Protocols by Charles Sieg

The research frontier

Research frontier: cancer

The results closest to cures or lasting remission across several cancers, plus the screening that catches cancer early, each with where it stands as of November 2026.

Overall evidence Mixed30 sourcesReviewed November 5, 2026

A few results rest on large trials or cohorts with long follow-up (melanoma immunotherapy, HPV vaccination, lung screening), several are small or early (CAR-T in brain tumors, personalized vaccines), and the multi-cancer blood test missed its main goal in a large randomized trial.

The short version

  • Almost nothing here is a proven cure. I sort every result into cure, durable remission, longer survival or slowing, or prevention and earlier detection, and I say which one it is.
  • The closest thing to cure-level data is in melanoma: among people alive and progression-free 3 years into nivolumab plus ipilimumab, 10-year melanoma-specific survival was 96%1. That is a selected group, and it is a survival number, not a cure.
  • Dostarlimab, for a genetic subtype of rectal cancer, produced complete responses that let most patients skip surgery, and the FDA granted it accelerated approval on October 19, 20262,3. The long-term cure rate is unknown.
  • HPV vaccination prevents cervical cancer: Scotland recorded no invasive cervical cancers in women vaccinated at 12 or 134.
  • Low-dose CT screening lowers lung-cancer deaths in heavy smokers5,6. The multi-cancer blood test did not meet its main goal in its largest randomized trial7.

How I sort results

  • Cure: the cancer is gone and stays gone for good. Nobody can prove that for one person, so studies use long follow-up as a stand-in.
  • Durable remission: the cancer is undetectable and stays that way for years, with no promise it never returns.
  • Longer survival or slowing: people live longer or the cancer grows more slowly, but it is still there or comes back.
  • Prevention or earlier detection: the cancer never starts, or it is caught while it can still be treated.

Anything that happened in one patient, a handful of patients, or mice is Early, no matter how exciting it looks.

Where each one stands (as of November 2026)

ResultWhat it actually isStatus
CAR-T in leukemia and lymphomaDurable remission in some patients; the lymphoma trial's authors see curative potential in a subsetTwo patients followed more than 10 years; a 101-patient trial followed 5 years8,9
GD2 CAR-T in diffuse midline gliomas, including DIPGEarly: tumor shrinkage, one long complete responsePhase 110
Dostarlimab, mismatch repair-deficient rectal cancerComplete responses, surgery avoidedFDA accelerated approval, October 19, 20263
Checkpoint inhibitors in advanced melanomaDurable remission for some, longer survival overallRandomized trial with 10-year follow-up1
Personalized mRNA vaccine in melanomaLower risk of recurrence, not yet shown to lengthen lifePhase 3 result in a company press release; investigational11
Personalized vaccine in pancreatic cancerEarlyPhase 1 data only12
Daraxonrasib (RAS inhibitor) in pancreatic cancerLonger survival, by monthsFDA approved August 26, 202613
Sotorasib (KRAS G12C inhibitor) in lung cancerSlowingPhase 3 trial: about one more month without progression14
HPV vaccinationPreventionPopulation results in three countries4,15,16
Low-dose CT, colonoscopy, multi-cancer blood testEarlier detectionVaries; see below

Why it works

Most of this frontier is the immune system pointed at cancer. Checkpoint inhibitors such as nivolumab, ipilimumab, pembrolizumab, and dostarlimab block PD-1 or CTLA-4, brakes that hold T cells back1,17,11. CAR-T cells are a patient's own T cells re-engineered to hunt one marker on the tumor: CD19 in leukemia and GD2 in the brain tumors below8,10. Personalized vaccines are built from the mutations in one person's tumor to train T cells against them18. Daraxonrasib works differently: it blocks the active form of RAS, both mutant and normal, and RAS mutations drive more than 90% of pancreatic cancers19. Vaccination and screening work earlier: high-risk HPV causes cervical cancer, so the vaccine heads it off, and screening looks for cancer before it spreads4,5.

What the research says

CAR-T: lasting remission in blood cancers

Early Two patients with chronic lymphocytic leukemia who reached complete remission in 2010 after CD19 CAR-T therapy still had detectable CAR-T cells more than ten years later, with sustained remission in both8. That shows a remission can last a decade. It is two people.

Moderate For a larger picture, the five-year follow-up of the single-arm ZUMA-1 trial covers 101 treated patients with refractory large B-cell lymphoma. Estimated five-year overall survival was 42.6%, and 31% were still in response at a median follow-up of 63.1 months9. The authors say the results support the "curative potential" of the treatment in a subset of patients. The data show a plateau, not a proven cure, and there was no control group.

CAR-T in a solid tumor: GD2 in diffuse midline gliomas

Early Diffuse midline gliomas with the H3K27M mutation include DIPG, a brainstem tumor. A phase 1 trial enrolled thirteen patients; eleven received GD2-targeted CAR-T cells into a vein, and those who benefited could get more infused into the fluid spaces of the brain. Nine had neurological benefit, four had major tumor reductions of 52%, 54%, 91%, and 100%, and one had a complete response ongoing more than 30 months after enrollment10. Three patients on the higher dose had dose-limiting cytokine release syndrome, and every patient had inflammation-related neurological side effects that needed intensive monitoring and care10. This is the most hopeful result I found in a solid tumor, and it is still a small safety trial with no control group.

