In 1971, President Nixon signed the National Cancer Act and declared war on cancer. He expected to win within a decade. Over fifty years and hundreds of billions of dollars later, cancer remains the second leading cause of death in the United States, killing approximately 600,000 Americans annually. The failure was not for lack of effort or resources. It reflected a fundamental misunderstanding of what cancer is.

Cancer is not a foreign invader like a pathogen. It is not a malfunction of a single system that can be repaired by a single intervention. It is an evolutionary process playing out inside a human body — cells exploiting the same darwinian logic that shapes species, adapting under selection pressure, diversifying, colonizing new territories. Understanding cancer means understanding evolution at the cellular scale.

The insights that have transformed cancer biology over the last thirty years — the molecular mechanisms, the hallmarks framework, the tumor microenvironment, the immune evasion strategies — have not yet cured cancer. But they have produced treatments that were inconceivable in 1971, and they have given researchers, for the first time, a coherent theory of what they are fighting.

"We now know that cancer is, in essence, a disease of the genome." — Bert Vogelstein, Johns Hopkins University

Key Definitions

Proto-oncogene — A normal gene that promotes cell growth and division, functioning properly in healthy tissue. When mutated or overexpressed, it becomes an oncogene — producing constant pro-growth signals even in the absence of external stimulation. Examples: RAS (mutated in ~30% of all cancers), MYC, HER2. Named by analogy to an accelerator pedal stuck in the pressed position.

Tumor suppressor gene — A gene that normally constrains cell growth, triggers repair of DNA damage, or initiates programmed cell death. Cancer requires inactivation of tumor suppressors to proceed. Examples: TP53 (mutated in ~50% of all human cancers, the most commonly mutated gene in cancer), RB1 (retinoblastoma), BRCA1/2 (DNA repair). Named by analogy to a malfunctioning brake.

Apoptosis — Programmed cell death — a genetically encoded suicide program that eliminates damaged, infected, or otherwise unwanted cells. Activated by severe DNA damage or by detection signals from neighboring cells. Cancer cells must evolve mechanisms to resist apoptosis to persist despite accumulating mutations.

Angiogenesis — The growth of new blood vessels. Tumors larger than approximately 1-2mm require their own blood supply; they achieve this by secreting VEGF (vascular endothelial growth factor) and other angiogenic factors that induce local vessel formation. Antiangiogenic drugs (bevacizumab/Avastin) attempt to starve tumors by blocking this process.

Metastasis — The spread of cancer cells from the primary tumor to distant organs via the bloodstream or lymphatic system, establishing secondary tumors (metastases). Metastasis accounts for over 90% of cancer deaths and represents the disease's most dangerous capability.

Tumor microenvironment (TME) — The non-cancerous cellular and molecular ecosystem surrounding a tumor, including immune cells, blood vessels, fibroblasts, and extracellular matrix. Profoundly influences tumor behavior, treatment response, and immune evasion.

Clonal evolution — The darwinian process by which cancer develops: a cell acquires a mutation that provides a growth advantage, expands clonally, and the expanded population is subjected to further selection for additional mutations. Cancer is therefore not a single event but an ongoing evolutionary process.

Driver mutation — A mutation that confers a growth advantage on the cell carrying it, actively contributing to cancer development. Distinguished from passenger mutations, which are neutral mutations that accumulate in dividing cells but do not contribute to oncogenesis. Identifying driver mutations is central to targeted therapy development.

The Warburg effect — Cancer cells preferentially use glycolysis (anaerobic fermentation of glucose to lactate) for energy even in the presence of oxygen, rather than the more efficient oxidative phosphorylation used by normal cells. First described by Otto Warburg in 1927. Exploited by PET scanning, which detects high glucose uptake in tumors.

The Origin of Cancer: Accumulating Mutations Over Decades

Cancer does not typically develop from a single catastrophic event. It develops from the accumulation of multiple mutations over months to decades — a slow biological arms race between the cell's growth-control systems and the genetic damage that accumulates with every cell division and every carcinogen exposure.

The human body contains approximately 37 trillion cells. Each cell division requires copying 3 billion base pairs of DNA. Despite highly accurate proofreading machinery (error rate approximately 1 in 10 billion base pairs per replication), the sheer scale of cellular turnover — approximately 3.8 million cell divisions per second across the body — means mutations are constantly arising. Most are repaired immediately. Some are not.

