by Azra Raza
The story does not end here, because there is yet another remarkable twist in the saga: p53 is kept in check by its controller, called Mdm2. As soon as p53 is switched on, it activates Mdm2 to assure its simultaneous degradation, thereby preventing its accumulation and overactivity. Artificially suppressing Mdm2 activity in turn would be expected to enhance p53’s activity. To study this effect, Mdm2 knockout (KO) mice were created that lacked the controller entirely. When a drug to stimulate p53 was administered to these KO mice, the results were nothing short of catastrophic. The mice essentially melted away due to massive, uncontrolled suicidal death of cells all over the body. The unintended consequences of tinkering with the p53 gene are brilliantly described in Sue Armstrong’s eminently readable book p53: The Gene That Cracked the Cancer Code.
To make the story of p53 even more complicated, in 2002, another group reported the generation of mice with extra copies of p53. These “Super p53” mice were protected from cancer and did not age prematurely, probably reflecting the fact that p53 was under normal regulatory control.
But p53 is also not the only answer to the issue of large animals and cancer. Whales don’t get cancer, but unlike elephants, even the gigantic bowhead whale—with a life span of over two hundred years—shows no extra copies of the tumor suppressor gene p53. One way to prevent cancer in large animals is by slowing down metabolism and reducing the production of DNA-damaging reactive oxygen species. Another, seen in naked mole rats, is activation of a different tumor-suppressing pathway signaling through hyaluronic acid.
Little of this has shown us how to avoid cancer in humans. But comparative biologic studies are certainly adding enormously to the body of knowledge that, one day, is sure to be extremely helpful for all animals on the planet and should continue.
WHETHER MUTATIONS TRIGGER the activity of oncogenes or alter the function of suppressor genes, statistical analyses conducted by Bert Vogelstein and Cristian Tomasetti indeed show that the number of times an organ’s stem cells divide determines how prone the organ is to cancer. In thirty-two different types of cancers, 66 percent of the mutations that drive the malignant process—which are known as founder mutations—were due to DNA replication errors.
Work done by Vogelstein’s group on colorectal cancer also showed the rate at which the mutations arise and which mutations actually tip a cell into cancer. Colorectal cancer develops slowly, transitioning through the three distinct phases of initiation, expansion, and metastasis, often taking two to three decades to reach the full-blown form that we see in advanced cases. Immortalization of the cell is most commonly due to acquisition of somatic mutations in the DNA. Some of these can be hereditary, while others are induced by environmental factors (much like benzene-caused mutations leading to secondary MDS and AML). But the vast majority of DNA mutations arise due to internal processes of the cell. An average of three copying errors occur with each round of DNA replication. In addition, mutations arise due to the quantum effects of base pairing between the two strands of DNA in a chromosome; mistakes induced by DNA polymerase, the enzyme that enables DNA molecules to copy themselves; metabolic DNA damage from reactive oxygen species; and hydrolytic deamination, which has the effect of converting DNA bases into different forms. All contribute significantly to DNA damage. Usually, there is one or very few driver mutations in vital genes that tip the cell into a malignant state. There are approximately 140 so-called driver genes, affecting just about a dozen major signaling pathways involved in the cell’s proliferation, differentiation, and normal functions that are responsible for the cancer phenotype. Genes that determine cellular fate and survival constitute about 90 percent of these while 5–10 percent control the rate of mutations of all genes. The most familiar of the last group are BRCA1 and BRCA2 genes where an inherited mutation leads to vastly increased risk of many types of cancers, especially those of breast and ovaries.
Such driver mutations might seem obvious targets for treatment, and they are in children, as the malignant cells in a child’s cancer are otherwise naive to the many “passenger mutations” that cells acquire over the course of decades. Cancer cells generally show one or two founder mutations but produce a scattering of daughter cells, each of which has acquired a different set of passenger mutations. A passenger mutation does not directly affect proliferative function, but by hitchhiking along the founder mutation, it can affect clonal expansion. As cancers grow, they evolve, continuously acquiring additional mutations and genetic diversity, so that an ecosystem of clones is produced bearing the original founder and a variety of additional passenger mutations.