Dostarlimab in mismatch repair-deficient rectal cancer

Moderate In 2022, 12 patients with mismatch repair-deficient stage II or III rectal cancer took dostarlimab every three weeks for six months, and all 12 had a clinical complete response, with no progression or recurrence over 6 to 25 months of follow-up17. A 2025 update included 117 patients with mismatch repair-deficient tumors. All 49 rectal cancer patients who finished treatment had a clinical complete response, 37 had one sustained at 12 months, and across the rectal and non-rectal groups, 82 of the 103 who finished treatment avoided surgery. Two-year recurrence-free survival was 92%2.

The global AZUR-1 trial, published in October 2026, enrolled 154 patients. Of the 144 with at least 12 months of follow-up, investigators judged that 68% had a complete response sustained for 12 months, or 76% counting patients with no tumor left at surgery, and no recurrences were reported in those responders over a median follow-up of 26.3 months20. The FDA's independent central review counted fewer: 47% of 116 evaluable patients, with another 38 patients (33%) in complete response but not yet followed for 12 months3. The trial's authors note that the central review's results were less mature20. On October 19, 2026, the FDA granted accelerated approval for adults with stage II or III rectal cancer that is MSI-H or mismatch repair-deficient by an FDA-authorized test; the recommended dose is 500 mg every three weeks for up to nine cycles3. An accelerated approval can later be verified or withdrawn, and the FDA's own pages track both outcomes3.

These results are in mismatch repair-deficient tumors, a subset of rectal cancers and 10% to 15% of nonmetastatic colon cancers17,21. In locally advanced mismatch repair-deficient colon cancer, the NICHE-2 trial gave nivolumab plus ipilimumab before planned surgery: 75 of 111 patients (68%) had no viable tumor left at surgery, and none had a recurrence at a median follow-up of 26 months21. The treatment came before surgery, so this is tumor kill, not a surgery-free cure.

Where it sits: durable remission with organ preservation, not yet a proven cure.

Melanoma and checkpoint inhibitors (CheckMate 067)

Moderate In advanced melanoma, the final analysis, with at least 10 years of follow-up, put median overall survival at 71.9 months with nivolumab plus ipilimumab, 36.9 months with nivolumab, and 19.9 months with ipilimumab alone1. Among patients alive and progression-free at 3 years, 10-year melanoma-specific survival was 96%, 97%, and 88%1. This is the strongest evidence I found that immunotherapy can produce remissions that last a decade, but those figures describe patients who were already doing well at 3 years, and not everyone gets there.

Personalized mRNA vaccines

Moderate in melanoma, Early in pancreatic cancer.

Melanoma. In the KEYNOTE-942 phase 2b trial of 157 patients, adding the individualized vaccine mRNA-4157 to pembrolizumab gave a hazard ratio for recurrence or death of 0.561 (95% CI 0.309 to 1.017, p = 0.053), with recurrence or death in 22% versus 40%22. That narrowly missed conventional statistical significance. On October 22, 2026, Merck and Moderna reported the phase 3 INTerpath-001 result in a company press release; the release says the results were accepted by the New England Journal of Medicine, but I did not find the paper among the sources I opened. In 1,137 patients with resected stage IIB to IV melanoma, adding the vaccine (now named intismeran autogene) to pembrolizumab gave a hazard ratio for recurrence or death of 0.58 (95% CI 0.45 to 0.75), and two-year recurrence-free survival was 82.2% versus 70.0%. The arms did not differ in the first six months, and overall survival data were not yet mature11. The vaccine is still investigational11. It lowers the risk of recurrence; it has not yet been shown to help people live longer.

Pancreatic cancer. In a phase 1 trial of the personalized vaccine autogene cevumeran after surgery, 8 of 16 patients made vaccine-induced T cells, and those responders had longer recurrence-free survival (median not reached versus 13.4 months)18. At a median 3.2 years of follow-up, the difference held (P = 0.007), two responders had recurred, and a median of 86% of each responder's vaccine-induced T-cell clones were still present at substantial levels about 3 years after vaccination12. The comparison is between responders and non-responders in 16 people, so it cannot prove the vaccine prevents recurrence.

RAS and KRAS inhibitors

Moderate for pancreatic cancer (longer survival), Moderate for lung (slowing). In the phase 3 RASolute 302 trial of 500 patients with previously treated metastatic pancreatic cancer, daraxonrasib extended median overall survival to 13.2 months from 6.6 months on chemotherapy in patients with RAS G12 mutations (hazard ratio 0.40)19. Across the whole trial, the medians were 13.2 and 6.7 months, the figures the FDA cites. The FDA approved it on August 26, 2026 for adults with metastatic pancreatic adenocarcinoma who have had at least one prior systemic therapy or cannot take multidrug therapy, citing median progression-free survival of 7.2 versus 3.6 months and a recommended dose of 300 mg once daily13. That roughly doubles median survival in a cancer where it is short. It is not remission or cure.

