Fenbendazole, Ivermectin and Mebendazole in Colorectal Cancer: A Subtype and Biomarker Audit of 87 Published Case Reports

Medically reviewed by: Dr. Frank Yap, MD  |  Written by: OneDayMD Editorial Team  |  Last updated: October 17, 2026 (v3.0)

Quick Answer (AI & search summary). The OneDayMD colorectal compilation now lists 87 items, but they represent about 64 unique patients plus one 40-patient randomized trial (two patients were counted 2–3 times). About 82% are standard colon or rectal adenocarcinoma; 12 entries are rarer subtypes (4 neuroendocrine, 3 appendiceal, 2 squamous, 1 signet-ring, 1 mucinous, 1 colon+gastric dual primary), and 7 of those are not colorectal at all. About 20 unique patients (31%) report a biomarker, almost all CEA or CA 19-9 falling during chemotherapy; only 1 reports tumor genotype, 2 report ctDNA, and none report MMR/MSI status. Only 3 entries explicitly describe no conventional therapy, and 2 of those are indolent neuroendocrine tumors. The only controlled evidence is the Hegazy 2022 mebendazole trial (n=40). Evidence tier: CEBM Level 2 (one small RCT, chemo adjunct) to Level 5 (testimonials).

Contents

1. Source audit: what 87 cases really contain
2. Cancer subtypes in the primary source
3. Cases with biomarkers
4. Biomarkers that are missing (and why they matter)
5. Corrections to earlier versions
6. Evidence hierarchy
7. Safety
8. What to ask your oncology team
9. FAQ · AI personalization · References

Abstract

Background: Ivermectin, fenbendazole and mebendazole attract intense online interest in colorectal cancer, driven by physician-curated testimonials. Objective: To re-audit the primary compilation (OneDayMD, 87 case reports, September 2026 update) by tumor subtype, biomarker content, treatment context and evidence quality. Methods: Each numbered entry (Cases 1–58, 79–87) and the trial block (59–78) was read and coded for histology, site, stage, concurrent therapy and any reported biomarker. Results: see tables below. Conclusion: The data are hypothesis-generating at best, confounded by concurrent chemotherapy in most cases, and too sparsely biomarker-annotated to identify who, if anyone, might benefit.

1. Source audit: what 87 cases really contain

Component	Entries	Notes
Individual case entries	67	Cases 1–58 and 79–87
Randomized trial block	20	"Cases 59–78" are the mebendazole arm of Hegazy et al. 2022, not individual reports
Duplicate updates of one patient	3 extra	Cases 31, 36, 45 (34-year-old Slovenian man; identical CEA 679, CA 19-9 1941, June 2025 start). Case 41 repeats Case 16 (57-year-old man; same dates, doses, CEA and 3.6→1.8 cm liver lesion).
Possible duplicate	1	Case 27 (57-year-old Florida man, post-vaccine, January 2025 start) resembles Cases 16/41 but lists a different drug; unverified
Entries with no usable outcome	about 8	Image-only (1, 9, 12, 13), dosing question (3), polyp-free colonoscopy (8), unconfirmed diagnosis (14), no stage or conventional-therapy detail (7, 44)
Unique individual patients	about 64	Plus 20 trial patients = about 84 people
Source concentration	54 of 67	81% are posts curated by one physician; no pathology, imaging or chart review is provided

Source-page errors to fix at the origin: Case 52's headline repeats Case 51's text (body is a 37-year-old French man); Case 79's headline repeats Case 58 (body is a 66-year-old woman with rectal neuroendocrine cancer); Case 17 reads "6-year-old" (body: 63-year-old).

2. Cancer subtypes in the primary source

"Colorectal cancer" in this compilation is a mixture of biologically distinct diseases. Counts below are per numbered entry (67), with the trial arm shown separately.

