Pancreatic cancer misses this list on points (43.6% vs 3.4%, a 40-point gap), but survival is about 13 times higher when it is caught early. Only 15% of cases are found localized, and the USPSTF grades routine screening D.
Back to the trend of the day: Sitcom star who beat cancer twice joins Detroit hockey fans' charity night
Poly Rating · from the trend of the day
Recommended Cancer Screenings Guidelines 2026: 10 Cancers
Colorectal cancer survival falls from 91% to 17% once it spreads, and a proven test starts at 45, yet only one in three cases is caught while still localized.
- No. 1: Colorectal cancer
- 10 places
01Final list
The leading AIs each built their own list of ten independently. Here is one result that is easy to check: every place has a reason, a source, and the list of AIs that included it.
A bright badge means this AI put the entry on its list. A dim badge means it did not.
- 1
Colorectal cancer
Widest confirmed cliff: 91.3% survive five years when localized vs 16.9% once spread. A grade A/B test starts at 45, yet only 34% are caught early.
Included it: Claude, ChatGPT, Gemini, Nemotron, Kimi, DeepSeek. Did not include: Meta.In their lists: Claude: 1 · ChatGPT: 3 · Gemini: 1 · Nemotron: 2 · Kimi: 5 · DeepSeek: 3
What each AI noted 6
- Claudeplace 1
The widest survival cliff of any cancer with a routine test: 91% localized vs 17% distant, a 74-point drop. Yet only 34% of cases are caught early.
Source ↗ - ChatGPTplace 3
The localized-to-distant survival difference is 74.4 points, while only about one-third of cases are localized.
Source ↗ - Geminiplace 1
A catastrophic 77-percentage-point survival cliff paired with the only routine USPSTF Grade A non-invasive screening test in modern medicine.
Source ↗ - Nemotronplace 2
76-point cliff; screening exists (Grade A) but only 38% are caught early — the gap between tool and uptake.
Source ↗ - Kimiplace 5
The biggest survival cliff on the whole board — 91% localized versus 15% distant — and the only cancer here whose routine test prevents the disease, not just finds it.
Source ↗ - DeepSeekplace 3
Colorectal has the biggest absolute cliff here — 74.4 points — and a proven A/B screening test, yet only 33.8% of U.S. cases are caught at a localized stage.
Source ↗
6 AIs put it here - 2
Cervical cancer
A 71-point cliff (91.8% vs 20.5%) behind the best-proven checkup, grade A from 21, yet only 41% are found localized. Older ACS figures: 91% vs 19%.
Included it: Claude, ChatGPT, Gemini, Nemotron, Kimi, DeepSeek. Did not include: Meta.In their lists: Claude: 2 · ChatGPT: 10 · Gemini: 2 · Nemotron: 3 · Kimi: 9 · DeepSeek: 7
What each AI noted 6
- Claudeplace 2
The best-proven checkup: a grade A test from age 21. Even so, only 41% of cases are found localized, and survival drops from 92% to 20.5% when it has spread.
Source ↗ - ChatGPTplace 10
The survival gap is substantial, but routine screening already targets precancer before it becomes invasive cancer.
Source ↗ - Geminiplace 2
Catching it in place yields a 91% survival rate, yet fewer than half of cases are caught before regional or distant spread.
Source ↗ - Nemotronplace 3
74-point cliff; screening (Grade A) lifts early diagnosis to 44%, but disparities persist.
Source ↗ - Kimiplace 9
Screening works — 91% localized survival versus 19% distant, a grade-A Pap/HPV test — yet about half of cases are still not found until the cancer has spread beyond the cervix.
Source ↗ - DeepSeekplace 7
Cervical cancer has a 71.3-point cliff and an A-grade routine test, yet 16% of U.S. cases still present at a distant stage.
Source ↗
6 AIs put it here - 3
Stomach cancer
A 70-point cliff (78.1% vs 8.1%) and no routine U.S. test; 35% are already distant at diagnosis, while Japan and Korea screen nationally from 40.
Included it: Claude, ChatGPT, Kimi, DeepSeek. Did not include: Meta, Gemini, Nemotron.In their lists: Claude: 5 · ChatGPT: 6 · Kimi: 7 · DeepSeek: 4
What each AI noted 4
- Claudeplace 5
A 70-point cliff (78% localized vs 8% distant), and only 32% are caught early. I found no U.S. general-population screening recommendation.