In KRAS G12C lung cancer, a phase 3 trial of 345 patients found sotorasib beat docetaxel on progression-free survival (5.6 versus 4.5 months, hazard ratio 0.66)14. That is slowing.

HPV vaccination and cervical cancer

Moderate In Scotland, no invasive cervical cancers were recorded in women vaccinated at 12 or 13. For those vaccinated at 14 to 22 with 3 doses, the rate was 3.2 per 100,000 person-years versus 8.4 in the unvaccinated4. In England, cervical cancer was 87% lower in women offered the vaccine at 12 to 13, and CIN3 precancer was 97% lower15. A Swedish cohort of 1,672,983 girls and women found an incidence rate ratio of 0.12 for vaccination before age 1716. These are observational studies, not randomized trials, but large ones in three populations agree.

Screening: lung, colon, and the multi-cancer blood test

Lung. Strong. The U.S. NLST randomized 53,454 heavy smokers to low-dose CT or chest X-ray. CT reduced lung-cancer deaths by 20.0% and deaths from any cause by 6.7%, but 24.2% of CT screens were positive, and 96.4% of those positives were false positives5. The NELSON trial found a rate ratio of 0.76 for lung-cancer death in men at 10 years, and 0.67 (not statistically significant) in women6. The U.S. Preventive Services Task Force recommends annual low-dose CT for adults 50 to 80 with a 20 pack-year history who smoke now or quit within 15 years (a 2021 statement)23.

Colonoscopy. Moderate, and humbling. In NordICC, an invitation to a one-time colonoscopy cut 10-year colorectal cancer risk from 1.20% to 0.98% (risk ratio 0.82), but colorectal cancer deaths were 0.28% versus 0.31% (risk ratio 0.90, 95% CI 0.64 to 1.16). Only 42.0% of those invited were screened24. The trial measured what an invitation does, and fewer than half went, so it is not a direct test of what a colonoscopy does for someone who has one.

Multi-cancer blood tests (Galleri). Mixed. The numbers matter here.

  • PATHFINDER 2 (35,878 people, single-arm): positive predictive value 60.3%, specificity 99.64%, and episode sensitivity 39.3% for all cancers (69.8% for a 12-cancer subgroup). After a positive test, 213 people (0.6%) had invasive procedures25. Roughly four in ten cancers were caught per testing episode, and six in ten positives were cancer25.
  • NHS-Galleri, the randomized trial of 142,250 people funded by GRAIL, did not reduce stage III or IV cancer, its primary end point (incidence rate ratio 1.03, 95% CI 0.92 to 1.14). Stage IV came out at 0.86 (95% CI 0.74 to 1.00)7. By round, positive predictive value was 58.0%, 50.4%, and 45.8%, specificity was 99.50% to 99.60%, and episode sensitivity for all cancers was 26.7% to 37.2%26. GRAIL's press release led with the stage IV result, put the false positive rate at 0.45%, and said in its subhead that the primary end point showed no reduction27.
  • Regulatory status as of November 2026: I found no FDA approval. GRAIL's May 30, 2026 release says the test has not been cleared or approved by the FDA27. An FDA advisory panel met on September 23, 202628. According to GRAIL's press release, the panel voted 7 to 2, with one abstention, that the benefits outweigh the risks, unanimously on safety, and 6 to 4 on effectiveness; GRAIL submitted its application on January 29, 2026, the FDA is not bound by the vote, and a final decision was expected in the coming months29.

What is unproven: that using the test lowers deaths from cancer. None of these trials has shown that, and the PATHFINDER 2 authors say randomized trials and longer follow-up are needed to judge its clinical usefulness7,25. GRAIL says the test is meant to add to guideline screening, not replace it29.

From Rhonda Patrick's interview with Peter Attia

In a FoundMyFitness clip posted January 10, 2024, from Rhonda Patrick's interview with Peter Attia, Attia says screening carries financial, procedural, and false-positive harms, that the Galleri test's sensitivity for breast cancer looks low at first glance while its specificity is very high, and that he uses it alongside imaging, not instead of it. The clip's summary says the validation evidence supports detection performance, not a mortality benefit30. That conversation predates the 2026 trials above.

Cautions

  • None of this is advice. These are trial results and approvals, not recommendations for any person. Anyone with a cancer diagnosis needs an oncologist who knows their tumor.
  • A response is not a cure. Complete response, remission, and survival are different outcomes, and the follow-up on the newest results here is about two to three years20,12.
  • Subtype matters. The dostarlimab results are in mismatch repair-deficient tumors, a subset of rectal cancers, and sotorasib was tested only in lung cancers with the KRAS G12C mutation17,14.
  • Press releases are not papers. I label company results as such and note where independent review gave a lower number than the investigators did3.
  • Screening has harms. False positives, extra scans, and invasive procedures are real, and a test that finds more cancer is not the same as one that saves lives5,25.
  • Things move. The FDA decisions and trial results on this page are as of November 2026.

