Why does cancer recur and metastasize? The answer reveals a new strategy for conquering malignant tumors

There are still many mysteries in the battle between cancer cells and the immune system. Every exploration brings new hope.

By Samuel F. Bakhoum (Memorial Sloan Kettering Cancer Center)

Compilation | Kestrel

It has been seven years since the incident, but the feeling of loss at that time is still clear.

One day in 2015, I received the DNA sequencing results of a patient's tumor tissue. Her lung cancer had metastasized to the brain, but I couldn't find any genetic changes in the report that could indicate a treatment target. At the same time, another thing caught my attention: the data showed that almost every chromosome had undergone many changes in structure and number.

Normal cells have two copies of each chromosome, but the chromosomes in cancer cells are different. The number of copies varies, from as few as one to as many as five or six, and sometimes chromosome fragments appear. This phenomenon of genomic disorder is called "chromosome instability," or CIN for short. Although cancer researchers and clinicians have long known that CIN is a sign of advanced cancer or metastasis, I have never seen such a serious case like this patient, who was only 59 years old, not to mention that she was diagnosed not long ago, and CIN is generally believed to occur gradually over a relatively long time span.

Oncologists are helpless because there is no good targeted therapy for CIN, and patients can only simply undergo multiple rounds of systemic chemotherapy and radiation therapy. My patient is no exception. In addition to the usual side effects, she also has to endure the neurotoxicity of brain radiation therapy, which may cause amnesia and other cognitive deficits.

This dilemma reminds me of my background in cell biology. In the past, my research work during my doctoral studies was to figure out how chromosomes are evenly distributed during normal cell mitosis. This process is complicated, but it happens every day in many tissues. Living organisms have evolved multiple backup mechanisms to ensure that this process does not go wrong. If something goes wrong, cells with abnormal chromosome numbers will be quickly eliminated. But cancer cells are different. They are very tolerant to chromosomal abnormalities, and such large-scale genetic changes are closely related to the progression of the disease. However, we do not know whether CIN plays an active role in the development and metastasis of tumors, and how it works.

A few years ago, I set out to investigate whether CIN drives cancer or is just a side effect of it. I collaborated with Lewis Cantley, then director of the Meyer Cancer Center at Weill Cornell Medicine and now at Dana Farber Cancer Institute. We genetically manipulated a variety of chromosomally unstable, metastatic cancer cells to reduce their CIN levels without affecting their other genetic abnormalities. Remarkably, the cancer cells that lost CIN also lost their ability to metastasize. Another surprise was that we discovered that CIN promotes metastasis by causing chronic inflammation. So, ultimately, it was the body's own immune response that allowed cancer cells to escape from the primary tumor and invade other organs.

In 2018, our work was published in the journal Nature. The results suggest that the step of making the genetic material unstable is itself crucial to the evolution of cancer. This could be the starting point for new treatment ideas: Can we find targets in this step to treat cancers with severe CIN? Can we find ways to stabilize the genome or reduce the chronic inflammation caused by CIN to curb cancer metastasis? Can we modify the immune system to eliminate cells with abnormal chromosome numbers?

To address these critical questions, my laboratory at Memorial Sloan Kettering Cancer Center (MSK or MSKCC) uses an interdisciplinary, cell biology-based approach that combines single-cell genomics, mathematical modeling, and clinical sampling. We believe that by integrating these approaches, we will understand how CIN changes cancer cell behavior and enhances cancer cell fitness to sustain cancer progression. Further, we aim to reveal which cellular pathways allow cancer cells to tolerate CIN and then treat cancer by targeting those pathways.

Also in 2018, Cantley, another colleague, Olivier Elemento, and I co-founded Volastra Therapeutics to complement our academic work on CIN. The company's researchers are developing therapies targeting CIN for a variety of cancers. Through this broad collaboration, we hope to explore and develop new treatments for cancer patients with chromosomal instability.

The overlooked sign of cancer

Since scientists sequenced the first cancer cell genome in 2006, we have been continuously learning more about which genetic changes drive cancer development and progression. Scientists have developed targeted therapies that act on genes that promote tumor development. The idea behind these therapies is that if we can inhibit these genes, the tumor will stop getting worse. Therefore, we need to sequence the genome of tumor tissue to find the cancer-promoting genes in each patient. This step has become a routine practice for many oncologists at MSK. However, when sequencing fails to help us find a target, the limitations of personalized cancer treatment become apparent: there are indeed some successful cases, but for most patients with advanced cancer, the effect is still very limited.

Even if targeted therapies can be used, they may only be effective at first, because tumors evolve. They often evade our drugs. And one of the most powerful weapons of cancer cells is CIN. Every time they divide, their chromosomes are randomly rearranged, thanks to CIN. As a result, errors in chromosome segregation accumulate, causing cells in cancerous tissue to be highly heterogeneous in chromosome composition and copy number. This phenomenon is called "aneuploidy." Indeed, advanced tumors that relapse after multiple rounds of treatment are characterized by highly unstable chromosomes and aneuploidy, and drugs that inhibit single mutant genes are no longer effective for such tumors, even though these drugs once caused the cancer to improve.