Source ↗ - ChatGPTplace 6
The distant-stage five-year survival is 8.1%, about one-tenth the localized figure.
Source ↗ - Kimiplace 7
A 68-point cliff (75% to 7%) with no routine US screening test — in a country where, unlike Japan and South Korea, no population program exists to catch it early.
Source ↗ - DeepSeekplace 4
Stomach cancer drops 70.0 points from localized (78.1%) to distant (8.1%), and about two-thirds of cases are still found after the cancer has spread.
Source ↗
4 AIs put it here - 4
Uterine (endometrial) cancer
Largest gap here (94.9% vs 19.9%, 75 points), but only one source gives it, so it sits below the top three. Two-thirds are found early; no routine test.
Included it: Claude. Did not include: ChatGPT, Meta, Gemini, Nemotron, Kimi, DeepSeek.In their lists: Claude: 6
What each AI noted 1
- Claudeplace 6
The cliff is as large as colon cancer's (95% vs 20%), but two in three cases are found localized, usually because symptoms show up early. There is no routine test.
Source ↗
1 AIs put it here - 5
Kidney cancer
Estimate: 93% localized vs 18% (ACS) or 15% (another AI's SEER reading) distant, a 75–78-point gap. About two-thirds are found early, often by chance on scans; no routine test.
Included it: Nemotron, Kimi. Did not include: Claude, ChatGPT, Meta, Gemini, DeepSeek.In their lists: Nemotron: 1 · Kimi: 8
What each AI noted 2
- Nemotronplace 1
A 78-percentage-point survival cliff — yet 65% are caught early with no routine screening test.
Source ↗ - Kimiplace 8
A 75-point cliff (93% to 18%) with no screening test — but unlike most cancers here, about two-thirds of kidney cancers are still found while localized, usually by accident on a scan.
Source ↗
2 AIs put it here - 6
Breast cancer (female)
A 66-point cliff (100% vs 33.8%). Mammograms every 2 years from 40, grade B since 2024, help find 64% of cases while localized.
Included it: Claude, ChatGPT, Gemini, Nemotron, Kimi. Did not include: Meta, DeepSeek.In their lists: Claude: 4 · ChatGPT: 5 · Gemini: 3 · Nemotron: 4 · Kimi: 10
What each AI noted 5
- Claudeplace 4
Screening is working: 64% are found localized and 5-year survival is about 100%, but it falls to 34% once the cancer is distant.
Source ↗ - ChatGPTplace 5
Localized survival is 100%; distant-stage survival is 33.8%, and 6% of cases are distant at diagnosis.
Source ↗ - Geminiplace 3
Nine in ten deaths occur after metastasis, yet localized detection preserves a 99.3% five-year survival rate.
Source ↗ - Nemotronplace 4
69-point cliff; 63% caught early with Grade A screening — the model of population impact.
Source ↗ - Kimiplace 10
The largest absolute number of lives in play: 99% survival when localized versus 32% distant, two-thirds already found early, and a grade-B mammogram program that now starts at 40.
Source ↗
5 AIs put it here - 7
Melanoma (skin cancer)
Estimate: A cliff of about 66 points (~100% vs 34%). Most are found localized (77% on SEER Stat Facts, ~83% on a SEER training page), yet skin checks get only a USPSTF 'I'.
Included it: Claude, ChatGPT, Gemini, Nemotron. Did not include: Meta, Kimi, DeepSeek.In their lists: Claude: 8 · ChatGPT: 8 · Gemini: 4 · Nemotron: 6
What each AI noted 4
- Claudeplace 8
Catch it early and survival is about 100%; at the distant stage it is 34%. Yet no routine skin-check test has earned a USPSTF grade.
Source ↗ - ChatGPTplace 8
Roughly 83% of melanomas are localized; the survival gap is still large at 65.9 points.
Source ↗ - Geminiplace 4
A near-perfect 99.6% survival when localized plummets to 35.1% once spread, but no national routine skin exam checkup is formally recommended.
Source ↗ - Nemotronplace 6
64-point cliff; 83% caught early visually, yet USPSTF finds insufficient evidence for whole-body exams (Grade I).