References

  1. Wolchok JD, Chiarion-Sileni V, Rutkowski P, Cowey CL, Schadendorf D, Wagstaff J, et al. Final, 10-Year Outcomes with Nivolumab plus Ipilimumab in Advanced Melanoma. The New England Journal of Medicine. 2025;392(1):11-22.Randomized trialPubMed 39282897 · DOI

    What it found: Final CheckMate 067 analysis with a minimum follow-up of 10 years in previously untreated advanced melanoma. Median overall survival was 71.9 months with nivolumab plus ipilimumab, 36.9 months with nivolumab, and 19.9 months with ipilimumab (hazard ratio for death 0.53, 95% CI 0.44 to 0.65, and 0.63, 95% CI 0.52 to 0.76, versus ipilimumab); among patients alive and progression-free at 3 years, 10-year melanoma-specific survival was 96%, 97%, and 88%.

    Limits: Both comparisons are against ipilimumab alone, and the 3-year landmark figures describe a selected group of patients who had already done well.

  2. Cercek A, Foote MB, Rousseau B, Smith JJ, Shia J, Sinopoli J, et al. Nonoperative Management of Mismatch Repair-Deficient Tumors. The New England Journal of Medicine. 2025;392(23):2297-2308.Cohort studyPubMed 40293177 · DOI

    What it found: Phase 2 study of six months of dostarlimab in 117 patients with stage I to III mismatch repair-deficient tumors. In the rectal cohort 49 patients completed treatment and all had a clinical complete response, and 37 had a sustained response at 12 months; across both cohorts 82 of the 103 who completed treatment avoided surgery. Two-year recurrence-free survival was 92% (95% CI 86 to 99), with median follow-up for recurrence of 20.0 months.

    Limits: Single-arm, with a median follow-up for recurrence of 20.0 months, so late recurrences and the long-term cure rate are not yet known.

  3. U.S. Food and Drug Administration. FDA grants accelerated approval to dostarlimab for locally advanced rectal cancer. FDA, Resources for Information, Approved Drugs, October 19, 2026.Regulatory noticeRead

    What it found: On October 19, 2026 the FDA granted accelerated approval to dostarlimab (Jemperli) for adults with stage II or III locally advanced rectal cancer that is MSI-H or dMMR, based on the single-arm AZUR-1 trial of 154 treated patients, at 500 mg every three weeks for up to nine cycles. Sustained clinical complete response at 12 months was 47% by independent central review (116 evaluable patients) and 68% by investigator assessment (144 patients), with median follow-up of 26.3 months.

    Limits: An accelerated approval based on one single-arm trial, with no survival or cure data yet; the independent review rate (47%) trails the investigator rate (68%), and the page notes 38 more patients in complete response by independent review with under 12 months of follow-up.

  4. Palmer TJ, Kavanagh K, Cuschieri K, Cameron R, Graham C, Wilson A, et al. Invasive cervical cancer incidence following bivalent human papillomavirus vaccination: a population-based observational study of age at immunization, dose, and deprivation. Journal of the National Cancer Institute. 2024;116(6):857-865.Cohort studyPubMed 38247547 · DOI

    What it found: Scottish population-based study of women born between January 1, 1988 and June 5, 1996, with data extracted in July 2020. No invasive cervical cancers were recorded in women vaccinated at age 12 or 13, regardless of dose number; for women vaccinated at 14 to 22 with 3 doses the rate was 3.2 per 100,000 person-years (95% CI 2.1 to 4.6), versus 8.4 per 100,000 person-years (95% CI 7.2 to 9.6) in unvaccinated women.

    Limits: An observational registry study, not randomized, and the women vaccinated youngest were only in their mid-twenties when the data were pulled.

  5. National Lung Screening Trial Research Team, Aberle DR, Adams AM, Berg CD, Black WC, Clapp JD, et al. Reduced lung-cancer mortality with low-dose computed tomographic screening. The New England Journal of Medicine. 2011;365(5):395-409.Randomized trialPubMed 21714641 · DOI

    What it found: Randomized trial of 53,454 high-risk U.S. participants given three annual low-dose CT scans (26,722) or chest radiographs (26,732). Lung-cancer deaths fell by 20.0% (95% CI 6.8 to 26.7) and all-cause deaths by 6.7% (95% CI 1.2 to 13.6) with CT, but 24.2% of CT screens were positive and 96.4% of those positives were false positives.

    Limits: Done at 33 U.S. medical centers in heavy smokers, so real-world benefit depends on who is screened and how nodules are followed.