The Two-Hit Hypothesis

In 1971, Alfred Knudson analyzed the epidemiology of retinoblastoma, a childhood eye cancer, and proposed what became one of the foundational insights of cancer biology: the two-hit hypothesis.

Knudson noticed that children with hereditary retinoblastoma developed tumors much earlier, and often bilaterally, compared to children with sporadic retinoblastoma. He proposed that cancer requires two mutations to inactivate a tumor suppressor gene (since we inherit two copies of each gene — one from each parent). Children who inherited one already-mutated copy needed only one additional mutation ("hit") to lose function entirely. Children without the inherited mutation needed two de novo mutations — a much rarer event, explaining the later onset of sporadic cases.

The prediction was confirmed when the retinoblastoma gene (RB1) was cloned in 1986, establishing the template for understanding tumor suppressor inactivation across all cancer types.

How Many Mutations Does It Take?

The number of driver mutations required for a cancer to develop depends on the cancer type and the tissue. Vogelstein's colorectal cancer model, developed over decades of sequencing studies, identified an average of four to five major driver mutations required for colon cancer development: an early APC mutation (founding the adenomatous polyp), followed by KRAS, SMAD4, TP53, and others over a 15-20 year progression from polyp to invasive cancer.

This multi-step model has three important implications:

First, cancer takes time — which is why it is primarily a disease of aging. The probability of accumulating enough mutations rises with each decade.

Second, early detection windows exist — colon cancer can be caught as a polyp years before it becomes invasive, which is why colonoscopy screening dramatically reduces colon cancer mortality.

Third, cancer can potentially be interrupted at multiple stages — not just treated after it becomes fully malignant.

The Hallmarks of Cancer: A Common Logic Beneath Two Hundred Diseases

There are approximately 200 distinct human cancer types, arising in different tissues with different genetics, different behaviors, and different treatment responses. What unifies them?

In 2000, Douglas Hanahan and Robert Weinberg published "The Hallmarks of Cancer" in the journal Cell — a paper that synthesized two decades of molecular discoveries into a coherent framework. Updated in 2011 and comprehensively revised in 2022, it remains the most cited paper in cancer biology. The framework identifies the biological capabilities that every cancer must acquire to develop and spread.

The Core Hallmarks

1. Sustaining proliferative signaling. Normal cells require external growth signals to divide. Cancer cells generate their own, typically through constitutively active oncogenes. The RAS protein is the most prominent example: mutated RAS (found in 30% of all cancers) continuously transmits 'divide' signals downstream regardless of whether external growth factors are present. It is, in the cellular sense, a perpetual motion machine.

2. Evading growth suppressors. The cell cycle is regulated at multiple checkpoints by proteins like RB1 and TP53 that halt progression when conditions are wrong. Cancer cells disable these checkpoints. TP53 mutation is so central to cancer development that p53 has been called "the guardian of the genome" — it is the last line of defense against cells with unrepaired DNA damage attempting to divide.

3. Resisting cell death. When cells sustain severe DNA damage, the normal response is apoptosis — cellular suicide. Cancer cells evolve resistance to apoptosis through overexpression of anti-apoptotic proteins (BCL-2, BCL-XL) or downregulation of pro-apoptotic proteins (BAX, PUMA). The discovery that follicular lymphoma invariably overexpresses BCL-2 (from the t(14;18) chromosomal translocation) led directly to the development of venetoclax, a BCL-2 inhibitor that restored apoptosis sensitivity in lymphoma and CLL.

4. Enabling replicative immortality. Normal cells can divide approximately 50 times (the Hayflick limit) before telomere shortening triggers senescence. Cancer cells bypass this by reactivating telomerase — the enzyme that maintains telomere length — present in embryonic cells but silenced in most adult tissues. Without telomerase activation, cancer cells cannot divide indefinitely.

5. Inducing angiogenesis. A tumor larger than approximately 1-2mm cannot be sustained by oxygen diffusion alone. To grow beyond a microscopic cluster, cancer must recruit blood vessels. Tumors do this by tilting the balance of angiogenic signals: secreting VEGF and FGF2 while suppressing angiogenesis inhibitors. The resulting tumor vasculature is chaotic and poorly organized — a feature that creates the hypoxic microenvironment that further drives aggressive cancer behavior.