Source ↗
4 AIs put it here - 8
Ovarian cancer
A 60-point cliff (91.9% vs 31.5%), but only 22% are found localized and 54% distant. The USPSTF grades routine screening D because trials showed no drop in deaths.
Included it: Claude, ChatGPT, Gemini, Nemotron, Kimi, DeepSeek. Did not include: Meta.In their lists: Claude: 7 · ChatGPT: 4 · Gemini: 6 · Nemotron: 7 · Kimi: 3 · DeepSeek: 5
What each AI noted 6
- Claudeplace 7
Fewer than 1 in 4 ovarian cancers (22%) are found localized, and over half (54%) are distant. At the localized stage, survival is 92%.
Source ↗ - ChatGPTplace 4
More than half of cases are distant at diagnosis, despite a 60.4-point survival gap by stage.
Source ↗ - Geminiplace 6
Survival drops from 93.1% to 31.5%, but because only 16% of tumors are localized at diagnosis, medicine currently advises against routine screening.
Source ↗ - Nemotronplace 7
62-point cliff but only 17% found early — the deadliest gap with no screening (Grade D).
Source ↗ - Kimiplace 3
A localized ovary tumor is survivable 92% of the time versus 31% once distant — and the USPSTF grades screening the general population a D, its vote of no confidence.
Source ↗ - DeepSeekplace 5
Ovarian cancer has a 60.4-point cliff and 54% of cases present distant, but the USPSTF gives average-risk screening a D — there is no proven routine test.
Source ↗
6 AIs put it here - 9
Prostate cancer
A 60-point cliff (100% vs 40.1%), but regional survival is also 100%, so stage matters less here. The PSA test gets only a grade C at 55–69.
Included it: Claude, ChatGPT, Gemini, Nemotron. Did not include: Meta, Kimi, DeepSeek.In their lists: Claude: 9 · ChatGPT: 7 · Gemini: 7 · Nemotron: 5
What each AI noted 4
- Claudeplace 9
Survival is 100% for localized and regional disease alike and 40% for distant. Because stage matters less here, the PSA test only gets a grade C.
Source ↗ - ChatGPTplace 7
A 59.9-point survival gap coexists with 69% of cases found localized; PSA screening is a personal decision, not a blanket recommendation.
Source ↗ - Geminiplace 7
Localized disease carries a 100% 5-year benchmark, but national guidelines relegate PSA blood testing to an individualized decision rather than an unconditional routine checkup.
Source ↗ - Nemotronplace 5
65-point cliff; 70% localized but Grade C screening means shared decision-making, not routine.
Source ↗
4 AIs put it here - 10
Lung cancer
A 55-point cliff (65.5% vs 10.5%), and 51% are already distant. A yearly low-dose CT is grade B, but only for people aged 50–80 with 20+ pack-years.
Included it: Claude, ChatGPT, Gemini, Nemotron, Kimi, DeepSeek. Did not include: Meta.In their lists: Claude: 3 · ChatGPT: 2 · Gemini: 5 · Nemotron: 8 · Kimi: 2 · DeepSeek: 2
What each AI noted 6
- Claudeplace 3
Only 24% of lung cancers are found localized, while 51% are already distant. A 55-point survival cliff sits behind a test that only smokers or former smokers qualify for.
Source ↗ - ChatGPTplace 2
51% of cases are distant at diagnosis; the localized-to-distant survival gap is 55 points.
Source ↗ - Geminiplace 5
America's top cancer killer offers an eightfold survival boost if caught early, but screening is strictly limited to high-risk heavy smokers.
Source ↗ - Nemotronplace 8
55-point cliff; Grade B screening for high-risk smokers only, yet just 24% caught early.
Source ↗ - Kimiplace 2
A proven screening scan exists, but it is reserved for heavy smokers — and roughly 4 in 5 cases are still found only after the cancer has left the lung.
Source ↗ - DeepSeekplace 2
Lung has the widest absolute survival cliff in this ranking — 55.0 points — but the USPSTF B test is reserved for high-risk adults 50–80 with a 20 pack-year history, not the general population.
Source ↗
6 AIs put it here
02Worth knowing
What the AIs noticed while building the list.
Only four cancers have a USPSTF grade A or B screening test: cervical, colorectal, breast, and lung (for heavy smokers only).
Having a test is not the same as using it: colorectal cancer has a grade A/B test, but only about one in three cases is found while still localized.