Expansion of a clone depends upon the fitness landscape between its genetic architecture and the microenvironment. A primary tumor in the stomach would have a very different soil to negotiate compared to one of its daughter cells that home in the liver as a metastasis. The founder mutations would be the same in the clones of cells growing in the stomach and the liver, yet their behavior and responsiveness to therapy would depend on the sum of passenger mutations and local signals in the soil. A drug targeting the founder mutation could get rid of the dominant clone of cells producing even dramatic tumor regressions, but the subclones waiting on the sidelines with a different genetic profile would eventually acquire a growth advantage and cause relapse. No—they cause relapse with a vengeance, because by definition, these are clones selected for survival precisely because they were resistant to therapy.
There are several questions that arise from the above conclusion. The first relates to prevention. If cancer always results from a cell’s intrinsic typo and has nothing to do with factors outside of the cell, such as the environment, then no amount of lifestyle changes would make a difference. However, this is not the case since we do see lifestyles affecting cancer incidence. For lung tumors, for example, DNA copying errors accounted for only 35 percent of the mutations while environmental factors accounted for 65 percent. A second question is, if a mutation can happen any time a cell is preparing to multiply, then why is it that cancers are more common in older age? Here Per Bak’s work and life come back into focus. MDS, which Per was suffering from, results both from factors intrinsic to the cell and the microenvironment surrounding the cell that appears to be full of inflammatory changes. Perhaps the only seed that can survive in such a toxic environment is a cell with a genetic mutation that has caused it to escape the normal growth-controlling signals. What possible changes in the microenvironment of the human body would increase the chances of a cell carrying a mutation to survive at the expense of normal cells? After reading about the phenomenon of self-organized criticality, I began to wonder about events preceding the intracellular gene-chromosome catastrophe causing a malignant transformation. The system could have already become unstable, poised for an avalanche at the least disturbance.
THE DISTURBANCE IN an unstable system poised for catastrophe may come from aneuploidy, a biological reality that challenges the current gene-centric obsession of cancer researchers. Humans inherit two sets of twenty-three chromosomes, one from each parent. A cell has aneuploidy if it contains fewer or more than those forty-six chromosomes. Aneuploidy arises during cell division due to unequal segregation of chromosomes with one daughter cell acquiring more and the other fewer than forty-six. What causes aneuploidy? Mutations in genes, especially those regulating repair of damaged DNA in a cell, can cause chromosomal instability and subsequent aneuploidy. As far back as 1902, the German scientist Theodor Boveri observed that if sea urchin eggs were aneuploid, embryos showed abnormal development. He proposed that having the incorrect chromosome number predisposed a cell toward cancer. Cells with aneuploidy produce abnormal amounts of proteins because of the number of functioning genes, interfering with vital proliferation and death signals. Roughly 90 percent of solid tumors and 75 percent of liquid cancers manifest aneuploidy.
Both genetic mutations and aneuploidy are hallmarks of malignancy, but the relative importance of each as the primeval cause of cancer has been a subject of debate for
decades. One side argues that aneuploidy comes first, and genetic mutations arise because of chromosome breaks, while the other suggests a driver role for genetic mutations with aneuploidy as the downstream consequence.
In 2017, researchers at Cold Spring Harbor conducted an experiment in which they cultured two groups of cells side by side. One had the normal number of chromosomes, and the other had one additional chromosome. The aneuploid cells grew slowly at the start, but eventually, a sudden burst of growth occurred and, almost overnight, they began dividing rapidly. As cells multiplied, more and more abnormalities appeared in their chromosome number. The lab dish seemed to recapitulate events in the body, where a primary tumor grows sluggishly for a while, abruptly bursting into metastases with newfound aggression. Cells with aneuploidy had a survival advantage over cells with a normal chromosome number. They also displayed genetic instability as aneuploidy sequentially worsened in daughter cells, some having more and some fewer chromosomes than the parent cells.
