells because they were derived from normal cells. Cancer cells weren’t alien intruders, but mutated versions of our own cells. We had seen the enemy, and it was ourselves.

NCI, “Age and Cancer Risk,” National Cancer Institute, April 29, 2015, https://www.cancer.gov/about-cancer/causes-prevention/risk/age.

Figure 6.3

The SMT was a breakthrough, promising new directions for research and new treatments. Cancer was now viewed as a cell-centered problem of genetic mutations. If we could find and then treat those mutations, logic followed, we could cure cancer. SMT led to some astounding predictions and some astounding successes. Rather than simply using the traditional tools of cancer medicine—to cut, burn, or poison—we could use incredibly precise molecular tools to develop entirely new pharmaceutical protocols to cure cancer. By the 1980s, the SMT made good on that promise, delivering one of the most spectacular weapons yet seen in the war on cancer.

THE PHILADELPHIA CHROMOSOME

In 1960, at the University of Pennsylvania in Philadelphia, researchers Peter Nowell and David Hungerford were studying human chromosomes in leukemia. Two patients with a rare type of blood cancer, chronic myelogenous leukemia (CML), shared a characteristic chromosomal abnormality. Odd. One of the chromosomes was consistently much smaller than normal.⁹ This was dubbed the “Philadelphia chromosome” for its city of discovery. When healthy cells divide normally, they provide exactly the same chromosomes to each new daughter cell. In the Philadelphia chromosome, a piece of chromosome 9 ended up on chromosome 12, and vice versa. This abnormality occurred in almost all cases of CML, and was exclusive to CML—no other types of cancer displayed this characteristic.

The Philadelphia chromosome produced an abnormal protein known as the bcr/abl kinase, a protein that toggles cell growth on or off precisely, depending upon the situation. The abnormal bcr/abl protein turned cell growth “on” and never turned it off. This uncontrolled growth eventually led to cancer. Researchers hunted for a drug to block this kinase, and in 1993, the drug firm Ciba-Geigy (now Novartis) selected the most promising candidate, called imatinib, to undergo human trials.

Human drug trials typically comprise three phases. Phase 1 studies are designed only to assess drug toxicity. This allows researchers to establish a safe dose so that further investigation may determine the drug’s effectiveness. In these early trials, imatinib improved CML in an astounding fifty-three of the fifty-four patients who took over 300 mg/day. This was a miracle. Researchers would have been happy if nobody died during this phase, but instead they found a virtual cure. Better yet, there was no evidence of significant drug toxicity at this dose.

The larger, phase 2 trials test efficacy, and about two thirds of investigational drugs end their journey here. Pharmaceutical researchers are generally happy if their drug kills a few cancer cells and manages not to kill any patients. Imatinib breezed through phase 2 like an Olympic hurdler. An unheard-of 95 percent of early-stage CML patients completely cleared their leukemic cells. Even more astounding, the Philadelphia chromosome could no longer be detected in 60 percent of treated patients. This drug didn’t just kill CML cancer cells; it was essentially curing the cancer.

It was a miracle drug, but more exciting, it provided a proof of concept for this new genetic paradigm of cancer. Imatinib was going to be the vanguard in the coming onslaught of new, targeted medications that promised superior efficacy with lower toxicity than standard treatments like chemotherapy. As we’ve discussed, chemo medications are selective poisons that kill cancer cells slightly faster than normal cells. If chemo could be considered a kind of carpet bombing, then this new generation of drugs would be the “smart bombs” of the cancer arsenal, homing in on specific targets to destroy cancer cells without causing much collateral damage.

Imatinib, known as Gleevec in the United States, is the unquestioned superstar of the genetics-focused approach to cancer treatment. Before the introduction of imatinib, CML was responsible for taking roughly 2,300 American lives per year; in 2009, after imatinib treatments began, annual CML deaths were reduced to 470. This oral medication with virtually no side effects was so dramatically successful that it was considered to herald an entirely new era of precision-targeted chemotherapy.

With the introduction of imatinib, science marked the dawn of a new age of genetic “cures” for cancer. On the cover of its May 28, 2001, issue, Time magazine proclaimed that “There is new ammunition in the war against cancer. These are the bullets.”—alongside a picture of imatinib. It was an entirely new and better way of treating cancer, just in time for the new century.

The genetic paradigm of cancer had proven its mettle in the crucible of battle. Finding the precise genetic abnormality led to identifying the abnormal protein, which led to discovering a medication to neutralize that protein, virtually curing that particular cancer. Yes, CML was a relatively rare cancer, but this was only the beginning. Soon, another major victory would be achieved in breast cancer, with the development of the drug trastuzumab. Unlike CML, breast cancer was a major-league cancer, second only to lung cancer in causing cancer deaths in women.

HER2/NEU

In 1979, researcher Robert Weinberg at the Massachusetts Institute of Technology was chasing oncogenes. He discovered a cancer-causing segment of DNA taken from neurologic tumors in rats that he named neu. The human equivalent was discovered in 1987 as human epidermal growth factor receptor 2 (HER2), so this gene became known as HER2/neu, a potent oncogene. Up to 30 percent of all breast cancer cases overexpressed the HER2/neu gene by up to one hundred times normal. Those cancers are much more aggressive and often deadlier than those without.