  6. de Koning HJ, van der Aalst CM, de Jong PA, Scholten ET, Nackaerts K, Heuvelmans MA, et al. Reduced Lung-Cancer Mortality with Volume CT Screening in a Randomized Trial. The New England Journal of Medicine. 2020;382(6):503-513.Randomized trialPubMed 31995683 · DOI

    What it found: NELSON randomized 13,195 men aged 50 to 74 to CT screening or no screening, with at least 10 years of follow-up. Lung-cancer mortality was 2.50 versus 3.30 per 1,000 person-years, a rate ratio of 0.76 (95% CI 0.61 to 0.94) at 10 years; in the 2,594 women the rate ratio was 0.67 (95% CI 0.38 to 1.14).

    Limits: The primary analysis was in men, and the result in women was not statistically significant.

  7. Sasieni P, Johnson P, Round T, Warwick J, Jones H, Dolly S, et al. Effect of Screening with Multicancer Early-Detection Test on Late-Stage Cancer Diagnosis. The New England Journal of Medicine. 2026;395(14):1374-1386.Randomized trialPubMed 42776074 · DOI

    What it found: NHS-Galleri randomized people aged 50 to 77 to a multi-cancer blood test (71,122) or usual care (71,128) for up to three annual visits. The primary end point, stage III or IV cancer across 12 prespecified cancers, was not reduced (incidence rate ratio 1.03, 95% CI 0.92 to 1.14, P = 0.63); stage IV cancer had an incidence rate ratio of 0.86 (95% CI 0.74 to 1.00).

    Limits: The trial missed its primary end point, the stage IV result sits at the edge of statistical significance, its end points are cancer stage rather than cancer deaths, and GRAIL funded it.

  8. Melenhorst JJ, Chen GM, Wang M, Porter DL, Chen C, Collins MA, et al. Decade-long leukaemia remissions with persistence of CD4+ CAR T cells. Nature. 2022;602(7897):503-509.Lab or mechanistic studyPubMed 35110735 · DOI

    What it found: Case study of two patients with chronic lymphocytic leukemia who reached complete remission in 2010 after CD19-directed CAR T cells and still had detectable CAR T cells more than ten years later with sustained remission. A highly activated CD4+ population came to dominate, with a repertoire led by a few clones.

    Limits: Two patients only, selected because they responded, so it shows that a lasting remission can happen but not how often.

  9. Neelapu SS, Jacobson CA, Ghobadi A, Miklos DB, Lekakis LJ, Oluwole OO, et al. Five-year follow-up of ZUMA-1 supports the curative potential of axicabtagene ciloleucel in refractory large B-cell lymphoma. Blood. 2023;141(19):2307-2315.Cohort studyPubMed 36821768 · DOI

    What it found: Five-year analysis of the single-arm phase 2 ZUMA-1 trial of axicabtagene ciloleucel in 101 treated patients with refractory large B-cell lymphoma: 83% objective response, 58% complete response, and 31% still in response at a median follow-up of 63.1 months. Estimated 5-year overall survival was 42.6% and 5-year disease-specific survival was 51.0%.

    Limits: A single-arm trial with no control group, and the "curative potential" in the title is the authors' reading of a plateau, not a proven cure.

  10. Monje M, Mahdi J, Majzner R, Yeom KW, Schultz LM, Richards RM, et al. Intravenous and intracranial GD2-CAR T cells for H3K27M+ diffuse midline gliomas. Nature. 2025;637(8046):708-715.Cohort studyPubMed 39537919 · DOI

    What it found: Phase 1 trial (published online November 13, 2024; in print January 2025) of GD2-CAR T cells that enrolled thirteen patients with H3K27M+ diffuse midline glioma (DIPG and spinal), 11 of whom received intravenous cells, 3 at the lower dose and 8 at the higher dose; 3 at the higher dose had dose-limiting cytokine release syndrome. Nine patients had neurological benefit, four had major tumor reductions (52%, 54%, 91%, and 100%), and one had a complete response ongoing more than 30 months after enrollment.

    Limits: A small phase 1 safety trial with no control group, so it cannot show a survival benefit or how many responses last.

  11. Merck & Co., Inc.; Moderna, Inc. Merck and Moderna Will Present Results From Landmark Phase 3 INTerpath-001 Trial for Intismeran Autogene Plus KEYTRUDA (pembrolizumab) at ESMO 2026 Presidential Symposium. Merck news release, October 22, 2026.ArticleRead

    What it found: Company press release of October 22, 2026, not a peer-reviewed paper; it says the results were accepted for publication in the New England Journal of Medicine. In the phase 3 INTerpath-001 trial of 1,137 patients with resected stage IIB to IV melanoma, adding intismeran autogene to pembrolizumab gave a recurrence-free survival hazard ratio of 0.58 (95% CI 0.45 to 0.75) at 24.6 months median follow-up, with 2-year recurrence-free survival of 82.2% versus 70.0%; overall survival was not yet mature.

    Limits: A sponsor press release with no peer review yet, no overall survival result, and no difference between the arms in the first six months.