In prostate cancer, 'earlier' is not always better: regional survival matches localized survival (100%), and overdiagnosis is why PSA testing gets only a grade C.
'Localized' is SEER summary stage, not stage I. Lead-time bias and overdiagnosis can widen stage gaps, and a 100% relative survival is a rounded figure, not zero deaths.
03How we counted
In plain words, no formulas.
1The question
Top 10 Cancers Where Catching It Early Matters Most — and Whether a Routine Checkup Exists
2Why this topic
Readers see which cancers have a large early-detection payoff and a proven checkup, which have a payoff but no routine test, and what to watch for in the second group. SEER and the USPSTF each publish only one half of this.
3Where it applies
United States, by cancer type
4What we looked at
- Survival cliff: the gap in 5-year relative survival between cancers found while still localized and cancers found after they have spread to distant organs
- Share of cases that are currently diagnosed at the localized stage
- Whether a routine screening test exists for the general population (USPSTF grade A/B), only for a high-risk group, or not at all
- National Cancer Institute, SEER Cancer Stat Facts — 5-year relative survival by stage (localized, regional, distant) and share of cases found at each stage, per cancer site (SEER 21, 2016–2022)
- American Cancer Society, Cancer Facts & Figures (SEER-based) — 5-year relative survival by stage using earlier diagnosis years (2014–2020 / 2015–2021); used where SEER Stat Facts figures were not retr
- U.S. Preventive Services Task Force, A and B recommendations — Which cancer screenings get grade A/B for people without symptoms, with eligible ages and risk groups
- U.S. Preventive Services Task Force, individual cancer statements — Grades C, D and I (individual decision, recommend against, insufficient evidence) for prostate, ovarian, pancreatic, skin and other
5Who ranked
The leading AIs, each building its own list of ten independently and naming a source for every place: Claude, ChatGPT, Meta, Gemini, Nemotron, Kimi, DeepSeek.
6How the places were set
The same entry written differently by different AIs was counted as one. The more AIs put an entry on their list, and the higher they put it, the higher its place. "AI agreement" shows how many of the AIs had it on their own list. Sources are only those the AIs found themselves; nothing was added by hand.
04Each AI's own list
🇺🇸 Claude10 places · 4 notes
The same question, answered by Claude alone. The final list above combines all the AIs.
- 1Colorectal cancer (colon and rectum)
The widest survival cliff of any cancer with a routine test: 91% localized vs 17% distant, a 74-point drop. Yet only 34% of cases are caught early.
Source ↗ - 2Cervical cancer
The best-proven checkup: a grade A test from age 21. Even so, only 41% of cases are found localized, and survival drops from 92% to 20.5% when it has spread.
Source ↗ - 3Lung cancer
Only 24% of lung cancers are found localized, while 51% are already distant. A 55-point survival cliff sits behind a test that only smokers or former smokers qualify for.
Source ↗ - 4Breast cancer (female)
Screening is working: 64% are found localized and 5-year survival is about 100%, but it falls to 34% once the cancer is distant.
Source ↗ - 5Stomach cancer
A 70-point cliff (78% localized vs 8% distant), and only 32% are caught early. I found no U.S. general-population screening recommendation.
Source ↗ - 6Uterine (endometrial) cancer
The cliff is as large as colon cancer's (95% vs 20%), but two in three cases are found localized, usually because symptoms show up early. There is no routine test.
Source ↗ - 7Ovarian cancer
Fewer than 1 in 4 ovarian cancers (22%) are found localized, and over half (54%) are distant. At the localized stage, survival is 92%.
Source ↗ - 8Melanoma of the skin
Catch it early and survival is about 100%; at the distant stage it is 34%. Yet no routine skin-check test has earned a USPSTF grade.
Source ↗ - 9Prostate cancer
Survival is 100% for localized and regional disease alike and 40% for distant. Because stage matters less here, the PSA test only gets a grade C.
Source ↗ - 10Pancreatic cancer
The starkest case of missing early detection: only 15% are found localized, 51% are already distant, and distant 5-year survival is 3.4%.
Source ↗
What Claude noticed
Colorectal cancer has the biggest localized-to-distant drop on the list (91.3% to 16.9%). Only 34% of cases are found localized, even though a grade A/B test has existed for decades.