Could the initial slow growth represent a phase of self-organization with the system persistently moving toward entropy, the population becoming increasingly more unstable just like sandpiles, ultimately reaching the state of self-organized criticality when any event could tip the system? Just as the last grain of sand causing collapse in a sandpile is no different from other grains, the cell causing a cataclysmic change may not be very different from others in the plate. The whole plate of cells becomes hypersensitive and unstable, prone to cataclysmic changes. In this setting, even a minor copying error in the DNA, a passenger mutation picked up as the cells divided, which otherwise would be of little consequence, could tip the system.
ON A BEAUTIFUL morning in early 2000, Harvey, Sheherzad, and I were enjoying a particularly spectacular sunrise over Lake Michigan from our living room window. Our apartment in Chicago overlooked Lakeshore Drive and Lincoln Park Zoo and provided a panoramic view of the city from the John Hancock building to the Sears Tower. Harvey was in a great mood. It was a happy morning as Sheherzad ran around, ecstatic at seeing her parents looking relaxed for a change. Harvey asked me if there was something special I wanted to do. He looked rested and well, so I made the impossible demand. Would he come with me for a jog by the lake? We used to love running together, but Harvey had not ventured out in months. His eyes lit up, and he said, “Why not?”
We had barely reached the Peggy Notebaert Nature Museum, a couple of blocks from our building, when Harvey slowed down.
“What’s wrong?” I asked.
“I’m not sure, but I feel I can’t breathe properly,” he replied.
We stopped and rested for a bit and tried again. Another block and the same thing again. We returned home. He started to get shorter of breath as the day wore on. I suggested we go to the ER, but he refused. I gave him Sheherzad’s nebulizer, and it helped him for a while. We spent an anxious day at home. Harvey went to lie down in the bedroom, watching the Ken Burns series on the Civil War. He became so engrossed that my periodic intrusions to check up on how he was feeling became annoying. I tried to leave him alone.
By now, Sheherzad was used to sudden cancellations of our best-laid plans and did not blink when I told her we would be eating at home. We went to bed early. At 4:00 a.m., Harvey woke me up, saying he needed help. He was sweating and looked like he was about to pass out, struggling to catch his breath. I wanted to call 911, but he asked me to drive him instead because the ambulance would take us to the nearest ER, while he wanted to reach Rush University hospital, where we worked. With the help of the housekeeper, I got him dressed and into the car. I had called the hospital in advance, and as we drove up, the crew was waiting with a wheelchair at the door. Harvey was intubated within minutes and placed on a ventilator.
It took days to get him off the machine. Following an exhaustive workup, including a bronchoscopy, no cause for the pulmonary issue was revealed. A diagnosis of adult-onset asthma was finally given, and he was eventually discharged on high-dose steroids and bronchodilators. Was this sudden onset of a brutal asthmatic attack in any way related to his lymphoma? In the absence of a history of lung problems, an association had to be considered, but a definitive answer was not possible. It was only a year later that the asthma was retroactively rediagnosed as a paraneoplastic manifestation of his primary cancer.
The ferocity of the brutal symptoms Harvey experienced resulted from a combat between lymphoma and a misdirected immune system, burning and blasting its way through his ravaged body in episodes that would last a few days or sometimes weeks, followed by an eerie calm, leaving him spent and exhausted in a way that no physical activity could possibly do. In November of 1999, we were in Manhattan for a brief meeting. We were staying at the Plaza, and Harvey had been excited to take Sheherzad to the Central Park Zoo. As he began to dress the next morning, he abruptly sat down and clutched his left calf. “Must be a cramp,” he said. By now, I assigned everything he experienced to the lymphoma, and he would lose his patience with me because he did not want to be constantly reminded of the diagnosis. By the time we landed back in Chicago, he was visibly limping. I forced him to see the internist. An ultrasound revealed deep-vein thrombosis, or DVT, in the calf. The universal verdict between his oncologist, pulmonologist, rheumatologist, and endocrinologist was that the DVT, night sweats, and migratory polyarthritis, even the asthma diagnosed the year before, were interconnected. Harvey’s symptoms could have been because of the lymphoma traveling throughout the body, causing local reactions extending from skin to lungs. But the same sorts of symptoms are seen with solid tumors confined to their organs of origin. How do we explain these?