  12. Sethna Z, Guasp P, Reiche C, Milighetti M, Ceglia N, Patterson E, et al. RNA neoantigen vaccines prime long-lived CD8+ T cells in pancreatic cancer. Nature. 2025;639(8056):1042-1051.Cohort studyPubMed 39972124 · DOI

    What it found: Extended follow-up of the phase 1 pancreatic cancer vaccine trial at a median of 3.2 years: 8 responders with vaccine-induced T cells and 8 non-responders, with median recurrence-free survival not reached in responders versus 13.4 months in non-responders (P = 0.007). Eighty-six percent of vaccine-induced T-cell clones per patient persisted at substantial levels about 3 years after vaccination, and two responders recurred.

    Limits: Sixteen patients in one phase 1 trial and a responder-versus-non-responder comparison that was not randomized.

  13. U.S. Food and Drug Administration. FDA approves daraxonrasib for metastatic pancreatic adenocarcinoma. FDA, Resources for Information, Approved Drugs, August 26, 2026.Regulatory noticeRead

    What it found: On August 26, 2026 the FDA approved daraxonrasib for adults with metastatic pancreatic adenocarcinoma who have had at least one prior systemic therapy or are not candidates for multiagent therapy, at 300 mg by mouth once daily on the label. In RASolute 302 (500 patients), median overall survival was 13.2 versus 6.7 months (hazard ratio 0.40, 95% CI 0.30 to 0.53), progression-free survival was 7.2 versus 3.6 months, and objective response was 30% versus 11%.

    Limits: A regulatory summary of one open-label trial; it lists warnings (skin and soft tissue toxicity, mouth sores, diarrhea, lung inflammation) and says nothing about cure.

  14. de Langen AJ, Johnson ML, Mazieres J, Dingemans AC, Mountzios G, Pless M, et al. Sotorasib versus docetaxel for previously treated non-small-cell lung cancer with KRAS G12C mutation: a randomised, open-label, phase 3 trial. The Lancet. 2023;401(10378):733-746.Randomized trialPubMed 36764316 · DOI

    What it found: CodeBreaK 200 randomized 345 patients with previously treated KRAS G12C-mutated lung cancer to sotorasib (171) or docetaxel (174). Median progression-free survival was 5.6 versus 4.5 months (hazard ratio 0.66, 95% CI 0.51 to 0.86, p = 0.0017), with serious treatment-related adverse events in 11% versus 23%.

    Limits: The abstract reports a gain of about one month in progression-free survival and no overall survival result, so this is slowing, not remission or cure.

  15. Falcaro M, Castañon A, Ndlela B, Checchi M, Soldan K, Lopez-Bernal J, et al. The effects of the national HPV vaccination programme in England, UK, on cervical cancer and grade 3 cervical intraepithelial neoplasia incidence: a register-based observational study. The Lancet. 2021;398(10316):2084-2092.Cohort studyPubMed 34741816 · DOI

    What it found: Register-based study of women in England aged 20 to under 30 (13.7 million person-years). Cervical cancer was reduced by 34% (95% CI 25 to 41) in those offered the vaccine at 16 to 18, 62% (52 to 71) at 14 to 16, and 87% (72 to 94) at 12 to 13; grade 3 cervical precancer (CIN3) fell by 39%, 75%, and 97%.

    Limits: An observational comparison with unvaccinated cohorts, with follow-up only into the early adult years.

  16. Lei J, Ploner A, Elfström KM, Wang J, Roth A, Fang F, et al. HPV Vaccination and the Risk of Invasive Cervical Cancer. The New England Journal of Medicine. 2020;383(14):1340-1348.Cohort studyPubMed 32997908 · DOI

    What it found: Swedish registry cohort of 1,672,983 girls and women aged 10 to 30, followed from 2006 through 2017. Cumulative incidence of cervical cancer by age 31 was 47 per 100,000 in vaccinated women versus 94 per 100,000 in unvaccinated women; the adjusted rate ratio was 0.12 (95% CI 0.00 to 0.34) for vaccination before age 17 and 0.47 (95% CI 0.27 to 0.75) at ages 17 to 30.

    Limits: Observational, with only 19 cancers among vaccinated women, and follow-up that ends at age 31.

  17. Cercek A, Lumish M, Sinopoli J, Weiss J, Shia J, Lamendola-Essel M, et al. PD-1 Blockade in Mismatch Repair-Deficient, Locally Advanced Rectal Cancer. The New England Journal of Medicine. 2022;386(25):2363-2376.Cohort studyPubMed 35660797 · DOI

    What it found: Phase 2 study of six months of dostarlimab every 3 weeks in mismatch repair-deficient stage II or III rectal cancer. All 12 patients who finished treatment with at least 6 months of follow-up had a clinical complete response (95% CI 74 to 100), with follow-up of 6 to 25 months and no adverse events of grade 3 or higher.

    Limits: Twelve patients followed 6 to 25 months, so it cannot show how long the responses last.

