My Healthcare News & Research — August 3, 2026 · High-Dose Vitamin D3 Added to Chemotherapy Did Not Slow Metastatic Colorectal Cancer in the SOLARIS Phase 3 Trial

On August 3, 2026, the Journal of the American Medical Association (JAMA) published online the full results of SOLARIS (Alliance A021703), a US trial that asked whether a high daily dose of vitamin D3, added to standard chemotherapy, could hold back colorectal cancer that had already spread to other organs. The trial enrolled 455 adults at 151 academic and community cancer centres across the United States. Every patient received the same kind of chemotherapy plus the drug bevacizumab; half also took 4,000 IU of vitamin D3 a day (after a two-week starting dose of 8,000 IU), and half took the standard 400 IU a day.

SOLARIS is a double-blind, randomised phase 3 trial: patients were assigned to a dose by chance, and neither they nor their doctors knew which dose they were taking. It was run to confirm an encouraging signal from a smaller phase 2 trial, SUNSHINE, published in 2019. It did not confirm it. The time until the cancer grew or the patient died (progression-free survival, the trial’s main measure) was a median of 11.8 months on high-dose vitamin D3 and 10.3 months on the standard dose, a difference that was not statistically significant (one-sided P = .25). Tumour shrinkage and overall survival were not different either.

The high dose did what it was meant to do in the blood: it lifted vitamin D levels quickly and kept them up, without extra side effects. It did not measurably change how the cancer behaved in the group as a whole. One prespecified subgroup, people whose cancer started on the left side of the colon, did better on the high dose; the authors call that finding exploratory. The headline result had first been presented at a cancer congress in September 2024; the August 2026 paper is the full, peer-reviewed report. Every number below is taken from that paper.

Table of Contents

  1. At a Glance
  2. What Was Done
  3. What Was Found, Number by Number
  4. Vitamin D Levels and Side Effects
  5. The Subgroup Signals (Exploratory)
  6. How Vitamin D Was Expected to Work
  7. How It Fits With Earlier Research
  8. What It Cannot Show, and What Remains Unknown
  9. What the Researchers and Others Said
  10. Sources and Primary Documents
  11. Key Research Papers
  12. Connections

At a Glance

Back to Table of Contents

What Was Done

Who could join. Adults aged 18 or older with colorectal adenocarcinoma (the common gland-cell type of bowel cancer) that was metastatic, or locally advanced and not removable by surgery, with tumours that could be measured on scans. They had to be fully active or limited only in strenuous activity (a performance score of 0 or 1 on the standard ECOG scale) and have normal organ function. Drug treatment for the advanced cancer must not have started; earlier chemotherapy around surgery was allowed if it had ended more than 12 months before the cancer came back.

Who could not. People with tumours known to be mismatch-repair deficient or “MSI-high” (a genetic subtype of the tumour), anyone with high blood calcium in the previous 6 months, anyone with symptomatic urinary stones in the past year, and anyone taking thiazide water pills, calcium supplements, or vitamin D at 2,000 IU a day or more in the past year.

The treatment everyone received. Every two weeks, each patient had either mFOLFOX6 (5-fluorouracil, leucovorin and oxaliplatin) or FOLFIRI (5-fluorouracil, leucovorin and irinotecan), the doctor’s choice, plus the antibody drug bevacizumab. Most patients (85.8% and 84.2% in the two groups) received mFOLFOX6.

The vitamin D comparison. Patients were randomised one-to-one. The high-dose group took two 4,000 IU capsules a day during the first two-week cycle (8,000 IU, a “loading” dose) and one 4,000 IU capsule a day after that. The standard-dose group took one 400 IU capsule plus one identical-looking dummy capsule during the first cycle, then one 400 IU capsule a day. Only the research pharmacist knew who was in which group. A supplement manufacturer supplied the capsules. Patients kept medication diaries and capsules were counted: average adherence was 96% in both groups.