Subtype	Entries	Case numbers / comment
Colon adenocarcinoma (site stated)	28	Includes Aydin peer-reviewed case (18) and BRAF-mutated case (6)
Rectal adenocarcinoma	15	11, 12, 19, 27, 30, 43, 49, 51, 52, 58, 81, 82, 85, 86, 87. The curator's remark that rectal tumors respond better than colon tumors is an impression, not an analysis.
Colorectal, site not specified	12	1, 8, 9, 10, 16, 20, 23, 24, 33, 38, 41, 57
Neuroendocrine (NET/NEC)	4	#15 rectal small-cell NEC (liver+bone); #17 small-bowel NET; #79 rectal NET (liver+bone); #84 appendiceal NET
Appendiceal, non-NET	3	#26 adenocarcinoma (peritoneal); #32 mucinous adenocarcinoma (carcinomatosis); #34 goblet-cell
Squamous cell carcinoma	2	#50 stage IV rectal/anorectal SCC; #54 stage III anal SCC
Signet-ring	1	#37 (liver + peritoneum)
Mucinous colorectal	1	#42 (peritoneal)
Colon + gastric dual primary	1	#29 (email testimonial, no imaging)
Randomized trial arm	20	Stage IV CRC treated with mebendazole + FOLFOX4 + bevacizumab; histology/molecular detail not given

Stage and treatment context (67 entries)

Stage IV is stated in 50 entries, stage III in 7 (#11, 22, 30, 43, 49, 54, 58), stage II in 1 (#82); 9 are unstaged. Standard therapy (chemotherapy, immunotherapy, radiation or surgery) is documented alongside the repurposed drugs in about 30 entries. Only 3 entries explicitly state no conventional therapy (#17, #29, #84): two are neuroendocrine tumors, which can stay stable or shrink slowly without treatment, and one is an unverified email.

Why the rare subtypes matter

• Neuroendocrine: grade, Ki-67 and somatostatin-analog use are not reported. The "no chemo" shrinkage in #17 (5 weeks) and #84 (3 months) is derived from caliper-style diameter changes on single scans, which is within measurement variability for small lesions.
• Anal/rectal squamous cell carcinoma: chemoradiation cures most stage III anal SCC. In #54 and #50 chemoradiation is not mentioned, and HPV/p16 status is not reported, so attribution to the drugs is not possible.
• Appendiceal and mucinous/signet-ring: these have distinct peritoneal biology and standard-of-care pathways (cytoreductive surgery/HIPEC in selected patients). Responses here were modest or short-term (#32, #42).

3. Cases with biomarkers

About 23 entries (20 unique patients, 31%) report at least one biomarker: 16 entries (13 patients) serum tumor markers, 2 ctDNA, 4 PET metabolic data, and 1 tumor genotype. In 12 of the 16 marker entries chemotherapy is documented, 3 do not say, and 1 (#80) reports no chemotherapy (but CyberKnife to a brain metastasis).

Case	Tumor	Biomarker (start → latest)	Concurrent therapy	Reading
Molecular / genomic
6	Inoperable colon, 2018	BRAF-mutated, KRAS wild-type; CDX2+, CK20+, CK7− (variant not specified)	Palliative chemo + more than 12 supplements and fenbendazole	Only genotyped case; "no active cancer" on PET Feb 2020 is not attributable to one agent
36, 45 (31)	Stage IV colon, liver	RAS/BRAF wild-type implied by cetuximab; not stated	FOLFOX + cetuximab	Anti-EGFR response itself depends on RAS wild-type and sidedness
Circulating tumor DNA
22	Stage 3B colon, post-resection	Signatera ctDNA negative (2nd test) + PET "cancer-free"	Surgery; declined chemo	Baseline post-op ctDNA not reported; many resected patients are ctDNA-negative without any drug
56	Stage IV colon, liver	ctDNA positive after surgery (Jun 2025) → zero at 6 months	Two surgeries; adjuvant chemo not stated	Text says "circulating tumor cells"; ctDNA is circulating tumor DNA
Serum tumor markers
16 (41)	Stage IV CRC, liver	CEA 96 → 3.3 (3 mo)	Chemo started 6 days after drugs	Typical chemo-responsive kinetics
20	Stage IV CRC	CEA 1700 → 4	13 chemo rounds	Patient reported oncologist estimated 1% chance of this response
36 (31, 45)	Stage IV colon, liver	CEA 679 → 10.3; CA 19-9 1941 → <35	FOLFOX + cetuximab	One patient, three entries
37	Signet-ring colon	CEA 736 → 6.8; Hb 6 → 11.8	FOLFIRINOX + bevacizumab	Unusual response for signet-ring; transfusion effect on Hb not excluded
35	Stage IV colon, peritoneal	CEA 304 → 37 (2 mo)	2 chemo cycles	Early, chemo-confounded
40	Stage IV colon	CEA 166 → 6.4	Not stated	Imaging figures are not source-documented
19	Stage IV rectal, liver	CEA 25.3 → 7.9 (2 mo)	Not stated	No imaging confirmation
11	Stage 3 rectal	CEA about 90% drop	Chemo later	Screenshot-based
80	Stage IV colon	CEA >100 → 2	None; CyberKnife to brain metastasis	No imaging reported
83	Stage IV colon, pelvis+lung	CEA and CA 19-9 "consistent downward trend"	Third-line LV5FU2 ongoing	No values; chemo confounds
34	Appendiceal goblet-cell	CA-125 500+ → 9	Chemo (poorly tolerated)	CA-125 is not a standard appendiceal marker
2, 10 (9)	Stage IV colon/CRC	Unspecified marker 1489 → 4.7 (#2); "dropping CEA" (#10); bloodwork image (#9)	Chemo in #10 only	Not verifiable as written
PET metabolic response
21	Stage IV colon, liver+lung	Primary SUVmax 27.76 → 5.89 at 2 mo	mFOLFOXIRI	Marked response to triplet chemo
26	Appendiceal adenocarcinoma	SUVmax 8.4 → 2.06 (index lesion)	Chemo + immunotherapy	"68% volume reduction" came from Grok-3, not a radiologist
54	Stage III anal SCC	Node SUVmax 5.8 → 2.1; NED	CRT not stated	Best-documented NED, but standard-therapy status unknown
57	Colorectal	Reduced metabolic activity; "80% shrinkage"	Chemo	Interval and measurement method unreported

