Tomatoes: cooked is where the signal is, and the supplement failed

Pool 30 studies covering 260 thousand people and 24 thousand prostate cancers, and eating more tomato goes with 19 percent less of it. Split that by form and the picture comes apart in a useful way. Cooked tomatoes and sauces carried the association, at 0.84. Raw tomatoes came out at 0.96, with a confidence interval running through 1 and a p value of 0.487. Then six randomised trials gave people lycopene or tomato extract in a capsule and measured the prostate marker the whole idea was built on. It moved by 0.12 nanograms per millilitre, with a confidence interval from minus 0.62 to plus 0.38. That is the shape of the evidence on this page. What holds up sits on cooked food eaten by people who chose to eat it. The pill version of the same idea has been tested head on and did nothing.

The numbers, per 100 g

How much is actually in it

• A sauce made from one tomato variety carried its pigment almost entirely in a different molecular shape. — 12 healthy subjects, 6 men and 6 women (n = 12) Evidence B source
  A 150 g serving with 15 g corn oil gave 13 mg lycopene of which 97.0% was cis-isomers from tangerine tomatoes, against 17 mg beta-carotene and 4 mg lycopene from the high-beta-carotene variety; area under the curve for total lycopene after the tangerine sauce was 870 ± 187 nmol.h/L with over 99% as cis-isomers, and the dose-adjusted response was high compared with the literature

What cooking and storage change

• The link between tomatoes and lower prostate cancer risk held for cooked tomatoes and sauces and vanished for raw ones. — 30 studies covering 24 222 cases among 260 461 participants (n = 260 461) Evidence C source
  Higher total tomato consumption gave RR 0.81 (95% CI 0.71 to 0.92, p = 0.001); tomato foods 0.84 (0.72 to 0.98) and cooked tomatoes and sauces 0.84 (0.73 to 0.98); raw tomatoes 0.96 (0.84 to 1.09, p = 0.487); dose-response was significant for total tomato, for cooked tomatoes and sauces, and for raw tomatoes
• Fifty five grams of tomato paste a day for twelve weeks measurably raised the skin's tolerance of ultraviolet light. — 20 healthy women, median age 33, skin phototype I or II, over 12 weeks, with no test of sunburn or sun protection (n = 20) Evidence B source
  Erythemal D30 rose to 36.6 ± 14.7 mJ/cm² after tomato paste against 23 ± 6.6 for control (P = 0.03) while minimal erythema dose did not differ significantly between groups; ultraviolet-induced matrix metalloproteinase-1 was lower after tomato paste (P = 0.04), procollagen I deposition rose (P = 0.05) and the mitochondrial DNA 3895-bp deletion was reduced (P = 0.01)

What it does to your health

• Given as a supplement in trials, lycopene moved the prostate marker by nothing at all. — 6 randomised trials pooled from 1036 records identified (n = 6 trials) Evidence A source
  Weighted mean difference in PSA against control was -0.12 ng/ml (95% CI -0.62 to 0.38, P = 0.64)

Safety, limits and interactions

• Pooled worldwide sampling found Salmonella on no tomatoes at all, and on one ready-to-eat salad in seven. — published prevalence surveys across eight food categories, five of which had enough data to pool (n = 5 pooled categories) Evidence B source
  Prevalence was 0.1% for fruits, 0.2% for leafy vegetables, 13.7% for mixed vegetables in ready-to-eat salads, 0.1% for tubercles and 0% for tomatoes; for fruits, tubercles and tomatoes neither preparation type nor sampling place changed prevalence significantly, while unprocessed leafy vegetables carried more than ready-to-eat ones
• Tomato products are one of the two main dietary routes for a mould toxin group that has no safety limit anywhere. — 29 dietary exposure assessment studies from Asia, Europe, Africa and the Americas (n = 29 exposure assessments) Evidence B source
  Cereal and tomato products are the primary sources of Alternaria mycotoxin intake for adults and children; roughly 70% of exposure assessments concluded one or more of these toxins poses a food safety risk; tenuazonic acid and alternariol are the most detected; most of these toxins are heat stable and are not easily removed by processing

Who this works differently for

• The one study in the corpus that counted nightshade eating in a skin disease found patients ate more of them, and found the same for junk food. — 50 patients with hidradenitis suppurativa and 50 controls matched on body mass index, sex and age (n = 100) Evidence C source
  Patients showed lower Mediterranean diet adherence, higher dietary glycaemic index, lower micronutrient intake, and more frequent consumption of junk food, dairy and nightshade vegetables; in multiple regression the Mediterranean diet score was the main variable associated with severity