6. Activating invasion and metastasis. This hallmark encompasses the dramatic process by which cancer cells escape the primary tumor, enter circulation, survive transit, and establish distant colonies — the process responsible for most cancer deaths.

7. Reprogramming energy metabolism (Warburg effect). Cancer cells preferentially use aerobic glycolysis rather than mitochondrial oxidative phosphorylation even when oxygen is available. Glycolysis is far less efficient (2 ATP per glucose vs. 36 ATP via oxidative phosphorylation) but generates metabolic intermediates that support rapid biosynthesis of proteins, lipids, and nucleic acids required for cell proliferation. PET scanning exploits this: FDG (radiolabeled glucose) accumulates preferentially in metabolically active tumors.

8. Evading immune destruction. The immune system continuously surveys for and eliminates abnormal cells. That cancer develops at all reflects cancer cells' ability to escape this surveillance — through MHC-I downregulation, checkpoint ligand expression, and creation of an immunosuppressive microenvironment.

9 and 10. Tumor-promoting inflammation and genome instability. The 2011 revision added these "enabling characteristics" — factors that facilitate acquisition of the core hallmarks. Chronic inflammation creates a mutagenic environment rich in reactive oxygen species and growth-promoting cytokines. Genome instability (from mutations in DNA repair genes) accelerates the mutation rate, speeding the evolutionary acquisition of cancer hallmarks.

Metastasis: How Cancer Kills

More than 90% of cancer deaths result not from the primary tumor but from metastasis — cancer cells that have left the original tumor, survived transport through the body, and established colonies in distant organs.

The metastatic cascade is an extraordinarily inefficient process. Studies introducing labeled cancer cells into the circulation find that fewer than 0.01% successfully establish distant metastases. The cells that succeed are those that have evolved a specific suite of additional capabilities beyond the hallmarks of primary tumor growth.

The Steps of Metastasis

Epithelial-mesenchymal transition (EMT). Most carcinomas originate in epithelial cells — cells that adhere tightly to each other and to a basement membrane, forming the sheets and tubes that line organs. To invade, these cells must switch to a mesenchymal phenotype: losing E-cadherin (the adhesion molecule that glues epithelial cells together), gaining migratory capabilities, and secreting matrix metalloproteinases (MMPs) that digest the extracellular matrix.

Intravasation and survival in circulation. Cancer cells that enter blood vessels face anoikis (death from loss of anchorage), mechanical shear forces, and attack from NK cells. Most die. Those that survive often do so in clusters — circulating tumor cell clusters are dramatically more efficient at metastasis than single cells.

Arrest and extravasation. Circulating tumor cells tend to arrest in the first capillary bed they encounter that is too narrow to traverse — explaining why colon cancer preferentially metastasizes to liver (portal circulation), and lung cancer to brain and adrenal glands. But the pattern is not purely mechanical: certain cancers have molecular affinities for specific tissues (Paget's "seed and soil" hypothesis from 1889, vindicated by modern molecular biology).

Colonization. This is the rate-limiting step. Arriving at a foreign organ and successfully establishing a growing colony requires the cancer cell to survive in an unfamiliar cellular niche, suppress the local immune response, and induce supportive stromal remodeling. Many disseminated tumor cells remain dormant for years or decades before either dying or reactivating — a phenomenon of immense clinical importance since it explains late recurrence in apparently cured patients.

The Tumor Microenvironment: Cancer as Ecosystem

A tumor is not simply a mass of cancer cells. It is a complex ecosystem — what is now called the tumor microenvironment (TME) — containing cancer cells, immune cells (T cells, macrophages, NK cells, regulatory T cells), fibroblasts, blood and lymphatic vessels, and extracellular matrix.

The TME is often the difference between a tumor that is contained and one that is lethal.

Immune Cells in the TME

Tumors are typically heavily infiltrated by immune cells — but these infiltrates frequently do not destroy the tumor. Instead, many of the immune cells are co-opted or dysfunctional:

Tumor-associated macrophages (TAMs) are typically polarized toward an anti-inflammatory M2 phenotype rather than the pro-inflammatory M1 phenotype that would kill cancer cells. M2 macrophages supply growth factors, matrix-remodeling enzymes, and immunosuppressive cytokines that promote tumor progression. High TAM infiltration correlates with poor prognosis in most cancer types.