Paraneoplastic syndromes can sometimes be the first presenting symptom of an unsuspected malignancy. They can affect any system or organ. They are tissue agnostic. David Ansari described the history of our knowledge of these syndromes by tracing the curious association between pancreatic cancer and thromboses.
The Manchester surgeon Charles White first demonstrated in 1784 that “milk leg” was not caused by retained milk or lochia, but rather by obstructing clots in the veins. In 1847, the German Rudolph [sic] Virchow [1821–1902] observed that venous thrombi often migrated to the lungs. In 1865, the French physician Armand Trousseau [1801–1867] described that migratory venous thromboses occurring during the course of his own pancreatic cancer. It has after that for a long time been considered a “truth” that carcinoma of the pancreas has an inherent and unique ability to induce a hypercoagulable diathesis that leads to clinically significant thrombosis. This has, however, later on been challenged and there have been voices stating that the relationship between cancer of the pancreas and thromboembolic disorders should be de-emphasized since it is neither unique nor especially in association with pancreatic carcinoma, and since it may be almost as frequently encountered in other visceral malignancies.
Harvey’s experience was also one of the many reminders that whatever the cause, cancer as a disease is more than a tumor confined to one organ. If not the primary tumor itself, then the immune reactions to cancer in the body can affect any system with cryptic, unanticipated displays, sometimes more painful than the tumor itself. Elimination of the underlying cancer is the only permanent treatment option. All other attempts are palliative and symptomatic to reduce pain and inflammation.
Night sweats are reminiscent of infectious episodes and suggest involvement of the immune system and the release of cytokines, a type of protein vital to the body’s way of responding to and fighting cancer. The body knows something is wrong. It mounts a ferocious immune reaction. Cancer cells escape the wrath by either expressing a signal on their surface that says, “Don’t eat me,” or cloaking the signal that says, “Eat me.” The immune response ends up causing more harm by damaging normal tissues instead of eliminating the cancer cells. The immune system is not always subdued in cancer patients but rather overactive, and it sometimes manifests both over- and underactivity.
Harvey’s life-threatening problems resulted from a weak
ened response of the immune system to repeated bouts of infections that landed him in the hospital several times a month during the last year of his life. He was given regular infusions of intravenous immunoglobulins to boost his immunity. At the same time, with the distressing night sweats and intensely painful polyarthritis, he manifested signs of an overactive, erratic immune response to the lymphoma. It is difficult to reconcile the ideas of a simultaneously underactive and overactive immune response. One possibility is that the cancer is masquerading as friend, fooling the immune system, but only partially. Another idea is that the syndromes result from chemicals and proteins secreted by the tumors and carried around the body via blood, setting up reactions in susceptible tissues. The converse is also possible in that the cancer itself arose because of a flawed immune system in the first place. And if cancer is a consequence, then what systemic changes in the body made the environment more hospitable to the survival of a mutated, transformed, malignant cell? Could it be the inflammatory response of an overreactive immune system?
A reductionist approach is the driving force for advances in the biomedical field, but at the expense of devaluing individual experience. Cancer, the disease in individual cases, can be a multi-organ illness even as the malignant cells remain confined to an organ. Only in its earliest manifestations is it defined by, or limited to, the properties of its individual components. The cause of remote, far-flung, body-wide effects of cancer cannot be traced back to the malfunction of individual malignant cells alone. Rather, single entities interacting with each other and with host defenses produce unpredictable complex behaviors as a collective. Immune cells seem not to recognize the cancer as alien and fail to eliminate it. But they can’t seem to ignore the cancer either, at least in some patients. The fired-up, activated immune system misses its real target, hurting the host more than the cancer. Depending upon what becomes the target of the immune attack, a bizarre panoply of paraneoplastic syndromes are experienced by patients. It is more like two states of water manifesting an unexpected emergent property. Upon freezing, water becomes ice. There is no change in the molecular components of water in liquid or solid form, so what accounts for the slipperiness of ice? The sum of individual parts cannot explain the complexity that emerges from the whole. Paraneoplastic syndromes seemed like an emergent property of cancer.