The newly established, soon-to-be giant drug company Genentech located the HER2/neu gene using DNA probes, but the question remained: how were they going to block it? Standard drugs are small molecules that can be synthesized in a chemical plant, but none of these specifically blocked the HER2 protein as imatinib had done so successfully with the bcr/abl kinase. But by the 1980s, the technology of the genetic revolution had advanced substantially, and Genentech pioneered an entirely new class of therapy that would offer another major leap forward in cancer treatment.

A healthy immune system produces proteins called antibodies to help fight off foreign invaders. Antibodies are very specific in their targets. For example, infection with the measles virus stimulates the body to produce antibodies that recognize measles. After you successfully fight off the infection, your body retains these antibodies. If you are reexposed to measles, your preexisting antibodies instantly recognize the virus and activate the immune system to destroy it. This is why it is rare to develop measles infections more than once in your life. Antibodies work by recognizing specific DNA sequences, and Genentech insightfully recognized that HER2/neu is also just a DNA sequence.

In a remarkable feat of genetic engineering, scientists at Genentech created a mouse antibody that could bind and block the HER2/neu protein. But a mouse antibody injected into a human would be instantly recognized as foreign and destroyed by the human immune system. Genentech’s ingenious solution was to create a mouse–human hybrid antibody to block the HER2/neu gene with high specificity, which became the drug called trastuzumab (Herceptin).

But there was yet another problem. Only about 30 percent of breast cancers carry the abnormal HER2/neu gene, so administering this very costly drug to every breast cancer patient would be extraordinarily wasteful and prohibitively expensive. So, in another innovative leap forward, scientists developed a simple test for the gene. Now only those patients with cancers overexpressing the abnormal HER2/neu were given trastuzumab.

This exciting development ushered in a new era in therapeutics. Drugs would not only be precision-guided weapons but also personalized. A drug did not need to work for every patient suffering from a disease in order to help a subset of those patients. We could define and treat only those expected to benefit. This approach saved money and spared patients from unnecessary side effects. Amazing. Medicine had finally found the holy grail of genetic therapy. If we could identify those few mutations that drove cancer for each specific person, we could then select the proper drug or antibody to use. Treatments could be personalized through genetic testing to reverse and potentially cure the disease.

Even before FDA approval, breast cancer patients implored Genentech to release the drug on compassionate grounds. Nobody yet knew if it worked, but patients with metastatic breast cancer had no other options, and trastuzumab was a gleaming beacon of hope. In 1995, Genentech set up the first-ever FDA-approved expanded-access program for cancer drugs. Its hunch was correct. By 1998, trastuzumab was approved by the FDA for HER2-positive breast cancer and ready for prime time. By 2005, human trials showed that Herceptin cut the risk of breast cancer deaths by about one third.¹⁰ The genetic age of precision, personalized cancer medicine had begun gloriously. It would be all rainbows and unicorns from here on, right?

CANCER PARADIGM 2.0

The genetic revolution had led us, by the early 2000s, to a major threshold. Our arsenal in the war on cancer had so far consisted of indiscriminate ways of killing cells: cutting (surgery), burning (radiation), and poisoning (chemotherapy). Blasting a cancer into oblivion seemed so crude compared to using highly specific gene-targeted antibodies to deliver a deadly payload of toxins. Only the “bad guys” were killed, sparing the collateral damage commonly seen with the older treatments. Victory seemed inevitable, as we landed body blow after body blow against cancer. We had new weapons that could penetrate cancer’s tough carapace. We had new defenses against cancer’s deadly pincers. The next step was to map out the one or two genetic mutations for each cancer the same way we had done for CML and the HER2/neu-positive subset of breast cancer.

Imatinib proved that the concept worked in “liquid” tumors of the blood, such as CML, and trastuzumab proved that this concept worked for solid tumors, too. It was only a matter of finding the mutations of various cancers and then designing the right drugs to destroy them.

Figure 6.4

The genomic revolution was unstoppable and showed no sign of slowing. Rather, the pace of technological advance and medical knowledge was accelerating. The new drugs, while difficult to develop, were priced accordingly, and profits from the first few were prodigious. Countless start-ups, big pharmaceuticals, and universities alike joined in the new gold rush. With the map of the human genome in hand, finding the mutations that were then blocking researchers from curing cancer would be as easy as shooting fish in a barrel.

Our understanding of cancer had progressed substantially from a “disease of excessive growth” to a “disease of genetic mutations that was causing the excessive growth” (see Figure 6.4). We had peeled back a layer of the truth of cancer’s origin: carcinogens cause cancer by causing gene mutations. Now that we understood the underlying root cause of cancer, we could develop lifesaving drugs.

As we rounded the corner of the twenty-first century, it seemed to many that we were on the precipice of a cancer-free world. Imatinib and trastuzumab were instant hits. But like so many other one-hit wonders, the first hit turned out to be the best.