By Billy Candra 
Breast cancers can be classified into subgroups that hint at the aggressiveness of the cancer and the likelihood that the patient will experience a recurrence years after their initial diagnosis. Now, groundbreaking research from Stanford Medicine has revealed that these molecular subgroups can be consolidated into three primary categories based on distinct structural variations within their DNA. These variations encompass critical genomic alterations such as repeats or amplifications of cancer-associated genes, known as oncogenes, found on chromosomes, and the presence of small, autonomous DNA circles untethered to the main genome, termed extrachromosomal DNA (ecDNA). Crucially, these foundational genomic variations are established remarkably early during cancer development and persist as the disease progresses and metastasizes, offering a stable blueprint of the tumor’s inherent characteristics.
This robust classification system represents a significant leap forward in oncology, promising to aid physicians in making more informed decisions regarding patient care. By illuminating the role these structural variations play in tumor evolution and identifying their vulnerabilities, the findings point toward the development of novel, targeted therapeutic interventions. Furthermore, this advanced stratification could empower clinicians to differentiate breast cancer patients who would most benefit from aggressive early intervention from those who might safely defer certain aspects of intensive treatment, thereby personalizing therapy and potentially mitigating unnecessary side effects.
The Longstanding Quest for Deeper Understanding
For decades, breast cancer classification has primarily relied on broad strokes, categorizing tumors based on the presence or absence of specific protein receptors. These include hormone-receptor positive (HR+), characterized by elevated levels of estrogen or progesterone receptors; HER2-positive (HER2+), marked by high levels of the HER-2 receptor; and triple-negative breast cancer (TNBC), which lacks all three receptors. While this system has guided treatment strategies, its limitations in predicting long-term outcomes and recurrence risk, particularly for HR+ cancers, have long been acknowledged.
Globally, breast cancer remains the most common cancer among women, accounting for over 2 million new cases annually and tragically claiming hundreds of thousands of lives each year. In the United States alone, approximately 1 in 8 women will develop invasive breast cancer in their lifetime. HR+ breast cancers represent the majority, roughly 70% of all cases, and are often successfully treated with hormone therapy, chemotherapy, surgery, and radiation. HER2+ cancers, accounting for 15% to 20% of cases, are aggressive but have seen remarkable improvements in outcomes since the advent of targeted therapies like trastuzumab (Herceptin) in the late 1990s. Triple-negative breast cancers, comprising about 10% to 15% of newly diagnosed cases, are notoriously difficult to treat due to their aggressive nature and propensity for early recurrence, lacking the specific targets amenable to hormone or HER2-directed therapies.
A Chronology of Discovery: From Receptors to Genomics
The journey toward this refined understanding of breast cancer’s genomic landscape has been a progressive one, spearheaded by researchers like Dr. Christina Curtis, the RZ Cao Professor and a professor of oncology, of genetics, and of biomedical data science at Stanford Medicine. Dr. Curtis, also the director of artificial intelligence and cancer genomics at the Stanford Cancer Institute, has dedicated over a decade to unraveling the evolutionary intricacies of breast cancer tumors.
- 2012: Uncovering Molecular Subgroups: Dr. Curtis and her colleagues first leveraged advanced machine-learning techniques to analyze DNA and RNA sequences from patients’ healthy cells and their breast tumors. This allowed for a molecular snapshot of genetic alterations and their impact on gene expression. This seminal study identified 11 clinically significant subgroups, a far greater resolution than receptor-based classifications, each associated with varied prognoses. However, the precise application of this information to guide patient care remained a challenge.
- 2019: Predicting Long-Term Recurrence: A subsequent, large-scale study involving 75,000 individuals with estrogen-receptor positive breast cancer highlighted a critical clinical dilemma: even after five years of hormone therapy, recurrences continued, even in patients initially deemed low-risk. Building on their 2012 findings, Dr. Curtis’s team demonstrated that overlaying receptor status with their previously defined molecular subgroups could accurately predict which HR+ tumors were prone to recur many years—even a decade or two—after initial diagnosis and treatment. Specifically, four of the eight HR+ subgroups showed a significantly elevated risk. Combining these high-risk groups revealed that a quarter of women with HR+ HER2-negative breast tumors faced a nearly 50% chance of recurrence decades later. This recurrence risk was alarmingly high, surpassing even that of TNBC before modern treatments and mirroring HER2+ outcomes before trastuzumab. The approach also identified TNBC patients less likely to recur beyond five years, allowing for more nuanced treatment intensity.
Despite these advancements, the fundamental drivers behind the differences in these subgroups remained elusive. "My lab has had a long-standing interest in understanding how aggressive breast tumors arise, why they are resistant to therapy and why they are prone to recur in distant organs," stated Dr. Christina Curtis, senior author of the latest research. "This research shows that breast tumors develop key structural variants that set the tumor on its course very early in its development. In short, some are born to be bad. It emphasizes the importance of robust biomarkers and of intervening early in the course of the disease."