  18. Rojas LA, Sethna Z, Soares KC, Olcese C, Pang N, Patterson E, et al. Personalized RNA neoantigen vaccines stimulate T cells in pancreatic cancer. Nature. 2023;618(7963):144-150.Cohort studyPubMed 37165196 · DOI

    What it found: Phase 1 trial in resected pancreatic cancer: 16 patients received the personalized vaccine autogene cevumeran with atezolizumab and 15 received mFOLFIRINOX. Vaccine-induced neoantigen-specific T cells appeared in 8 of 16 patients, and at 18 months median follow-up non-responders had a median recurrence-free survival of 13.4 months while the responders' median had not been reached (P = 0.003).

    Limits: A small, single-arm phase 1 trial where the responder comparison is not randomized, so it cannot show the vaccine itself prevents recurrence.

  19. O'Reilly EM, Wainberg ZA, Hendifar AE, Borad MJ, Pietrantonio F, Pant S, et al. Daraxonrasib or Chemotherapy in Previously Treated Metastatic Pancreatic Cancer. The New England Journal of Medicine. 2026;395(4):325-337.Randomized trialPubMed 42223072 · DOI

    What it found: Randomized phase 3 RASolute 302 trial of 500 patients with previously treated metastatic pancreatic cancer (248 on daraxonrasib, 252 on chemotherapy), 91.8% with RAS G12 mutations. In the RAS G12 population median overall survival was 13.2 versus 6.6 months (hazard ratio 0.40, P<0.001) and median progression-free survival was 7.3 versus 3.5 months.

    Limits: An open-label trial in second-line patients that measures survival in months, so it shows longer survival and not remission or cure.

  20. Cercek A, Bachet JB, Folprecht G, Bando H, Seligmann J, Starling N, et al. Clinical Complete Response with Dostarlimab Monotherapy in dMMR/MSI-H Locally Advanced Rectal Cancer: First Results from the Global AZUR-1 Trial. Annals of Oncology. 2026.Cohort studyPubMed 42927858 · DOI

    What it found: Interim analysis (data cutoff April 20, 2026) of the single-arm, registrational AZUR-1 trial of dostarlimab 500 mg every 3 weeks for 9 cycles: 154 patients enrolled, median follow-up 26.3 months, and 144 with at least 12 months of follow-up. Investigator-assessed sustained clinical complete response at 12 months was 68%, and 76% counting pathologic complete responses; grade 3 or higher immune-mediated adverse events occurred in 3%.

    Limits: An interim, single-arm analysis published ahead of print; the investigator-assessed rate is higher than the FDA's independent central review rate (47%), which the authors say was less mature.

  21. Chalabi M, Verschoor YL, Tan PB, Balduzzi S, Van Lent AU, Grootscholten C, et al. Neoadjuvant Immunotherapy in Locally Advanced Mismatch Repair-Deficient Colon Cancer. The New England Journal of Medicine. 2024;390(21):1949-1958.Cohort studyPubMed 38838311 · DOI

    What it found: NICHE-2, a single-arm phase 2 trial of neoadjuvant immunotherapy before surgery in 115 patients with locally advanced mismatch repair-deficient colon cancer. Of 111 patients in the efficacy analysis, 109 (98%) had a pathological response and 75 (68%) had a pathological complete response, and no patient had recurrence at a median follow-up of 26 months.

    Limits: Single-arm with a median follow-up of just over two years, and treatment was given before planned surgery, so it shows tumor kill, not a surgery-free cure.

  22. Weber JS, Carlino MS, Khattak A, Meniawy T, Ansstas G, Taylor MH, et al. Individualised neoantigen therapy mRNA-4157 (V940) plus pembrolizumab versus pembrolizumab monotherapy in resected melanoma (KEYNOTE-942): a randomised, phase 2b study. The Lancet. 2024;403(10427):632-644.Randomized trialPubMed 38246194 · DOI

    What it found: Randomized phase 2b trial in 157 patients with completely resected high-risk melanoma (107 on mRNA-4157 plus pembrolizumab, 50 on pembrolizumab alone). Recurrence or death occurred in 24 (22%) versus 20 (40%), hazard ratio 0.561 (95% CI 0.309 to 1.017, p = 0.053), and 18-month recurrence-free survival was 79% versus 62%.

    Limits: A phase 2b trial where the primary comparison missed conventional statistical significance (p = 0.053) and follow-up was about two years.

  23. US Preventive Services Task Force, Krist AH, Davidson KW, Mangione CM, Barry MJ, Cabana M, et al. Screening for Lung Cancer: US Preventive Services Task Force Recommendation Statement. JAMA. 2021;325(10):962-970.GuidelinePubMed 33687470 · DOI

    What it found: The Task Force recommends annual low-dose CT screening for adults aged 50 to 80 with a 20 pack-year smoking history who smoke now or quit within the past 15 years (grade B), replacing the 2013 statement that set 55 to 80 and 30 pack-years.

    Limits: A 2021 statement that I did not check against any newer update, and it covers only people with a heavy smoking history.