Regulatory T cells (Tregs) infiltrate many tumors and suppress effector T cell function, preventing anti-tumor immunity. Treg depletion improves outcomes in mouse tumor models.

Exhausted T cells. T cells that recognize tumor antigens but are chronically exposed to antigen in the suppressive TME develop an exhausted phenotype — still present but dysfunctional, expressing PD-1, TIM-3, LAG-3, and TIGIT. This exhaustion program is reversible: it is exactly what checkpoint inhibitor drugs target when they block PD-1/PD-L1 or CTLA-4 interactions.

Immune Evasion and the Checkpoint Revolution

The modern immunotherapy revolution was built on understanding how cancer evades the immune system.

In the 1990s and 2000s, James Allison (MD Anderson) and Tasuku Honjo (Kyoto University) worked independently on immune checkpoints — molecular brakes on T cell activation that normally prevent autoimmunity. Allison focused on CTLA-4; Honjo on PD-1. Both recognized that cancer was co-opting these checkpoints to suppress anti-tumor immunity.

Allison developed an anti-CTLA-4 antibody (ipilimumab). Early clinical trials in metastatic melanoma — a cancer with a median survival of 8-9 months — produced something that had rarely been seen in oncology: durable complete responses. Approximately 20% of patients who received ipilimumab were alive at 10 years — in a disease that was almost uniformly fatal within two years.

Anti-PD-1 drugs (pembrolizumab, nivolumab) proved even more broadly effective, with better tolerability. Combined anti-CTLA-4 + anti-PD-1 therapy produced response rates of ~50% in melanoma.

The checkpoint revolution spread to other cancer types. In non-small-cell lung cancer — historically with poor outcomes — PD-L1-high tumors treated with pembrolizumab alone showed 5-year survival rates of approximately 30-35% compared to historical rates of ~5% for stage IV disease.

Allison and Honjo shared the 2018 Nobel Prize in Physiology or Medicine.

The limitation of checkpoint immunotherapy is that it only works when there are pre-existing tumor-infiltrating lymphocytes to reinvigorate — "cold" tumors with sparse immune infiltration respond poorly. Understanding why some tumors are "hot" and others "cold" is a central challenge of current cancer immunology.

The Cancer Genome: Mutation Signatures and Tumor Evolution

Whole-genome sequencing of cancers has revealed that each cancer genome carries a characteristic pattern of mutations — a "mutational signature" reflecting the specific mutational processes that acted on it.

The COSMIC (Catalogue of Somatic Mutations in Cancer) database has catalogued over 80 distinct mutational signatures. Signature 4, characterized by C→T and CC→TT substitutions at specific contexts, is the signature of tobacco smoking — identifiable in virtually every lung cancer from smokers. Signature 7 (C→T at dipyrimidine sites) reflects UV radiation damage in melanomas. Signature 3 reflects defective homologous recombination repair — a hallmark of BRCA1/2-mutated tumors.

These signatures have clinical implications. Tumors with a high mutational burden (many mutations, producing many neoantigens for T cells to target) tend to respond better to checkpoint immunotherapy — which is why microsatellite-unstable tumors (with defective DNA mismatch repair) respond well to anti-PD-1 therapy across multiple cancer types, leading to the first tumor-agnostic FDA approval.

Intratumoral Heterogeneity

One of the most challenging recent insights is that tumors are not genetically uniform. A single tumor contains multiple genetically distinct subclones — a tree structure of clonal evolution where the founding clone has diversified into branches carrying different mutations.

This heterogeneity explains treatment resistance: a drug targeting mutation X will kill subclones harboring that mutation but will spare subclones that have already evolved an alternative pathway. The surviving subclone then repopulates the tumor — resistant to the original drug.

Charles Swanton's TRACERx study, following lung cancer patients through treatment and recurrence by sequencing multiple tumor regions and circulating tumor DNA, has documented this evolutionary branching in real time, demonstrating that metastases can arise from different subclonal branches of the primary tumor.

Prevention: What Actually Reduces Cancer Risk

Cancer's complexity does not mean it is unpreventable. Epidemiological estimates suggest that approximately 40-50% of cancers are attributable to modifiable risk factors.