Unveiling the Genomic Architecture: The Three Main Groups
The latest research, published January 8 in Nature, delves deeper into the genomic architecture—the mutations and structural variations within a cancer cell’s DNA—of nearly 2,000 breast cancers across all stages, from ductal carcinoma in situ (stage 0) to advanced metastatic disease (stage 4). Lead authors included former postdoctoral scholar Kathleen Houlahan, PhD, postdoctoral scholar Lise Mangiante, PhD, former research assistant Cristina Sotomayor-Vivas, and graduate student Alvina Adimoelja. The team’s unbiased assessment allowed them to categorize tumors into three distinct groups based on fundamental oddities in their genomes:
- The "Complex Amplification & ecDNA" Group: This group, comprising high-risk HR+ subgroups and strongly overlapping with HER2+ cancers, is characterized by complex yet localized amplifications of cancer-associated genes and the presence of extrachromosomal DNA (ecDNA). Oncogenes, which are genes that can promote cell growth and division, are frequently found on ecDNA. These small, circular DNA molecules are untethered to the main chromosomes and can exist in high copy numbers, often ignoring normal cellular regulatory mechanisms. Recent studies have increasingly implicated ecDNAs as potent drivers of rapid cancer growth, evolution, and drug resistance, making their presence a significant indicator of aggressive disease. "Here we have two different molecular subtypes, which we treat differently in the clinic but that strongly overlap in their patterns of chromosomal instability," Dr. Curtis noted, highlighting a crucial convergence of distinct clinical entities at the genomic level.
- The "Globally Unstable" Group: Primarily encompassing a subset of triple-negative tumors, this group exhibits widespread genomic instability, accumulating alterations across the entire genome rather than localized amplifications. A notable characteristic within this group is a deficiency in DNA repair mechanisms, meaning the cells are less able to fix errors that arise during DNA replication, leading to a cascade of mutations. As Dr. Curtis described, "The whole genome shows scars. It’s not limited to particular oncogenes," indicating a fundamental breakdown in genomic integrity. This pattern is reminiscent of the genomic havoc observed in cancers driven by BRCA1 and BRCA2 mutations.
- The "Relatively Stable Genome" Group: This category includes the more common, garden-variety HR+ HER2-negative breast cancers that typically present with standard risks of recurrence. These tumors are distinguished by their comparatively stable genomes, lacking the dramatic structural rearrangements seen in the other two high-risk groups. This genomic stability correlates with their generally favorable prognosis when treated with conventional therapies.
A pivotal finding from the study is that these defining structural variations are not late-stage developments but are present in the earliest stages of the disease, even in ductal carcinoma in situ. Furthermore, they are consistently maintained as the tumors grow and spread throughout the body, underscoring their foundational role in cancer pathogenesis. The researchers also found that these genomic architectures correlated with how immune cells infiltrate and respond to the tumor, opening avenues for immunotherapy research.
Implications for Clinical Practice and Future Therapies
Understanding the foundational importance of these structural variations and their impact on genomic architecture offers profound insights into cancer development and, critically, hints at entirely new therapeutic options. This research heralds a new era of precision oncology, moving beyond broad classifications to target the specific genomic vulnerabilities inherent to a tumor’s "birth."
- Personalized Treatment Stratification: The ability to classify tumors based on these fundamental genomic alterations allows for unprecedented patient stratification. For instance, identifying patients with high-risk HR+ tumors driven by complex amplifications and ecDNA early in their disease course could justify more aggressive initial treatment, intensive monitoring, or enrollment in clinical trials targeting these specific mechanisms. Conversely, patients with more stable genomes and lower recurrence risks might safely undergo less intensive treatments, sparing them from the harsh side effects of unnecessary therapies. This stratification system could be particularly impactful for the ~13% of patients with DNA repair-deficient, estrogen-receptor positive breast cancers, who might benefit from existing drugs designed to target impaired DNA repair pathways, such as PARP inhibitors, currently used for BRCA1/2-mutated cancers.
- Novel Targeted Interventions: The study highlights specific vulnerabilities. Tumors reliant on focal amplifications and ecDNA, for example, might be susceptible to compounds that directly target their respective drivers or the ensuing replication stress they induce. This could involve developing drugs that inhibit the replication of ecDNA or interfere with the specific oncogenes they carry. For tumors with globally unstable genomes and DNA repair deficiencies, strategies to exploit these weaknesses, potentially through synthetic lethality approaches, could prove effective. Still other therapeutic approaches may focus on directly targeting the mutational processes that propagate these catastrophic genomic events.
- Early Detection and Intervention: The observation that these critical mutational events occur "decades prior to the diagnosis of the tumor" underscores immense opportunities for earlier interventions. This could include improved screening methods capable of detecting these genomic alterations even before macroscopic tumor formation, or preventative strategies for high-risk individuals.
- Informed Prognosis and Patient Counseling: For patients and their families, a more accurate understanding of recurrence risk, especially for late recurrences in HR+ disease, can provide clarity and empower them to make informed decisions about long-term surveillance and lifestyle choices.
Dr. Curtis emphasized the significance of these early events: "Despite the complexity of their genomes, there are constraints and only so many evolutionary paths for a tumor to follow. We now have an understanding of how and when these complex alterations arise and their accompanying vulnerabilities." This perspective suggests that while cancer genomes may seem chaotic, there are underlying patterns that can be deciphered and exploited.
This landmark research was supported by critical funding from the National Institutes of Health (grants CA261719 and CA252457) and the Breast Cancer Research Foundation, underscoring the collaborative effort and significant investment in advancing cancer science. Dr. Christina Curtis is also a member of Bio-X and the Stanford Cancer Institute, as well as a Chan Zuckerberg Biohub investigator, further illustrating the interdisciplinary nature of this cutting-edge work. As this new genomic classification system becomes integrated into clinical research and practice, it promises to revolutionize how breast cancer is diagnosed, treated, and ultimately, overcome, offering renewed hope for patients worldwide.