What was measured. The main outcome was progression-free survival: the time from randomisation until scans showed the cancer growing (by the standard RECIST 1.1 rules) or the patient died. Secondary outcomes were objective response (the share of patients whose tumours shrank by a set amount), overall survival, blood vitamin D levels, and side effects. Blood was drawn for 25-hydroxyvitamin D, the form measured in the standard vitamin D test, before treatment, after cycles 4 and 8 (the first two restaging scans), and when treatment stopped. The samples were frozen and tested together in one batch at a certified laboratory whose staff did not know which group each sample came from.

How big it needed to be. The trial was planned to collect 273 progression events, enough for 90% power to detect a hazard ratio of 0.7 (planned on a median of 14.3 months on the high dose against 10 months on the standard dose). Enrolment was raised from 400 to 445 because more patients than expected dropped out. A planned halfway check in August 2022 did not trigger an early stop. Patients joined between October 29, 2019 and December 15, 2022, and the data were frozen on July 15, 2024, after a median follow-up of 20 months.

Back to Table of Contents

What Was Found, Number by Number

A hazard ratio compares how often an event (here, cancer growth or death) happens in one group against the other at any moment. A value of 1 means no difference; below 1 favours the high dose. The 95% confidence interval (CI) is the range of values the data are compatible with; when it spans 1, the difference between the groups is not statistically significant.

Who was in the trial. Of the 455 patients, the median age was 59 and 39.8% were women; 67.0% described themselves as White, 17.4% as Black or African American, 5.5% as Asian and 10.1% as another race. The cancer started in the left colon in 38.2% (high dose) and 39.8% (standard dose), in the right colon in 35.1% and 33.6%, and in the rectum in 21.5% and 20.4%. RAS gene changes were somewhat more common in the standard-dose group (44.9% vs 38.2%). The two groups received similar numbers of chemotherapy cycles.

Main outcome: progression-free survival.

Secondary outcomes.

In plain terms, the medians lean slightly towards the high dose for cancer growth and slightly the other way for survival, but the hazard-ratio intervals all cross 1 and none of the comparisons reached statistical significance. By the rules the trial set itself before it began, high-dose vitamin D3 made no measurable difference to how long the cancer was held back or how long patients lived.

Back to Table of Contents

Vitamin D Levels and Side Effects

Blood levels. 351 patients had a usable blood sample from before treatment. The trial used the cut-offs of a 2018 vitamin D guideline paper: below 20 ng/mL counted as deficient and 20–29 ng/mL as insufficient. The US Preventive Services Task Force has noted that “no one serum vitamin D level cutpoint defines deficiency,” so these labels are conventions, not fixed biological lines.

So the trial achieved a clear and lasting gap in vitamin D status between the two groups. The authors point out that this matters on its own: patients receiving chemotherapy have historically responded weakly to vitamin D supplements, and this regimen corrected insufficiency “quickly and durably.”

Side effects. 421 patients (212 high dose, 209 standard dose) took at least one dose and were counted for safety. Serious side effects were similar in both groups:

These safety figures apply to a closely monitored trial population that excluded people with recent high calcium, recent kidney stones, thiazide use or calcium supplements, and whose blood was checked repeatedly.

Back to Table of Contents

The Subgroup Signals (Exploratory)

Before the trial began, the investigators planned to look at whether the effect differed across groups of patients defined by known prognostic factors. They used a looser threshold for these “interaction” tests (P below .10) than for the main result. Two subgroups stood out:

Both patterns held after statistical adjustment for other patient factors. The authors write that the tumour-side finding is consistent with subgroup analyses from earlier clinical trials and from translational studies, but that “these findings remain exploratory and require further investigation.”

That caution has a reason. When a trial misses its main goal, splitting patients into many subgroups raises the chance that one comparison looks favourable by luck alone, and each subgroup is much smaller than the trial as a whole. In SOLARIS, the upper end of the left-sided confidence interval reaches 1.00, the line of no difference. A subgroup signal of this kind is a question for a future trial to test, not an answer.

Back to Table of Contents

How Vitamin D Was Expected to Work

Vitamin D made in the skin or swallowed in food and supplements is first converted in the body to 25-hydroxyvitamin D, the stable form that circulates in the blood and that the standard test measures. The SOLARIS authors describe it as “the best indicator of vitamin D status.” A second step turns it into the active hormone, 1,25-dihydroxyvitamin D, also called calcitriol. A 2014 review in Nature Reviews Cancer describes vitamin D as “not really a vitamin but the precursor to the potent steroid hormone calcitriol.” The animation How Your Skin Makes Vitamin D shows the first steps.