Risk Factor Estimated % of Cancer Deaths Cancers Primarily Affected
Tobacco smoking ~30% Lung, bladder, kidney, head/neck, esophagus
Obesity and overweight ~8-10% Colon, breast, endometrial, liver, kidney, esophageal
Excess alcohol ~4-6% Liver, breast, colon, esophagus, head/neck
Infections (HPV, H. pylori, HBV/HCV) ~4-6% Cervical, gastric, liver
UV radiation ~4% Melanoma, skin cancers
Physical inactivity ~2-3% Colon, breast, endometrial

The World Cancer Research Fund's continuous update project has identified the strongest dietary/lifestyle risk factors: processed meat (Class 1 carcinogen, ~18% increased colon cancer risk per 50g/day), red meat (Class 2A), alcohol (Class 1 carcinogen for multiple cancer types), body fatness, and physical inactivity.

Screening interventions with demonstrated mortality reduction: colonoscopy and fecal immunochemical testing (FIT) for colorectal cancer; low-dose CT for heavy smokers (NLST trial: 20% lung cancer mortality reduction); mammography for breast cancer (efficacy debated, ~15-20% mortality reduction); HPV vaccination (near-complete prevention of HPV-attributable cancers when administered before exposure); hepatitis B vaccination (prevents HBV-associated hepatocellular carcinoma).

Why Cancer Is So Hard to Cure

Understanding the hallmarks, the TME, intratumoral heterogeneity, and the evolutionary dynamics of cancer makes the challenge clear: there is no single intervention that can address a disease with ten enabling capabilities, evolving in real time under the selection pressure of treatment.

The history of targeted therapy illustrates the problem. Imatinib (Gleevec), which targets the BCR-ABL fusion oncogene in chronic myelogenous leukemia, was heralded as a miracle drug when it produced ~95% major cytogenetic response rates in 2001 — compared to ~30% with previous treatments. For most CML patients, it remains highly effective with continuous treatment. But BCR-ABL is unusually targetable: it is the single defining event of CML, with limited subclonal heterogeneity. When similar targeted therapies were applied to cancers with more complex genomic landscapes (lung cancer with EGFR mutations, colorectal cancer with KRAS), resistance emerged within months as subclones bearing resistance mutations expanded.

The field is increasingly moving toward combination approaches — targeting multiple pathways simultaneously to prevent escape — and toward harnessing the adaptive immune system, which can theoretically generate responses against the full heterogeneous population of a tumor.

For related articles, see how the human immune system works, how genetic engineering works, what causes chronic inflammation, and why we age.

References

Tags

cancer biology oncology genetics tumor cell biology health

Frequently Asked Questions

What actually causes cancer at the cellular level?

Cancer develops when a cell accumulates mutations in specific categories of genes that normally regulate cell growth, division, and death. The key categories are: proto-oncogenes (normal genes that promote cell growth, which become oncogenes when mutated, causing cells to receive constant 'grow and divide' signals even without external stimulation — like a stuck accelerator); tumor suppressor genes (genes that put the brakes on cell division, such as TP53 [the 'guardian of the genome,' mutated in ~50% of all human cancers] and BRCA1/2 [involved in DNA repair], which when mutated remove critical growth constraints); and DNA repair genes (which normally fix mutations before they become permanent, so when they malfunction, mutation rates accelerate dramatically). Critically, cancer does not typically develop from a single mutation — it requires the accumulation of multiple driver mutations, typically six to ten, over months to decades. This is why cancer risk rises steeply with age: more cell divisions mean more copying errors. A cell dividing in 2024 carries accumulated mutations from every division its ancestors made since 1950. The sources of these mutations are diverse: replication errors (inherent copying mistakes during cell division), carcinogen exposure (tobacco smoke, UV radiation, certain chemicals), viral DNA integration (HPV, hepatitis B/C, EBV), and inherited predispositions (germline mutations in tumor suppressors like BRCA1 that start the clock earlier).

What are the hallmarks of cancer and why do they matter?

