Two words get used interchangeably in headlines and clinics. They mean different things, and the difference is starting to matter.

Walk into almost any cancer conversation today and you will hear early detection and screening used as if they were synonymous. They are not.

Screening is a system. Early detection is an outcome.

Screening is structured by design.

• A defined population.

• A validated test.

• A fixed interval.

• A predefined follow-up pathway.

This framework traces back to Wilson and Jungner's 1968 WHO monograph and still underpins national screening policy today. [1]

Early detection is the broader goal: identifying a cancer earlier in its course than it would otherwise be found. Structured screening is one pathway to get there. Others include risk-based surveillance for BRCA1/BRCA2 or Lynch syndrome carriers [2], hepatocellular carcinoma surveillance in patients with chronic hepatitis B [3], and incidental findings on scans ordered for unrelated reasons.

Screening is one pathway to early detection. In practice, most cancers are still diagnosed after symptoms appear. In the US, that figure likely exceeds 80%. [4]

The mismatch between programs and biology

Screening programs are built around defined populations, set ages, and fixed intervals. Cancer biology does not follow those rules.

A tumor's status as “early” or “late” is shaped by accumulated mutations, clonal expansion, and immune engagement. By the time a cancer is large enough to appear on CT or shed enough signal to be detectable, it has typically been evolving for years [5].

The gap is showing up in real time. Current guidelines recommend colorectal cancer screening begin at age 45. Yet across 14 countries we are seeing a rise in colorectal cancer incidence for adults under 50, while rates stabilize in older adults. [6] Many younger patients are diagnosed before screening age. The pathway was not built to find them.

A case in practice

Consider a 60-year-old asymptomatic man who underwent multi-cancer early detection testing as part of a routine health assessment. The blood test returned a cancer signal. PET-CT showed focal uptake in the right colon. Colonoscopy confirmed stage III colorectal cancer. He underwent surgery with curative intent. [7]

Without that result, he would likely have remained asymptomatic for months or longer. By the time symptoms prompted investigation, the window for curative surgery may have closed. The test did not make the diagnosis. The clinical team did. But the signal arrived early enough that the full diagnostic pathway could run its course before late-stage symptoms took over.

This is where MCED fits into practice: not as a replacement for structured screening, but as a tool that can surface signals in patients who fall outside existing programs and in cancers that no current program screens for.

Multi-cancer early detection: filling a gap

Roughly 8 in 10 cancer deaths worldwide come from cancers with no general-population screening test today. [8] Three cancers account for a large share of that burden:

• Lung cancer. The leading cause of cancer death globally, responsible for nearly 1.8 million deaths per year, and increasingly diagnosed in patients who have never smoked. [8]

• Hepatocellular carcinoma. One of the fastest-rising causes of cancer mortality worldwide, with particularly high rates across East Asia driven by chronic hepatitis B prevalence. [3]

• Pancreatic cancer. Rarer, but almost always diagnosed late, with persistently low five-year survival rates. [9]

None of these has a general-population screening test. Patients are most often diagnosed after symptoms appear, by which point options have narrowed. MCED is designed to look for signals from these cancers earlier, offering a possible route to diagnosis where today there is none.

That breadth comes with interpretive demands. Cancers occur at very different frequencies across populations. A "cancer signal detected" result carries different implications depending on who is being tested and which cancer the signal points toward. This is why MCED tests report a signal alongside a predicted tissue of origin rather than a binary result, and why MCED is designed to complement existing screening rather than replace it.

Two ideas worth understanding

Two concepts appear repeatedly in coverage of new detection tests. Both are worth knowing before reading the next headline.

LEAD-TIME BIAS

Suppose two patients have the same cancer and will die at the same age. The difference is when their cancer is found. From the date of diagnosis, the screened patient appears to live twice as long. The death date has not changed. The extra "survival" comes from detecting the cancer earlier, not from altering its course.

This is why "screened patients live longer after diagnosis" is not, on its own, evidence that a test works. The only reliable measure is randomized trial data on overall mortality.

OVERDIAGNOSIS

Some cancers found by screening would never have caused harm in the patient's lifetime. The tumor grows slowly. The patient dies of something else first. Treatment adds side effects without adding life. Estimating how often this happens requires modeling what would have occurred without detection. The best estimates for breast and prostate cancer screening suggest meaningful but not enormous rates. [10]

Neither concept means screening does not work. They mean that "we caught more cancers earlier" is not on its own proof that a test is working.

What a blood test is actually reading

Research on blood-based detection has not converged on a single approach. Some methods measure circulating tumor DNA (ctDNA) directly, reading the genetic alterations shed by cancer cells. Others use methylation-based detection, identifying chemical marks on DNA that correlate with cancer presence. Each approach has distinct strengths and limitations. [11]

One number anchors the interpretation: positive predictive value (PPV). Of all people who receive a positive result, what fraction actually have cancer? Even a highly specific test will generate false positives when the underlying cancer is rare. This is why a structured diagnostic follow-up (imaging, endoscopy) is the designed next step after a positive MCED signal. The test opens a clinical conversation; it does not close one.

The bottom line

Screening is a structured system with trial-level evidence behind it. It works well for a defined set of cancers. Early detection is the broader goal of finding cancer earlier than it would otherwise be found, by any pathway.

That gap, between what structured screening covers and what cancer biology demands, is precisely where multi-cancer detection is designed to help. The technology brings its own interpretive questions around positive predictive value, lead time, and overdiagnosis. These are not reasons to dismiss it. They are the questions that will shape how it is used well.

Patients deserve answers before symptoms arrive. The work is finding them.

References

1. Wilson JMG, Jungner G. Principles and Practice of Screening for Disease. Public Health Papers No. 34. World Health Organization; 1968.

2. Daly MB, Pal T, Maxwell KN, et al. NCCN Guidelines: Genetic/Familial High-Risk Assessment: Breast, Ovarian, and Pancreatic, Version 2.2024. J Natl Compr Canc Netw. 2023; 21(10): 1000–1010.

3. Omata M, Cheng AL, Kokudo N, et al. Asia–Pacific clinical practice guidelines on the management of hepatocellular carcinoma: a 2017 update. Hepatol Int. 2017; 11(4): 317–370.

4. Sarma EA, Kobrin SC, Thompson MJ. A proposal to improve the early diagnosis of symptomatic cancers in the United States. Cancer Prev Res (Phila). 2020; 13(9): 715–720.

5. Hu Z, Li Z, Ma Z, Curtis C. Multi-cancer analysis of clonality and the timing of systemic spread in paired primary tumors and metastases. Nat Genet. 2020; 52(7): 701–708.

6. Sung H, Siegel RL, Rosenberg PS, et al. Colorectal cancer incidence trends in younger versus older adults: an analysis of population-based cancer registry data. Lancet Oncol. 2024. doi:10.1016/S1470-2045(24)00600-4.

7. Tucker S, et al. Paper presented at: European Society for Medical Oncology (ESMO) Congress; November 2023; Madrid, Spain.

8. Bray F, Laversanne M, Sung H, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2024; 74(3): 229–263.

9. Li J, Li Y, Chen C, et al. Recent estimates and predictions of 5-year survival rate in patients with pancreatic cancer: a model-based period analysis. Front Med (Lausanne). 2022; 9: 1049136.

10. Welch HG, Black WC. Overdiagnosis in cancer. J Natl Cancer Inst. 2010; 102(9): 605–613.

11. Rendek T, Pos O, Duranova T, et al. Current challenges of methylation-based liquid biopsies in cancer diagnostics. Cancers (Basel). 2024; 16(11): 2001.