What is still unknown

• Europe asked its members to start measuring these toxins in tomato products and stopped short of setting a limit. — EU Member States and food business operators Evidence E source
  Member States are asked to monitor alternariol, alternariol monomethyl ether and tenuazonic acid in food, with processed tomato products named first among eight food groups; the annex gives indicative levels of 10, 5 and 500 µg/kg respectively for processed tomato products

The bottom line

Twenty women ate 55 grams of tomato paste a day in olive oil for twelve weeks, and their skin took measurably more ultraviolet before it reddened. That is a genuine randomised result, and it is twenty women with the palest two skin types, taking most of a small tin of paste every day for three months. Nothing in it says you can use less sunscreen, and the trial never tested that. The thing we would rather you knew about tomatoes is duller and better established. Processed tomato products are one of the two main dietary routes for a family of mould toxins called Alternaria toxins, alongside cereals. Those toxins are mostly heat stable and processing does not easily remove them, and around 70 percent of the exposure assessments in that review concluded one or more of them poses a food safety risk. Europe has asked its member states to start measuring three of them in processed tomato products, with indicative levels of 10, 5 and 500 micrograms per kilogram, and has stopped short of setting a legal limit. There is nothing you can do about that in a shop, and you should still know it exists. On the contamination people actually worry about, pooled worldwide sampling found Salmonella on zero percent of tomatoes and on 13.7 percent of mixed vegetables in ready-to-eat salads. The tomato is not the risky item in that bowl. And if you have heard that nightshades make skin disease worse, here is the one study in our corpus that counted. Patients with hidradenitis suppurativa ate nightshades more often, and they also ate more junk food and more dairy. The strongest thing in the model was how far their diet sat from a Mediterranean one. One line is missing here and we would rather name it than let you find the hole yourself. We hold no evidence card about people who should go carefully with tomatoes, so everything above is about the food and nothing is about a particular reader. That silence is our gap and not a clean bill.

Sources

1. tos-r5-07 · randomised crossover trial
   The serving size was 150 g (containing 15 g of corn oil), tangerine sauce containing 13 mg of lycopene (97.0% as cis-isomers) and high-beta-carotene sauce containing 17 mg of total beta-carotene (1.6% as the 9-cis-isomer) and 4 mg of lycopene.
   J Agric Food Chem, 2007 · checked 2026-09-22
2. tos-r5-01 · systematic review and dose-response meta-analysis of observational studies
   However, no associations were found for raw tomatoes (RR = 0.96, 95% CI: 0.84-1.09, p = 0.487).
   Prostate Cancer Prostatic Dis, 2018 · checked 2026-09-22
3. tos-r5-05 · randomised controlled trial
   Mean ± SD erythemal D(30) was significantly higher following tomato paste vs. control (baseline, 26·5 ± 7·5 mJ cm(-2); control, 23 ± 6·6 mJ cm(-2); tomato paste, 36·6 ± 14·7 mJ cm(-2); P = 0·03), while the MED was not significantly different between groups
   Br J Dermatol, 2011 · checked 2026-09-22
4. tos-r5-02 · systematic review and meta-analysis of randomised controlled trials
   The pooled analysis of the 6 studies showed no significant differences in PSA levels in subjects treated with lycopene or tomato extract containing lycopene (WMD= -0.12 ng/ml; 95% CI: -0.62, 0.38 ng/ml; P = 0.64) compared to the control.
   Complement Ther Med, 2022 · checked 2026-09-22
5. tos-r5-03 · systematic review and meta-analysis of prevalence studies
   Results showed prevalence for Salmonella spp. of 0.1%, 0.2%, 13.7%, 0.1%, and 0% for fruits, leafy vegetables, mixed vegetables related to ready-to-eat salads (RTE), tubercles, and tomatoes, respectively.
   World J Microbiol Biotechnol, 2021 · checked 2026-09-22
6. tos-r5-04 · systematic literature review
   Cereal and tomato products are the primary sources of Alternaria mycotoxin intake for both adults and children.
   Compr Rev Food Sci Food Saf, 2025 · checked 2026-09-22
7. tos-r5-08 · cross-sectional case-control study
   Their micronutrient intake was also reduced, while the consumption of junk food, dairy products, and nightshade vegetables was more frequent.
   Medicina (Kaunas), 2024 · checked 2026-09-22
8. tos-r5-06 · EU Commission Recommendation
   Member States, in close cooperation with the food business operators should monitor the Alternaria toxins alternariol, alternariol monomethyl ether and tenuazonic acid in food, in particular in processed tomato products, paprika powder, sesame seeds, sunflower seeds, sunflower oil, tree nuts, dried figs and cereal-based foods for infants and young children.
   Commission Recommendation (EU) 2022/553 on monitoring the presence of Alternaria toxins in food · checked 2026-09-22