In 2000, Robert Weinberg and Douglas Hanahan published what became the most cited paper in cancer biology, 'The Hallmarks of Cancer,' identifying the shared capabilities that all cancers must acquire. Updated in 2011 and again in 2022, the current framework identifies ten hallmarks: (1) Sustaining proliferative signaling — cancer cells generate their own growth signals rather than waiting for external cues; (2) Evading growth suppressors — disabling tumor suppressors like RB and TP53 that would normally halt growth; (3) Resisting cell death — blocking apoptosis (programmed cell death), the mechanism that normally eliminates damaged cells; (4) Enabling replicative immortality — co-opting telomerase to bypass the Hayflick limit of approximately 50 cell divisions that normal cells respect; (5) Inducing angiogenesis — recruiting new blood vessel formation to feed the growing tumor; (6) Activating invasion and metastasis — acquiring the ability to detach, migrate through tissue, enter blood vessels, and establish distant colonies; (7) Reprogramming energy metabolism — shifting to aerobic glycolysis even in the presence of oxygen (the Warburg effect, first described 1927); (8) Evading immune destruction — escaping surveillance by T cells and NK cells; (9) Tumor-promoting inflammation — co-opting inflammatory cells to supply growth factors and matrix-remodeling enzymes; (10) Genome instability and mutation — accelerating the mutation rate to speed up evolution. The hallmarks framework matters because it explains why cancer is so difficult to treat — it is not a static disease but an evolving ecosystem with multiple redundant pathways.

How does cancer spread (metastasize) to other parts of the body?

Metastasis — the spread of cancer to distant organs — is responsible for over 90% of cancer deaths and is the process that makes cancer genuinely dangerous rather than a localized problem. Metastasis is a multi-step process: (1) Local invasion: cancer cells must first break out of their original tissue compartment, dissolving the extracellular matrix using enzymes called matrix metalloproteinases (MMPs) and losing E-cadherin, the cellular adhesion molecule that keeps epithelial cells attached to their neighbors (this transition is called epithelial-mesenchymal transition, or EMT); (2) Intravasation: cells must enter the bloodstream or lymphatic system, a process that normally triggers anoikis (death from loss of anchorage) — metastatic cells must evolve resistance to anoikis; (3) Survival in circulation: circulating tumor cells travel through vessels where they face mechanical shear stress, immune attack from NK cells, and anoikis — very few survive (estimates suggest 1 in 10,000 entering circulation successfully metastasizes); (4) Extravasation: surviving cells must exit the bloodstream at a distant site; (5) Colonization: this is the major bottleneck — arriving at a new tissue and establishing a viable tumor colony. Different cancers have characteristic metastatic destinations: prostate cancer preferentially seeds bone; colon cancer seeds liver; lung cancer seeds brain and adrenal glands. These patterns reflect both biomechanical factors (circulatory anatomy) and molecular affinities between cancer cells and the target organ's microenvironment — what Stephen Paget called 'seed and soil' in 1889.

What is the tumor microenvironment and why does it matter?

A tumor is not simply a mass of cancer cells — it is a complex ecosystem that includes cancer cells, immune cells, fibroblasts, blood vessels, lymphatic vessels, and extracellular matrix, all communicating through growth factors, cytokines, and metabolites. This surrounding context is the tumor microenvironment (TME), and it profoundly influences whether cancer progresses, responds to treatment, or is contained. The TME is often immunosuppressive: regulatory T cells (Tregs), tumor-associated macrophages (TAMs, typically polarized to an anti-inflammatory M2 phenotype), and myeloid-derived suppressor cells (MDSCs) dampen anti-tumor immunity. Cancer cells actively shape this immunosuppressive environment by secreting TGF-beta, IL-10, and VEGF, and by expressing checkpoint ligands like PD-L1, which binds PD-1 on T cells to functionally 'switch them off' — the exact mechanism exploited by checkpoint immunotherapy drugs like pembrolizumab (Keytruda) and nivolumab (Opdivo). Cancer-associated fibroblasts (CAFs) produce collagen and growth factors that structurally support tumor growth and create a stiff, hypoxic environment that further suppresses immune function. Understanding the TME revolutionized oncology: checkpoint immunotherapy drugs that reinvigorate tumor-infiltrating lymphocytes (TILs) have produced durable remissions in cancers that were previously rapidly fatal, including metastatic melanoma and non-small-cell lung cancer.

How does cancer evade the immune system?