The SOLARIS paper sets out the laboratory reasoning behind the trial. Colorectal cancer cells carry the vitamin D receptor (VDR) and the enzyme 1-α-hydroxylase, which makes calcitriol from 25-hydroxyvitamin D. That means the active hormone can be produced inside the tumour and the tissue around it. When calcitriol binds the receptor, it switches target genes on and off. In laboratory studies this has been linked to:

One more piece of background explains the high dose. A 2012 study of 50 colorectal cancer patients given 2,000 IU a day for six months found that blood levels rose by at least 10 ng/mL at three months in 92% of patients not on chemotherapy, but in only 39% of those receiving it. The SOLARIS authors cite this study when they note that patients on chemotherapy have historically shown weaker responses to vitamin D supplements. In SOLARIS, the blood levels did rise. What it did not show was a change in the cancer.

Back to Table of Contents

How It Fits With Earlier Research

Observational studies: higher levels, longer survival. Several cohorts have linked higher blood vitamin D to better outcomes in colorectal cancer. In 304 patients from the Nurses’ Health Study and the Health Professionals Follow-Up Study, those in the highest quarter of pre-diagnosis 25-hydroxyvitamin D had a hazard ratio for death of 0.52 (95% CI 0.29–0.94) compared with the lowest quarter (2008). In 1,598 patients with stage I–III disease, the highest third of post-surgery levels had a hazard ratio of 0.68 (95% CI 0.50–0.90) for death from colorectal cancer (2014). In 1,041 patients with advanced or metastatic disease in the CALGB/SWOG 80405 chemotherapy trial, 63% were deficient and 31% insufficient at the start, and the top fifth of levels had a hazard ratio for death of 0.66 (95% CI 0.53–0.83) against the bottom fifth (2019). Studies like these show that two things go together; they cannot show that raising vitamin D causes the better outcome, because patients with higher levels can differ in many other ways.

SUNSHINE, the phase 2 trial (2019). The same research group randomised 139 patients at 11 US centres to high-dose or standard-dose vitamin D3 with mFOLFOX6 plus bevacizumab. Median progression-free survival was 13.0 months (95% CI 10.1–14.7) against 11.0 months (95% CI 9.5–14.0); the log-rank P value was .07, short of significance, but a supporting adjusted hazard ratio was 0.64 (one-sided 95% CI 0–0.90; P = .02). Tumour response (58% vs 63%) and overall survival (24.3 vs 24.3 months) did not differ. The SUNSHINE authors wrote that the findings “warrant further evaluation in a larger multicenter randomized clinical trial.” SOLARIS was that trial.

Why the two trials may have differed. The SOLARIS authors offer three possible explanations:

AMATERASU (2019). A double-blind trial at one university hospital in Japan gave 417 patients with stage I–III cancers of the digestive tract (48% colorectal) either 2,000 IU of vitamin D3 a day or a placebo after surgery. Five-year relapse-free survival was 77% vs 69%, a difference that was not significant (hazard ratio 0.76, 95% CI 0.50–1.14; P = .18); overall survival was 82% vs 81%. In a subgroup with starting levels of 20–40 ng/mL, relapse-free survival was 85% vs 71% (hazard ratio 0.46, 95% CI 0.24–0.86) — another subgroup finding, not the main result.

VITAL (2019). In 25,871 US adults (men aged 50 and over, women aged 55 and over) given 2,000 IU of vitamin D3 a day or placebo for a median of 5.3 years, vitamin D did not lower the number of new invasive cancers (hazard ratio 0.96, 95% CI 0.88–1.06). For colorectal cancer specifically the hazard ratio was 1.09 (95% CI 0.73–1.62). Deaths from cancer had a hazard ratio of 0.83 (95% CI 0.67–1.02), not statistically significant. VITAL asked about preventing cancer in the general population, a different question from SOLARIS.