Researchers have known for decades that aneuploidy is a hallmark of human cancer, but it was not until 1997 that Christoph Lengauer and Bert Volgelstein of the Johns Hopkins University School of Medicine first demonstrated the role of CIN in promoting heterogeneity in cancer cells. Through their work, it was soon understood that CIN has the potential to stimulate tumor progression and malignancy: CIN can regulate the number of chromosome copies and, therefore, the number of copies of genes on these chromosomes. More recently, Stephen Elledge of Harvard Medical School found that human malignancies do increase the number of copies of chromosomes carrying oncogenes as much as possible and reduce the number of copies of chromosomes carrying tumor suppressor genes to improve their own fitness.

Despite the importance of CIN in human cancer, the laboratory still focuses on gene mutations. The methodological revolution brought about by the next-generation sequencing technology has guided the academic community to focus on the contribution of individual genes to cancer occurrence, and has led to many important discoveries, expanding our understanding of the role of many genes in the process of tumorigenesis. However, this method also has disadvantages. It ignores large-scale chromosomal aberrations and their effects on gene function and cancer cell behavior. Purifying DNA from whole tumor tissue samples and sequencing it can certainly allow us to see the genetic information on chromosomes more clearly, but it is impossible to locate DNA fragments with altered sequences on chromosomes, and it blurs the heterogeneity of chromosome copy number between cells.

Over the past decade, researchers have begun to focus more attention on large-scale chromosomal changes. In 2010, Robert Benezra and others at MSK published an important study showing that tumors that have acquired CIN no longer rely on the oncogenes that initially caused the cancer. Specifically, when researchers upregulated the oncogene KRAS to induce lung cancer in mice and then canceled the upregulation, tumor regression was observed; but if genetic engineering methods were used to further artificially introduce CIN into cancerous cells, tumor regression could not be observed. The so-called targeted therapy is to specifically inhibit oncogenes like KRAS to prevent them from driving tumor development, but malignant tumors will gradually become resistant to this therapy, and this work tells us how resistance is generated.

Since then, Charles Swanton's group at University College London and the Crick Institute has provided strong evidence that CIN is important in human cancer. In 2017, by following up on lung cancer patients, Swanton's team demonstrated that chromosomal instability - rather than the number of point mutations in the tumor genome - is associated with reduced overall survival. Their subsequent studies have shown that CIN is likely to play an important role in all aspects of tumor biology, whether it is tumor metastasis or cancer cells escaping immune surveillance. The gradual change in the number of chromosome copies accompanying each division of cancer cells provides tumors with the ability to evolve under various selection pressures.

As these scientists continue to study the role of CIN in cancer, the boundaries between cancer genomics and cell biology are gradually disappearing. On the one hand, the genetic code carried by chromosomes can be deciphered using complex genomic methods; on the other hand, the chromosome life cycle and separation behavior during cell division can basically be tracked using high-resolution optical microscopy. For example, chromosomes that make mistakes during the separation process will eventually appear in a "micro-nuclei" that contains a small piece of DNA, separated from the cell nucleus, the "main nucleus". Micronuclei have always been regarded as a feature that distinguishes cancer cells from surrounding normal tissues. Work by multiple research groups has shown that the nuclear membrane that envelops micronuclei often breaks, spilling chromosomes into the cytoplasm, exposing them to nucleases and being degraded like other proteins, becoming broken into pieces.

Figure 1. Microscopic photo of micronuclei. The small black dots in the cells are micronuclei.
https://doi.org/10.1016/B978-0-12-800764-8.00006-9)

After large-scale chromosome breakage, some fragments are lost, while others are randomly connected together, and the direction and order of the connection between the fragments may be wrong, forming new, highly abnormal chromosomes. This process is called "chromothripsis". Recently, researchers have found that chromothripsis is an important mechanism that stimulates the deterioration of cancer. In addition to the step-by-step changes in chromosome number, through large-scale chromosome rearrangements, the broken chromosomes quickly form a jackpot that can cause cancer - once cancer occurs, it is already very serious. This process has the potential to rapidly amplify oncogenes and lose tumor suppressor genes. We now also know that chromothripsis may place oncogenes near highly active chromosomal regions and promote the formation of circular extrachromosomal DNA, both of which will prompt tumor cells to quickly develop resistance to targeted therapies.