Prostate cancer shows that 'earlier' does not always mean 'better'. Regional survival is as high as localized survival (100%) and only distant disease drops, to 40.1%. This is part of why PSA screening gets only a grade C.
Lung and pancreatic cancer are the two where over half of cases are distant at diagnosis (51% each). Lung now has a screening test for a high-risk group; pancreas does not.
In 2024 the Task Force moved the start of routine mammograms from a choice at 40–49 to a grade B recommendation at 40, every 2 years. It kept the evidence for women 75 and older as insufficient.
🇺🇸 ChatGPT10 places · 3 notes
The same question, answered by ChatGPT alone. The final list above combines all the AIs.
- 1Pancreatic cancer (United States)
About half of cases are already distant when found, where five-year relative survival is 3.4%.
Source ↗ - 2Lung and bronchus cancer (United States)
51% of cases are distant at diagnosis; the localized-to-distant survival gap is 55 points.
Source ↗ - 3Colorectal cancer (United States)
The localized-to-distant survival difference is 74.4 points, while only about one-third of cases are localized.
Source ↗ - 4Ovarian cancer (United States)
More than half of cases are distant at diagnosis, despite a 60.4-point survival gap by stage.
Source ↗ - 5Female breast cancer (United States)
Localized survival is 100%; distant-stage survival is 33.8%, and 6% of cases are distant at diagnosis.
Source ↗ - 6Stomach cancer (United States)
The distant-stage five-year survival is 8.1%, about one-tenth the localized figure.
Source ↗ - 7Prostate cancer (United States)
A 59.9-point survival gap coexists with 69% of cases found localized; PSA screening is a personal decision, not a blanket recommendation.
Source ↗ - 8Melanoma of the skin (United States)
Roughly 83% of melanomas are localized; the survival gap is still large at 65.9 points.
Source ↗ - 9Liver and intrahepatic bile duct cancer (United States)
The localized survival figure is about 44%, but distant-stage survival is only a few percent.
Source ↗ - 10Cervical cancer (United States)
The survival gap is substantial, but routine screening already targets precancer before it becomes invasive cancer.
Source ↗
What ChatGPT noticed
The biggest measured survival cliffs do not necessarily mean there is a proven population screening test: pancreatic and ovarian cancer have steep stage differences, while USPSTF recommends against general-population screening for both.
Lung cancer combines a large stage gap with a screening test, but eligibility is limited to a defined smoking-history group.
Colorectal screening can prevent some cancers by finding and removing precancerous growths, so its value is not fully represented by stage-specific survival comparisons.
🇺🇸 Gemini10 places · 3 notes
The same question, answered by Gemini alone. The final list above combines all the AIs.
- 1Colorectal Cancer
A catastrophic 77-percentage-point survival cliff paired with the only routine USPSTF Grade A non-invasive screening test in modern medicine.
Source ↗ - 2Cervical Cancer
Catching it in place yields a 91% survival rate, yet fewer than half of cases are caught before regional or distant spread.
Source ↗ - 3Female Breast Cancer
Nine in ten deaths occur after metastasis, yet localized detection preserves a 99.3% five-year survival rate.
Source ↗ - 4Melanoma of the Skin
A near-perfect 99.6% survival when localized plummets to 35.1% once spread, but no national routine skin exam checkup is formally recommended.
Source ↗ - 5Lung and Bronchus Cancer
America's top cancer killer offers an eightfold survival boost if caught early, but screening is strictly limited to high-risk heavy smokers.
Source ↗ - 6Ovarian Cancer
Survival drops from 93.1% to 31.5%, but because only 16% of tumors are localized at diagnosis, medicine currently advises against routine screening.
Source ↗ - 7Prostate Cancer
Localized disease carries a 100% 5-year benchmark, but national guidelines relegate PSA blood testing to an individualized decision rather than an unconditional routine checkup.
Source ↗ - 8Urinary Bladder Cancer
Survival tumbles from 70% to 8% upon metastasis, yet routine urinalysis screening remains unendorsed due to excessive false-positive rates.
Source ↗ - 9Pancreatic Cancer
A localized diagnosis preserves a 44% 5-year outlook, but silent progression means only 1 in 8 patients are diagnosed before spread.
Source ↗ - 10Thyroid Cancer
Survival drops 47 points once distant, yet screening healthy necks is discouraged because it creates an epidemic of harmless microtumor diagnoses.