Pooled trials (2019). A meta-analysis of randomised trials in Annals of Oncology found no effect of vitamin D supplements on how many people developed cancer (relative risk 0.98, 95% CI 0.93–1.03; 10 trials), but a lower risk of dying from cancer (relative risk 0.87, 95% CI 0.79–0.96; 5 trials), which the authors found “largely attributable to interventions with daily dosing.” That analysis asked about cancer across whole trial populations, not about treating people already living with metastatic disease.

Taken together: blood vitamin D tracks with colorectal cancer outcomes in observational studies, but randomised trials that raise vitamin D, in the general population (VITAL), after surgery (AMATERASU) and now in metastatic disease (SOLARIS), have not met their main goals. The site’s page on vitamin D3 and cancer prevention covers the wider research.

Back to Table of Contents

What It Cannot Show, and What Remains Unknown

Limits the authors state.

Limits that follow from the design.

Still unknown.

Back to Table of Contents

What the Researchers and Others Said

The authors. In the paper’s discussion, the investigators write that “the addition of high-dose vitamin D3 to standard chemotherapy cannot be recommended for unselected patients with previously untreated mCRC” (metastatic colorectal cancer). They also list what the trial did show: the regimen corrected low vitamin D in patients on chemotherapy, it was well tolerated alongside chemotherapy and bevacizumab, and patients took it reliably. Among the trial’s strengths they name its double-blind, placebo-controlled design, its size, its mix of academic and community centres and its diverse patient population. They describe it as an NCI-funded trial addressing a question “for which no other funding source exists.”

The first presentation (2024). The headline result was first presented at the European Society for Medical Oncology Congress in Barcelona in September 2024. The Dana-Farber Cancer Institute’s news release of September 15, 2024 (abstract LBA26) and the ecancer news service reported that high-dose vitamin D3 did not delay progression more than the standard dose, raised no new safety concerns, and showed a possible benefit in left-sided disease requiring further study. The ecancer report summarised the conclusion that high-dose vitamin D3 “cannot be recommended as a treatment for patients with untreated metastatic colon cancer.”

Coverage of the paper. Reports in trade outlets for oncology and pharmacy (August 6, August 7 and August 31, 2026) repeated the paper’s figures: 11.8 vs 10.3 months, hazard ratio 0.92, and the left-sided subgroup at 13.8 vs 10.2 months. The ASCO Post quoted the investigators’ description of that subgroup finding as exploratory. The figures in that coverage match the paper. PubMed lists no linked editorial for the paper.

Disclosures. The trial was funded by the National Cancer Institute. A supplement manufacturer supplied the vitamin D3 and dummy capsules. Three authors reported research funding to their institutions and/or consulting fees from drug, diagnostics or supplement companies, including the capsule supplier; the other authors reported none. The lead author is an associate editor of JAMA, as the paper’s author list states.

Since then. The paper appeared in the September 1, 2026 print issue of JAMA (volume 336, issue 9), and the authors’ manuscript became free to read in PubMed Central on September 5, 2026.

Back to Table of Contents

Sources and Primary Documents

Every trial figure above was taken from the full text of the SOLARIS paper (the NIH author manuscript in PubMed Central) and checked against the PubMed abstract. The DOI and PMID were confirmed at Crossref and NCBI.

  1. Ng K, Ou FS, Zemla T, Jackson NA, Kalyan A, Devoe C, et al. (2026). Addition of High-Dose Vitamin D3 to Standard Treatment in Patients With Metastatic Colorectal Cancer: The SOLARIS Randomized Clinical Trial (Alliance A021703). JAMA 336(9):746-754. — doi:10.1001/jama.2026.9350 · PubMed PMID: 42545685
  2. ClinicalTrials.gov. Vitamin D3 With Chemotherapy and Bevacizumab in Treating Patients With Advanced or Metastatic Colorectal Cancer (Alliance A021703). — NCT04094688
  3. Dana-Farber Cancer Institute news release (September 15, 2024). High-dose vitamin D3 does not provide benefit for metastatic colorectal cancer. — dana-farber.org
  4. ecancer (September 16, 2024). ESMO 2024: High-dose vitamin D3 does not provide benefit for metastatic colorectal cancer. — ecancer.org
  5. The ASCO Post (August 31, 2026). High-dose vitamin D3 does not improve PFS in metastatic colorectal cancer. — ascopost.com

Back to Table of Contents

Key Research Papers

Background papers on vitamin D and colorectal cancer, in the order they appear above. Each entry was checked against its PubMed record before it was written down.