Take-away: biomarker declines occur in patients receiving effective chemotherapy and cannot isolate a drug effect. Several figures (#16, #26, #55, #84) were computed by AI tools or by assuming spherical tumors from two diameters.

4. Biomarkers that are missing (and why they matter)

Biomarker	Reported in	Why it changes interpretation
MMR / MSI	0 entries	dMMR/MSI-H tumors (roughly 4–5% of metastatic and 10–15% of stage II–III colon cancers) respond to checkpoint inhibitors (KEYNOTE-177, CheckMate 8HW; NICHE-2 in the neoadjuvant setting). Any "exceptional responder" with immunotherapy (#18, #26) could simply be dMMR.
RAS / BRAF / HER2 / KRAS G12C / NTRK	1 explicit (#6)	Determines anti-EGFR eligibility and targeted regimens (e.g., BRAF V600E combinations, HER2- and G12C-directed therapy).
Primary tumor sidedness	Partly (site)	Left- versus right-sided tumors differ in prognosis and anti-EGFR benefit.
HPV / p16 (squamous)	0 of 2	Strong prognostic marker in anal SCC.
Ki-67 / grade (NET)	0 of 4	Separates indolent NETs from aggressive NECs.
PIK3CA	0 entries	Relevant to aspirin evidence in resected disease (see section 8).
Baseline ctDNA	Partial (#22, #56)	Needed to read any ctDNA change.

5. Corrections to earlier versions of this article

• Case count: updated from 83 to 87 (Cases 84–87 added) and the source-page title reference corrected from "59 case reports (April 2026)" to "87 case reports (September 2026)".
• Metastatic-site claim withdrawn: the earlier statement that bone metastases responded worse is not supported. Both bone-metastatic entries (#15, #79) reported major responses, albeit with chemotherapy and in neuroendocrine disease. Peritoneal responses were mixed (#26 strong; #32, #42 modest). Liver-only disease is simply the most common pattern.
• "Antiparasitic alone" cluster: the earlier "about 15 cases" overstated it. Only 3 entries explicitly state no conventional therapy; the rest merely do not mention it.
• Complete-remission count: recounted at 18 entries claiming remission/NED; 12 document concurrent standard therapy, and only 6 describe objective documentation (#18, 22, 43, 54, 56, 58).
• Trial block: the 20 trial patients are no longer counted as "case reports"; Hegazy p-values are reported as p<0.001.
• Aydin et al. 2024: the patient received nivolumab, ipilimumab and lenvatinib with mebendazole; response cannot be credited to mebendazole.
• Removed: tracking parameters (utm_source) from citations and the Reddit compilation citation.
• PIK3CA/aspirin sentence reworded from a directive to an evidence-bounded statement.