Source ↗
What Gemini noticed
Only 3 cancer sites in the United States (colorectal, cervical, and breast) have USPSTF Grade A or B recommendations for universal routine screening across average-risk adults.
The deadliest cancers by total annual mortality—lung and pancreatic—have the lowest localized detection rates (21% and 12%), severely blunting the curative window.
Prostate and thyroid cancers achieve near-100% 5-year survival at localized stages, but widespread routine checks are discouraged to prevent overdiagnosis and aggressive treatment of indolent nodules.
🇺🇸 Nemotron10 places · 3 notes
The same question, answered by Nemotron alone. The final list above combines all the AIs.
- 1Kidney and Renal Pelvis Cancer
A 78-percentage-point survival cliff — yet 65% are caught early with no routine screening test.
Source ↗ - 2Colon and Rectum Cancer
76-point cliff; screening exists (Grade A) but only 38% are caught early — the gap between tool and uptake.
Source ↗ - 3Cervical Cancer
74-point cliff; screening (Grade A) lifts early diagnosis to 44%, but disparities persist.
Source ↗ - 4Female Breast Cancer
69-point cliff; 63% caught early with Grade A screening — the model of population impact.
Source ↗ - 5Prostate Cancer
65-point cliff; 70% localized but Grade C screening means shared decision-making, not routine.
Source ↗ - 6Melanoma of the Skin
64-point cliff; 83% caught early visually, yet USPSTF finds insufficient evidence for whole-body exams (Grade I).
Source ↗ - 7Ovarian Cancer
62-point cliff but only 17% found early — the deadliest gap with no screening (Grade D).
Source ↗ - 8Lung and Bronchus Cancer
55-point cliff; Grade B screening for high-risk smokers only, yet just 24% caught early.
Source ↗ - 9Thyroid Cancer
47-point cliff but near-100% localized survival; overdiagnosis concern drives Grade D against screening.
Source ↗ - 10Testicular Cancer
26-point cliff — smallest on list — but 99% localized survival and 68% caught early without screening (Grade D).
Source ↗
What Nemotron noticed
Cervical and colorectal cancers are the only two with Grade A recommendations and survival cliffs above 70 pp — they are the clearest 'screening wins' where early detection transforms outcomes.
Ovarian cancer's 62-pp cliff is wasted: only 17% are found early, and USPSTF recommends against screening (Grade D) because trials showed no mortality reduction and harms from false positives.
Melanoma's 83% localized diagnosis rate — highest of all — happens largely through patient/partner skin checks and clinician exams, yet USPSTF says evidence is insufficient (Grade I) to recommend routine whole-body skin exams.
🇨🇳 Kimi10 places · 4 notes
The same question, answered by Kimi alone. The final list above combines all the AIs.
- 1Pancreatic cancer
Catching it early is worth 41 points of survival (44% vs 3%), yet only about 1 in 6 cases is found that early and no routine screening test exists for anyone.
Source ↗ - 2Lung cancer
A proven screening scan exists, but it is reserved for heavy smokers — and roughly 4 in 5 cases are still found only after the cancer has left the lung.
Source ↗ - 3Ovarian cancer
A localized ovary tumor is survivable 92% of the time versus 31% once distant — and the USPSTF grades screening the general population a D, its vote of no confidence.
Source ↗ - 4Esophageal cancer
A 43-point survival cliff (48% to 5%) — one of the widest of any cancer — with no routine screening test for the general population at all.
Source ↗ - 5Colorectal cancer
The biggest survival cliff on the whole board — 91% localized versus 15% distant — and the only cancer here whose routine test prevents the disease, not just finds it.
Source ↗ - 6Liver cancer
Even the lucky minority caught early (42% of cases) survives only 37-38% of the time — the worst 'early' survival of any major cancer — and only high-risk patients get surveillance.
Source ↗ - 7Stomach cancer
A 68-point cliff (75% to 7%) with no routine US screening test — in a country where, unlike Japan and South Korea, no population program exists to catch it early.
Source ↗ - 8Kidney (renal) cancer
A 75-point cliff (93% to 18%) with no screening test — but unlike most cancers here, about two-thirds of kidney cancers are still found while localized, usually by accident on a scan.