  24. Bretthauer M, Løberg M, Wieszczy P, Kalager M, Emilsson L, Garborg K, et al. Effect of Colonoscopy Screening on Risks of Colorectal Cancer and Related Death. The New England Journal of Medicine. 2022;387(17):1547-1556.Randomized trialPubMed 36214590 · DOI

    What it found: NordICC randomized people aged 55 to 64 in Poland, Norway, and Sweden, inviting 28,220 to a one-time colonoscopy (11,843, or 42.0%, were screened) and following 56,365 as usual care. At a median of 10 years the colorectal cancer risk was 0.98% versus 1.20% (risk ratio 0.82, 95% CI 0.70 to 0.93), and the risk of colorectal cancer death was 0.28% versus 0.31% (risk ratio 0.90, 95% CI 0.64 to 1.16).

    Limits: Only 42.0% of those invited were screened, so it measures an invitation rather than a colonoscopy, and the confidence interval for colorectal cancer death is wide.

  25. Nabavizadeh N, McDonnell C, Kurbegov D, Matrana M, Gadgeel S, Kim RH, et al. Performance and safety of a multi-cancer early detection test: the PATHFINDER 2 study. Nature Medicine. 2026;32(10):3760-3769.Cohort studyPubMed 42773207 · DOI

    What it found: Prospective single-arm study of 35,878 people aged 50 or older, with 32,007 analyzable for performance. The cancer detection rate was 0.54%, positive predictive value 60.3% (95% CI 54.5 to 65.8), specificity 99.64%, and episode sensitivity 39.3% for all cancers and 69.8% for a prespecified 12-cancer subgroup; 213 participants (0.6%) had invasive procedures after a positive test.

    Limits: A single-arm study with 12 months of follow-up and several authors who work for or hold stock in the test's maker, and it cannot show that the test saves lives.

  26. Neal RD, Dolly S, Johnson P, Jones H, Kumar SH, Lee LYW, et al. Performance of a multi-cancer early detection test in the randomized controlled NHS-Galleri trial. Nature Medicine. 2026;32(10):3770-3780.Randomized trialPubMed 42773209 · DOI

    What it found: Test-performance analysis of the intervention arm of NHS-Galleri (142,250 participants randomized in all). Positive results were returned for 722 of 70,325 (1.03%) in round 1, 518 of 64,498 (0.80%) in round 2, and 561 of 62,323 (0.90%) in round 3, with 937 cancers found and positive predictive value of 58.0%, 50.4%, and 45.8% by round; specificity was 99.50% to 99.60%.

    Limits: Describes how the test performed in the intervention arm, not whether screening improves outcomes.

  27. GRAIL, Inc. GRAIL Reports Full Results From NHS-Galleri Trial Demonstrating Substantial Reduction in Stage IV Cancer Diagnoses at 2026 ASCO Annual Meeting. GRAIL press release, May 30, 2026.ArticleRead

    What it found: Company press release, not peer-reviewed at the time. In NHS-Galleri (142,250 participants aged 50 to 77) the test missed its primary endpoint (incidence rate ratio 1.03, 95% CI 0.92 to 1.14), while stage IV diagnoses fell 14% (incidence rate ratio 0.86); 1,801 participants had a positive result, 937 were diagnosed with cancer, positive predictive value was 52.0%, and the false positive rate was 0.45%.

    Limits: A sponsor press release whose headline stresses the secondary result, and the release itself says the test has not been cleared or approved by the FDA.

  28. U.S. Food and Drug Administration. September 23, 2026: Molecular and Clinical Genetics Panel of the Medical Devices Advisory Committee Meeting Announcement. FDA Advisory Committee Calendar, September 23, 2026.Regulatory noticeRead

    What it found: FDA notice that a device advisory panel would meet on September 23, 2026 to discuss the Galleri test under a premarket approval application, described as a prescription-only, sequencing-based test of cell-free DNA methylation for early detection of multiple cancer types in adults aged 50 or older, to be followed by diagnostic workup.

    Limits: A meeting announcement written before the meeting, so it records the proposed use but not the panel's vote or the FDA's decision.

  29. GRAIL, Inc. FDA Advisory Committee Votes In Favor of Approval of GRAIL's Galleri Multi-Cancer Early Detection Test. GRAIL press release, September 23, 2026.ArticleRead

    What it found: Company press release reporting that the FDA's device panel voted 7 to 2 with 1 abstention that benefits outweigh risks, unanimously on safety, and 6 to 4 on effectiveness; GRAIL submitted its premarket approval application on January 29, 2026, and the FDA's decision was still pending, expected in the coming months.

    Limits: A sponsor's account of the vote, which does not mention cancer deaths, and the FDA is not bound by the panel's advice.

  30. Patrick R, Attia P. How Cancer Risk and Screening Shape Early Detection. FoundMyFitness, January 10, 2024.PodcastListen

    What it found: Clip from Rhonda Patrick's interview with Peter Attia, in which Attia says screening carries financial, procedural, and false-positive harms, that the Galleri blood test's sensitivity for breast cancer looks low at first glance while its specificity is very high, and that he uses it alongside imaging, not instead of it; the page notes the evidence supports detection performance but not a mortality benefit.

    Limits: A January 2024 conversation that predates the 2026 trial results, and it describes one physician's personal screening practice, not a recommendation.