Early-onset cancers are reshaping the cancer landscape, and this Cell Perspective lays out how researchers could uncover hidden causes across the life course and turn those insights into smarter prevention.
Perspective: Accelerating discovery of cancer causes for prevention in the era of rising early-onset cancers. Image Credit: Lightspring / Shutterstock
A recent Perspective article published in the journal Cell reviewed the key milestones in cancer etiology research and highlighted contemporary challenges that impede progress.
Early-onset cancers, i.e., those diagnosed before age 50, have been rapidly increasing worldwide. Globally, they account for nearly 50 million disability-adjusted life years (DALYs) and nearly one million deaths, mounting significant societal, economic, and personal burdens. While cancer mortality has decreased among older people in the United States (US), mortality under age 50 has plateaued overall since the 1990s and increased for endometrial and colorectal cancers.
These increases exhibit strong birth-cohort effects, with Millennials and Generation X having higher risks at the same ages as earlier-born cohorts. This shift underscores the need to expedite the identification of novel causes and translate these insights into prevention and interception strategies. In the present Perspective, the authors reviewed key milestones in the discovery of cancer causes and outlined contemporary barriers to progress.
Cancer cause discovery: historical perspectives
A 1981 study articulated two strategies for discovering cancer causes: a mechanistic approach involving experimental testing of candidate agents and a black-box epidemiology strategy. Over the past decades, advances in genomics, molecular epidemiology, mechanistic biology, causal inference, and prospective cohorts have transformed cancer cause discovery and explanation.
The convergence of epidemiological and mechanistic evidence underpins the classification of group 1 carcinogens by the International Agency for Research on Cancer (IARC). Alcohol intake, obesity, and tobacco consumption represent important avoidable causes of cancer. As group 1 carcinogens, alcohol and tobacco exemplify how mechanistic research and epidemiology converge.
For tobacco, observations from the 18th century linked pipe and snuff use to lip cancer and nasal polyps. By the mid-20th century, studies demonstrated a substantially higher risk of lung cancer in heavy smokers. Further, observations from the 20th century linked alcohol to upper aerodigestive tract cancers, with lower risks in abstinent groups.
In 1987, alcohol was classified as a group 1 carcinogen for cancers of the liver, oral cavity, esophagus, larynx, and pharynx. Subsequent evaluations expanded this list to include colorectal and female breast cancers, and mechanistic studies linked alcohol to acetaldehyde toxicity, oxidative stress, inflammation, hormonal changes, and interactions with tobacco. Since the 1970s, the increasing prevalence of obesity has promoted epidemiological investigations, which linked increased body weight to cancer death. IARC evaluations from 2002 and 2016 show that avoiding weight gain reduces the risk of at least 13 cancers.
Discovery of major causes of cancer: Tobacco, alcohol, obesity, and genetics
Contemporary barriers to the discovery of cancer causes
Age at diagnosis indicates when the disease is detected, and depends on screening, healthcare access, diagnostic pathways, and age of onset. This process is continuous, varying by people and over time. As such, age at diagnosis is a limited proxy for identifying the unique biology of early-onset tumors. Therefore, analyses will benefit by treating age as a continuous variable, modeling both period and birth-cohort effects, and interpreting molecular as well as exposure patterns.
Early understanding of established cancer causes is mainly based on simplified measures, such as single-time-point assessments and questionnaire recall. However, these snapshots missed timing, trajectories, intensity, and cumulative exposures over the life course, resulting in an underestimation of preventable burden and risk. Moving forward, efficient, innovative, and objective characterization of exposures that captures timing, intensity, trajectories, and clustering is needed. The authors also noted that the exposome is a useful framework, but not a complete answer on its own, because real-world exposures are numerous, dynamic, and difficult to disentangle.
Mechanistic evidence in human cells, tissues, or experimental systems can enhance hazard evaluation by showing how exposure impacts cells and tissues. Nevertheless, translational limitations from the laboratory to humans are substantial. In the future, while cancer cause discovery will continue to be guided by consistency across epidemiological studies, embedding experimental models as a complementary layer for hypothesis testing could achieve maximal impact.
Frameworks for accelerated cancer cause discovery
The authors proposed three frameworks to accelerate the discovery of cancer causes: tissue-ecosystem-anchored, biological-state-based, and dynamic. The tissue ecosystem-anchored framework reframes cancer risk as an emergent feature of dynamic tissue ecosystems, focusing on how cumulative exposures across key life stages generate persistent biological signatures that affect somatic evolution, tissue susceptibility, and tumorigenesis.
Linking such tissue-level signatures to upstream drivers allows for cancer cause discovery and the identification of modifiable exposures for prevention. The biological state-based framework conceptualizes cancer risk as a continuous, evolving process in which physiological changes and exposures accumulate over the life course. It emphasizes quantifying tissue states preceding clinical detection to enhance prediction and enable precision screening and prevention.
The dynamic framework characterizes cancer preventability by synthesizing evidence from mechanistic, implementation, and population sciences to guide feasible, high-impact prevention approaches. It involves modeling cancer preventability at the individual level, informed by natural history, and incorporating changes in exposure across life stages and birth cohorts.
Concluding remarks
In sum, the Perspective highlighted the need for closer integration between epidemiological and mechanistic studies and proposed three frameworks for accelerating cancer cause discovery. It also emphasized that genetics alone is unlikely to explain the rapid rise in early-onset cancers, although inherited susceptibility may help determine who is most vulnerable to modern exposures. Advances in these frameworks will depend on how well non-genetic exposures and genetic susceptibility can be measured over the life course and across generations, and will require sustained inter-disciplinary collaboration.
Researchers develop low-cost scalable sensors to study genetic brain disorders