Source ↗ - 9Cervical cancer
Screening works — 91% localized survival versus 19% distant, a grade-A Pap/HPV test — yet about half of cases are still not found until the cancer has spread beyond the cervix.
Source ↗ - 10Breast cancer (female)
The largest absolute number of lives in play: 99% survival when localized versus 32% distant, two-thirds already found early, and a grade-B mammogram program that now starts at 40.
Source ↗
What Kimi noticed
The single widest survival cliff in the data belongs to a cancer that already has a routine test: colorectal (91% localized vs 15% distant), while the cancers with no test at all (pancreas, ovary, liver, esophagus) have the smallest share of early diagnoses.
The prostate paradox: the survival cliff is enormous (>99% localized vs 38% distant) yet the USPSTF gives PSA screening only a C — an individual decision — because overdiagnosis harms outweigh benefits; specialty societies disagree and recommend testing.
Melanoma is the counterexample where early detection 'already won' without any organized screening program: ~83% of melanomas are found localized and 5-year survival is 97.6% at that stage, driven by skin-awareness behavior rather than a national test.
Lung cancer shows the cost of a narrow screening rule: LDCT screening has a B grade, but only for heavy smokers aged 50-80 — and about 80% of cases are still diagnosed after the tumor has spread.
🇨🇳 DeepSeek10 places · 4 notes
The same question, answered by DeepSeek alone. The final list above combines all the AIs.
- 1Pancreatic cancer (U.S., 2026)
Only 15.4% of U.S. pancreatic cancers are found localized, and survival runs 43.6% localized versus 3.4% distant — a 40-point cliff with no average-risk screening test.
Source ↗ - 2Lung and bronchus cancer (U.S., 2026)
Lung has the widest absolute survival cliff in this ranking — 55.0 points — but the USPSTF B test is reserved for high-risk adults 50–80 with a 20 pack-year history, not the general population.
Source ↗ - 3Colorectal cancer (U.S., 2026)
Colorectal has the biggest absolute cliff here — 74.4 points — and a proven A/B screening test, yet only 33.8% of U.S. cases are caught at a localized stage.
Source ↗ - 4Stomach cancer (U.S., 2026)
Stomach cancer drops 70.0 points from localized (78.1%) to distant (8.1%), and about two-thirds of cases are still found after the cancer has spread.
Source ↗ - 5Ovarian cancer (U.S., 2026)
Ovarian cancer has a 60.4-point cliff and 54% of cases present distant, but the USPSTF gives average-risk screening a D — there is no proven routine test.
Source ↗ - 6Oral cavity and pharynx cancer (U.S., 2026)
Oral cavity survival falls 52.7 points from localized to distant, yet routine oral screening has an I statement — not enough evidence to recommend it for asymptomatic adults.
Source ↗ - 7Cervical cancer (U.S., 2026)
Cervical cancer has a 71.3-point cliff and an A-grade routine test, yet 16% of U.S. cases still present at a distant stage.
Source ↗ - 8Bladder cancer (U.S., 2026)
Bladder’s localized-to-distant cliff is 63.4 points, but 50% of cases are in situ and routine screening still has an I grade — early detection is common but not systematically proven.
Source ↗ - 9Esophageal cancer (U.S., 2026)
Esophageal cancer has a 43.3-point cliff, only 19.1% are localized, and there is no average-risk national screening recommendation.
Source ↗ - 10Liver and intrahepatic bile duct cancer (U.S., 2026)
Liver cancer is the exception where 44.7% are localized, but localized survival is still only 37.4% and distant disease 3.6% — early stage helps, but the absolute prognosis remains poor.
Source ↗
What DeepSeek noticed
Lung cancer’s 55-point localized-to-distant gap is the largest absolute cliff in this top 10, yet the B-grade test is not universal: it is limited to adults 50–80 with a 20 pack-year smoking history.
Pancreatic cancer has the smallest localized share in this ranking — 15.4% — and a 40-point survival cliff, but the USPSTF gives average-risk screening a D rather than a routine recommendation.
Colorectal cancer has the widest absolute survival cliff at 74.4 points and a proven A/B test, yet only about one in three U.S. cases is diagnosed while still localized — a test existing is not the same as a test being used.
Cervical cancer’s 71.3-point cliff is the largest among cancers with an A-grade routine test; 41.2% are localized, but 16% still present distant, showing that access and follow-up matter alongside test existence.