Although researchers have long recognized the chromosomal chaos in cancer, our understanding of the process of chromothripsis has only become clearer with the advent of cutting-edge genomics and microscopy techniques. In 2015, David Pellman and colleagues at Dana-Farber Cancer Institute used microscopy to capture and genomic profile individual cancer cells with chromosome segregation errors. Using this approach, which they called Look-seq, the researchers showed how the complex patterns of chromosomal rearrangements common in human cancer genomes emerge within a single cell cycle. Through a variety of pathways, CIN appears to promote both gradual and episodic bursts of tumor genome evolution. This is probably what happened to my patient, who had large-scale chromosomal aberrations soon after diagnosis.

Mechanism of micronucleus formation

During cell division, many chromosome segregation errors may lead to micronucleus formation. Even if the chromosomes are not missegregated in the end, micronuclei may still form. For example, in the case shown in Figure 2, although the chromosomes are finally evenly distributed to the two daughter cells, the left daughter cell still forms a micronucleus due to the delayed separation. These events are not mutually exclusive or independent, but each occurrence will aggravate the confusion of chromosomes.

Figure 2. Incorrect attachment.

When microtubules at both poles of a dividing cell attach to the same centriole (top), the attached chromosome lags behind the other chromosomes in separation and is often subsequently encapsulated in a micronucleus, even though it eventually makes it to the cell it is destined for (bottom left).

Figure 2. Aneuploidy.

If chromosome segregation does go wrong, either due to microtubule misattachment or for some other reason, the missegregated chromosome may be enclosed in the nuclear membrane along with the other chromosomes. If it is not enclosed, the resulting aneuploid cell may have a lagging chromosome, increasing the risk of micronucleus formation.

Figure 3. Chromosome fusion.

Shortened or damaged telomeres make chromosome fusion more likely. Such fusion events can produce chromosomes with two centrioles. During the next cell division, the chromosomes with two centrioles will be torn apart and divided into two daughter cells. These damaged chromosomes will either be immediately sequestered into micronuclei or will be sequestered into micronuclei again during the next cell division because they cannot complete replication normally.

CIN and inflammation

When studying how chromosomal instability affects cancer metastasis, we made an unexpected discovery. Specifically, cancer cells that have already developed CIN have activated inflammatory signaling pathways, which can allow cancer cells to produce and secrete a variety of inflammatory factors related to cancer metastasis. This was quite confusing at first, because these cells were only raised in the laboratory and were not planted in experimental animals, so there was no chance of contact with immune cells. So what triggers these "inflammatory responses"?

After searching for a long time under the microscope, we not only observed that micronuclei accounted for the majority of cells with CIN, but also found that those cells containing ruptured micronuclei carried an immune-related enzyme called cGAS. cGAS was first discovered by James Chen of the University of Texas Southwestern Medical Center in 2013. It is a double-stranded DNA receptor located in the cytoplasm. So we imagined that after the micronucleus ruptured, the chromosomes exposed in the cytoplasm might be recognized by cancer cells as danger signals, just as cells recognize invading pathogenic DNA. Of course, we then confirmed that ruptured micronuclei can strongly activate cGAS and its related protein STING, thereby activating the innate immune response. But unlike acute viral infections that can be cleared within a few days, micronucleus rupture events in the cytoplasm of cancer cells follow one after another, causing the inflammatory pathway to remain activated and inflammation to persist.

By now, it was clear that cancer cells must have exploited a protective immune pathway to be able to break through these inflammatory defenses. Although activation of innate immune signaling pathways may protect the body from tumors early in tumor development, at a certain stage, tumor cells are able to bypass these protective mechanisms, tolerate the inflammatory response caused by CIN, and gradually exploit these pathways to drive tumor growth. The ability of cancer cells to sustain inflammation is crucial to their ability to metastasize to another organ. Immune cells are the most mobile cells in the body. Within hours of an infection or wound, they can migrate through the vascular system to inflamed tissues with increased hydrostatic pressure to reach damaged areas. Cancer cells use CIN and other genomic abnormalities to mimic this physiological process to achieve metastasis.

The association between chronic inflammation and cancer is long-standing. In fact, the core characteristics of inflammation described by the ancient Roman encyclopedist Aulus Cornelis Celsus also apply to cancer - redness, heat, pain and swelling. For hundreds of years, clinicians have often referred to tumors as "unhealing wounds" because they are constantly inflamed. We are not yet fully aware of the role of inflammatory signaling pathways in the development of cancer, but by linking intrinsic genomic abnormalities (such as chromosomal instability) with persistent inflammation in cancer, we can find that CIN can not only drive the genetic heterogeneity of tumors, but also stimulate cancer metastasis through non-genetic mechanisms (i.e., simulating inflammatory responses).

How does micronucleus disruption promote cancer development?

Figure 4. Micronucleus disruption promotes cancer development.