The immune system surveys the body for abnormal cells continuously — a process called immunosurveillance. That cancer exists at all in immunocompetent individuals demonstrates that cancer cells can evolve immune evasion. The mechanisms are diverse and constitute an active area of research. Downregulation of MHC class I: cancer cells reduce surface expression of MHC-I, the molecular 'display case' that shows T cells what a cell is manufacturing internally. When MHC-I is low, cytotoxic T cells cannot recognize and kill the cell — but this simultaneously makes the cell visible to NK cells, which kill cells lacking MHC-I (an immunosurveillance trade-off). Checkpoint exploitation: PD-L1 expression on tumor cells binds PD-1 on T cells, delivering an 'off' signal that exhausts or anergizes T cells. Similarly, CTLA-4 competes with CD28 for CD80/86 co-stimulation, blocking T cell activation. Loss of neoantigens: cancer cells under immune pressure can lose immunogenic mutations entirely, evolving away from immune recognition. Immunosuppressive cytokines: TGF-beta and IL-10 secreted by tumor cells and TAMs suppress effector T cell function. T cell exhaustion: chronic antigen exposure in the TME leads to a dysfunctional T cell phenotype (expressing PD-1, TIM-3, LAG-3) that is poorly cytotoxic. Checkpoint immunotherapy essentially reverses T cell exhaustion by blocking PD-1/PD-L1 and CTLA-4 interactions — releasing the brakes on anti-tumor T cells.

Why do some people get cancer and others don't?

Cancer risk is determined by the interplay of three broad factors: heritable predisposition, environmental exposure, and statistical chance (replication errors). Heritable predisposition: approximately 5-10% of cancers are associated with high-penetrance germline mutations — notably BRCA1/2 (breast, ovarian, prostate cancer), TP53 mutations (Li-Fraumeni syndrome), APC (familial adenomatous polyposis, near-certain colon cancer), and MLH1/MSH2/MSH6 (Lynch syndrome, colon and endometrial cancer). These high-penetrance mutations are relatively rare; more common are low-penetrance SNPs that modestly elevate risk across populations. Environmental exposure: tobacco smoking accounts for approximately 30% of all cancer deaths; it introduces carcinogens (polycyclic aromatic hydrocarbons, nitrosamines) that form DNA adducts, causing characteristic C→T transitions at specific sequence contexts — a mutational signature now detectable by whole-genome sequencing. UV radiation causes C→T and CC→TT transitions in melanocytes (melanoma signature). Obesity increases risk of at least 13 cancer types through chronic inflammation and elevated IGF-1/insulin/estrogen signaling. Statistical chance: Cristian Tomasetti and Bert Vogelstein's controversial 2015 Science paper estimated that two-thirds of cancer risk is attributable to random replication errors — 'bad luck' in the form of unavoidable copying mistakes during normal cell division. This fraction varies dramatically by cancer type: brain cancer and childhood leukemia are largely random; colon and lung cancer have large preventable fractions.

How does cancer treatment work — what does each major approach do?

Modern cancer treatment encompasses six major modalities, often used in combination. Surgery remains the primary treatment for solid tumors when localized — physically removing the tumor before metastasis occurs; the challenge is achieving clear margins (no cancer cells at the surgical edge) and managing micrometastases too small to detect. Radiation therapy kills cancer cells by creating DNA double-strand breaks that the cell cannot repair, exploiting the fact that rapidly dividing cells with impaired DNA repair (a cancer hallmark) are more radiosensitive than normal tissue. Conventional chemotherapy targets rapidly dividing cells with agents that damage DNA (platinum compounds, alkylating agents), interfere with DNA synthesis (antimetabolites like 5-FU), or disrupt cell division (taxanes, vinca alkaloids) — the side effects (hair loss, nausea, bone marrow suppression) reflect collateral damage to normal rapidly-dividing cells. Targeted therapy: tyrosine kinase inhibitors and monoclonal antibodies against specific molecular drivers — imatinib (Gleevec) against BCR-ABL in CML, trastuzumab (Herceptin) against HER2 in breast cancer, erlotinib against EGFR in lung cancer — achieve remarkable initial responses but are often limited by acquired resistance. Immunotherapy: checkpoint inhibitors (anti-PD-1, anti-CTLA-4), CAR-T cell therapy, and cancer vaccines reinvigorate or engineer the immune system against tumor cells; checkpoint inhibitors produce durable remissions in ~20-30% of responsive cancer types. Hormone therapy: blocks the hormones that drive hormone-sensitive cancers — antiandrogens (enzalutamide) in prostate cancer, aromatase inhibitors in estrogen-receptor positive breast cancer.

Share this article