  1. Feldman D, Krishnan AV, Swami S, Giovannucci E, Feldman BJ (2014). The role of vitamin D in reducing cancer risk and progression. Nature Reviews Cancer 14(5):342-357. — PubMed PMID: 24705652
  2. Fakih MG, Andrews C, McMahon J, Muindi JR (2012). A prospective clinical trial of cholecalciferol 2000 IU/day in colorectal cancer patients: evidence of a chemotherapy-response interaction. Anticancer Research 32(4):1333-1338. — PubMed PMID: 22493367
  3. Ng K, Meyerhardt JA, Wu K, Feskanich D, Hollis BW, Giovannucci EL, et al. (2008). Circulating 25-hydroxyvitamin D levels and survival in patients with colorectal cancer. Journal of Clinical Oncology 26(18):2984-2991. — PubMed PMID: 18565885
  4. Zgaga L, Theodoratou E, Farrington SM, Din FV, Ooi LY, Glodzik D, et al. (2014). Plasma vitamin D concentration influences survival outcome after a diagnosis of colorectal cancer. Journal of Clinical Oncology 32(23):2430-2439. — PubMed PMID: 25002714
  5. Yuan C, Sato K, Hollis BW, Zhang S, Niedzwiecki D, Ou FS, et al. (2019). Plasma 25-Hydroxyvitamin D Levels and Survival in Patients with Advanced or Metastatic Colorectal Cancer: Findings from CALGB/SWOG 80405 (Alliance). Clinical Cancer Research 25(24):7497-7505. — PubMed PMID: 31548349
  6. Ng K, Nimeiri HS, McCleary NJ, Abrams TA, Yurgelun MB, Cleary JM, et al. (2019). Effect of High-Dose vs Standard-Dose Vitamin D3 Supplementation on Progression-Free Survival Among Patients With Advanced or Metastatic Colorectal Cancer: The SUNSHINE Randomized Clinical Trial. JAMA 321(14):1370-1379. — PubMed PMID: 30964527
  7. Urashima M, Ohdaira H, Akutsu T, Okada S, Yoshida M, Kitajima M, et al. (2019). Effect of Vitamin D Supplementation on Relapse-Free Survival Among Patients With Digestive Tract Cancers: The AMATERASU Randomized Clinical Trial. JAMA 321(14):1361-1369. — PubMed PMID: 30964526
  8. Manson JE, Cook NR, Lee IM, Christen W, Bassuk SS, Mora S, et al. (2019). Vitamin D Supplements and Prevention of Cancer and Cardiovascular Disease. New England Journal of Medicine 380(1):33-44. — PubMed PMID: 30415629
  9. Keum N, Lee DH, Greenwood DC, Manson JE, Giovannucci E (2019). Vitamin D supplementation and total cancer incidence and mortality: a meta-analysis of randomized controlled trials. Annals of Oncology 30(5):733-743. — PubMed PMID: 30796437
  10. US Preventive Services Task Force; Krist AH, Davidson KW, Mangione CM, et al. (2021). Screening for Vitamin D Deficiency in Adults: US Preventive Services Task Force Recommendation Statement. JAMA 325(14):1436-1442. — PubMed PMID: 33847711

PubMed Topic Searches

  1. PubMed: Vitamin D3 supplementation in metastatic colorectal cancer trials
  2. PubMed: Blood vitamin D and colorectal cancer survival
  3. PubMed: Vitamin D receptor, calcitriol and colorectal cancer
  4. PubMed: Vitamin D and tumour side in colorectal cancer
  5. PubMed: Vitamin D supplements and cancer mortality, meta-analyses

Back to Table of Contents

Connections

Back to Table of Contents