The nuclear membrane of the micronucleus is very fragile and often breaks, causing the chromosomes to scatter into the cytoplasm. The chromosomes in the cytoplasm are cut into small fragments by nucleases, and these fragments are either lost, randomly connected together, or connected end to end to form circular extrachromosomal DNA. This process is called "chromosome fragmentation". The complex rearranged chromosomes formed by this process can drive the formation of tumors.

At the same time, DNA remaining in the cytoplasm triggers the cGAS-STING inflammatory pathway. It is generally believed that this pathway evolved from a mechanism to resist viral infection. cGAS binds to the DNA scattered by the rupture of micronuclei and catalyzes the generation of 2'3'-cyclic guanosine monophosphate (cGAMP), thereby activating the STING protein and downstream inflammatory pathways. Since there are many micronuclei in cancer tissues, it is possible that this pathway is always activated, triggering a persistent inflammatory response and driving tumor growth and metastasis.

Targeting CIN

Unlike cancer cells, normal cells cannot tolerate chromosome segregation errors. Research led by the late Angelika Amon of MIT found that aneuploidy is associated with multiple cellular defects, such as metabolic disorders and mitochondrial dysfunction, as well as cellular stress responses induced by protein misfolding. In fact, the human body has evolved multiple mechanisms that can eventually eliminate aneuploid cells. Duan Compton and others at the Geisel School of Medicine at Dartmouth found that in response to chromosome segregation errors, normal cells quickly activate the tumor suppressor p53, stop cell division, and prevent the spread of aneuploid cells. These important protective mechanisms are designed to maintain the integrity of the genome, and generally, with them, the genome is fine. But in cancer cells, these lines of defense are breached. Therefore, understanding how tumor cells respond to CIN may provide inspiration for cancer treatment.

There is growing interest in the mechanisms by which tumor cells tolerate CIN. Recently, several groups have discovered through genetic screening that some genes and some cellular activities are essential for tumor cells with high levels of CIN, and their deficiency is lethal. One of them is the kinesin Kif18a, which plays a role during mitosis and chromosome movement. For cancer cells with CIN, the Kif18a protein is essential for cell division, but not for cancer cells without CIN. Interestingly, if a mouse lacks a functional Kif18a protein, it can still survive, but only shows some minor defects. Then this kinesin may become a safe and effective therapeutic target. There is now a phase 1 clinical trial testing the efficacy of Kif18a inhibitors in patients with advanced cancer.

Several research groups are exploring another therapeutic strategy, namely, inhibiting targets that allow tumor cells to overcome chronic inflammation. For example, the ENPP1 protein was first discovered by Timothy Mitchison of Harvard Medical School and Lingyin Li of Stanford University (also at Harvard at the time). Later, our group found that it was selectively upregulated in cancer cells with chromosomal instability. The ENPP1 protein is an enzyme located on the outer surface of cancer cells that can degrade the signaling molecule cGAMP that stimulates immune response. After the extracellular cGAMP is degraded, immune cells can no longer detect cancer cells. The degradation of cGAMP can also produce adenosine, which in turn exacerbates immune disorders and promotes cancer cell migration. The ability of cancer cells to turn enemies into friends and use the immune system's protective mechanisms for their own benefit has impressed us.

Volastra researchers are continuing to gain a deeper understanding of the biological mechanisms of CIN, and by combining computational and genetic screening methods, they are slowly discovering some cancer treatment strategies. The current lead candidate drug targets the process of microtubule binding to chromosomes, which can selectively kill cancer cells with chromosomal instability without harming other cells. The drug is planned to be launched into clinical trials in 2023. Other treatment strategies we are exploring include: regulating spindle formation, changing the organization of chromosomes during cell division, and using CIN-driven inflammatory responses to fight cancer.

Finding pathways associated with CIN that can be targeted therapeutically is exciting and will help to break the barrier of drug resistance for some cancers. CIN is an attractive drug target because only cancer cells show chromosomal instability, so treatment targeting CIN will not harm innocent patients - this is the holy grail of cancer treatment. Over the past decade, the fields of cell biology, genomics and cancer biology have become increasingly intertwined, and interdisciplinary research methods and collaborations between academia and industry will lead to new discoveries. The ultimate goal of all this is to serve patients - patients like me whose tumors have metastasized to the brain, and other patients who have very limited current treatment options.

How Chaos in Chromosomes Helps Drive Cancer Spread

Read the original article:

https://www.the-scientist.com/features/how-chaos-in-chromosomes-helps-drive-cancer-spread-69695

Special Tips

1. Go to the "Featured Column" at the bottom of the menu of the "Fanpu" WeChat public account to read a series of popular science articles on different topics.

2. Fanpu provides a function to search articles by month. Follow the official account and reply with the four-digit year + month, such as "1903", to get the article index for March 2019, and so on.

Copyright statement: Personal forwarding is welcome. Any form of media or organization is not allowed to reprint or excerpt without authorization. For reprint authorization, please contact the backstage of the "Fanpu" WeChat public account.