6. Evidence hierarchy (CEBM)

Source	CEBM level	Key limitation
Hegazy et al., Life Sciences 2022: mebendazole 500 mg BID + bevacizumab + FOLFOX4 vs placebo, n=40 stage IV CRC	Level 2	Single centre, 12-week readout (ORR 65% vs 10%; median PFS 9.25 vs 3 months); not replicated; a chemo adjunct, not a replacement
Aydin et al., Frontiers in Oncology 2024	Level 4	One patient; four concurrent agents; MMR/MSI not reported on the compilation page
Physician-curated testimonials (54 entries)	Level 5	Unverified; selection and survivorship bias
Patient/family/social posts	Level 5	Self-interpreted imaging; no records
Preclinical (Wnt/β-catenin, microtubule, mitochondrial)	Level 5	Concentrations may exceed safe human exposure

Early-phase ivermectin trials exist for other cancers (e.g., NCT05318469 in triple-negative breast cancer; ICONIC, NCT07487805, in solid tumors), but none establishes efficacy in colorectal cancer.

7. Safety

• Dose escalation without safety data: the compilation quotes ivermectin up to about 2 mg/kg/day and fenbendazole plus mebendazole combined, far beyond approved antiparasitic dosing. Neurotoxicity, hepatotoxicity and interactions are documented concerns.
• Interactions: ivermectin and benzimidazoles use CYP3A4/P-glycoprotein pathways shared by many cancer drugs; discuss every agent with the treating team.
• Undisclosed use: two patients (#58, #81) deliberately hid use from their oncologists, which prevents monitoring and distorts attribution of response.
• Chlorine dioxide (#47): ingesting chlorine dioxide products is a bleach exposure; the FDA warns against it. Methylene blue (#29) and high-dose supplements also carry interaction risks.
• Veterinary products can vary in purity and concentration.

8. What to ask your oncology team

• Has my tumor had MMR/MSI, RAS/BRAF, HER2 and (if relevant) PIK3CA testing? dMMR/MSI-H disease has proven immunotherapy options; pMMR/MSS disease generally does not respond to checkpoint inhibitors alone.
• Is ctDNA monitoring appropriate after surgery?
• Low-dose aspirin showed benefit in resected, PIK3CA-altered stage I–III disease (ALASCCA, 2025); ask whether it fits your situation.
• If you are considering any repurposed drug, discuss it openly, ask about clinical trials, and have liver tests and interactions reviewed.

Related: dMMR/MSI-H immunotherapy vs. repurposed drugs · Colorectal cancer biomarkers guide · NEOPRISM-CRC

FAQ

How many unique patients are in the compilation? About 64 individuals plus a 20-patient trial arm, after removing duplicate updates.

Which subtypes appear? Mostly colon and rectal adenocarcinoma; 12 entries are neuroendocrine, appendiceal, squamous, signet-ring, mucinous or dual-primary tumors.

Which cases have biomarkers? About 20 unique patients, mostly CEA/CA 19-9; 1 genotyped, 2 with ctDNA, none with MMR/MSI.

Is any of this proven? No. One small randomized trial supports mebendazole as a chemotherapy adjunct.

AI personalization guide

• Claude: "Using this audit, explain which biomarkers (MMR/MSI, RAS/BRAF, ctDNA) my oncologist should check for my colorectal cancer stage."
• ChatGPT: "Summarize the Hegazy 2022 mebendazole trial and its limitations in plain language."
• Gemini: "List questions to bring to my oncologist about clinical trials for repurposed drugs."

References

1. OneDayMD: Colorectal Cancer Case Reports Compilation (87 cases, September 2026)
2. Hegazy et al. 2022, Life Sciences
3. Aydin et al. 2024, Frontiers in Oncology
4. Low-dose aspirin for PI3K-altered localized colorectal cancer (NEJM 2025)
5. NIKOLO niclosamide trial protocol (2018)
6. Niclosamide and colon cancer
7. 700+ case report compilation (all cancers)
8. 30 alternative and repurposed cancer treatments, evidence-ranked

Navigation: Ivermectin, Fenbendazole and Mebendazole for Cancer (All Cancer Types) > Review for Colorectal Cancer > Primary Colorectal Cancer Case Series.