Effect of Intake of Lycopene-Containing Products on Vascular Endothelial Function in Healthy Adults: A Systematic Review with Meta-Analysis

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The meta analysis consider an interesting compound Lycopene and a very hot topic the cardiovascular and endothelial health

The paper fit the journal's aims and scopes

Paper is carefully prepared and well organised not to long and clearly stating objectives and methods, 

The introduction offer sufficient information to understand the topic and the paper

Methods in my opinion are clearly described

A lot of information are available as supplementary fiiles

Results are summarised in sintetic way and relevantinformations are reported

I appreciated also the statement in the first paragraph of the discussion "This study overcomes key limitations of previous meta-analysis by incorporating 289 more recent evidence and encompassing a broader range of intervention designs, outcome 290 measures, intake amounts, and durations. Our meta-analysis demonstrated a significant 291 improvement in FMD, indicating that continuous lycopene intake enhances vascular en-292 dothelial function in healthy adults. Collectively, these findings provide up-to-date evi-293 dence supporting the beneficial role of lycopene in vascular endothelial function." that clearly describe the novelty of th epresent paper compared to previous and offer reader some new informations

 

I have some comments

-is it possible to obtain some further informations related to the type of intervention, i see in the table 1 way of supplementing lycopene are very different ranging from juice (watermelon) to tomato paste, thus the compound is from one side in a water based media in another in a lipidic based or high lipid food. 

Is there any comment about this can differences be observed?

A final comment is if in the different considered study there is clear indication of how lycopene was measured and if the different values reported as doses can be really be comparable.

 

Author Response

【Comment 1】

-is it possible to obtain some further informations related to the type of intervention, i see in the table 1 way of supplementing lycopene are very different ranging from juice (watermelon) to tomato paste, thus the compound is from one side in a water based media in another in a lipidic based or high lipid food. 

Is there any comment about this can differences be observed?

【Response 1】

Types of intervention may influence the result, but unfortunately, sensitivity analyses considering the form and type of the intervention and other factors could not be conducted because of the limited number of included studies. Based on this, we added the following description in the Discussion section as a limitation.

“First, the number of interventional studies included in the meta-analysis was limited. We conducted leave-one-out analysis as a sensitivity analysis, but sensitivity analyses considering factors such as study design, the form and type of the intervention, lycopene intake amounts or duration, and ethnicity could not be conducted because of the limited number of included studies. Insufficient information from sensitivity analysis makes it difficult to discuss the dependence of results on a specific study design, the effect of ingredients other than lycopene, the relationship between the duration and amount of lycopene intake and its effect, ethnic differences in the effects of lycopene. Regarding the effect of ingredients other than lycopene, a recent meta-analysis showed that intake of L-citrulline, which is a major ingredient in watermelon, significantly improved FMD [51]. A broader evidence base is needed, accounting for the source of lycopene intake, the duration/amount of lycopene intake, and regional differences. As additional RCTs accumulate, it will become possible to perform sensitivity analyses and clarify the effects of lycopene on vascular endothelial function.” (L.424 – L. 437)

We added the following reference in the reference section.

51. Luo, P.; Li, Z.; Liu, K.; Gao, W. Effects of L-citrulline supplementation and watermelon intake on arterial stiffness and endothelial function in middle-aged and older adults: a systematic review and meta-analysis of randomized controlled trials. Front Nutr 2025, 12, 1632952. DOI: 10.3389/fnut.2025.1632952. (L. 635 – L. 637)

 

【Comment 2】

A final comment is if in the different considered study there is clear indication of how lycopene was measured and if the different values reported as doses can be really be comparable.

【Response 2】

Lycopene contents in the intervention were determined by HPLC [35,38], nutrient analysis by Food Processor SQL Edition [34], standard nutritional values ​​for the product [39]. It is unclear in two included studies [36,37]. Due to differences in measurement methods, it may be difficult to simply compare lycopene content across studies. Therefore, the statement related to the dose-duration conclusion in the Discussion section was corrected as follows.

“Furthermore, one study not included in the meta-analysis showed a significant in-crease in RH-PAT with 15 mg lycopene intake for 8 weeks [39]. Based on the available evidence, it can be hypothesized that the beneficial effect of lycopene intake on vascu-lar endothelial function may require a daily intake of at least 15 mg for 12 weeks, although this hypothesis warrants further investigation, and the proposed dose and du-ration thresholds may change as further evidence accumulates.” (L. 370 – L. 375)

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript presents a systematic review and meta-analysis on the effect of oral intake of lycopene-containing products on vascular endothelial function in healthy adults, focusing mainly on flow-mediated dilation (FMD). The topic is relevant to Nutrients and the manuscript is generally well organized. The authors provide a registered protocol, a PRISMA checklist, detailed search strategies, study characteristics, risk-of-bias assessment, and supplementary material. The manuscript is clearly written overall.

However, the current version substantially overstates the strength, novelty, and specificity of the evidence. The review is based on only six included studies, and the quantitative FMD meta-analysis includes only four studies with 182 participants. The statistically significant pooled effect is strongly driven by one recent study authored by the present Authors and funded/sponsored by the same company/research center employing all Authors. In addition, most FMD studies used tomato- or watermelon-based products rather than isolated lycopene, making it difficult to attribute the observed effects specifically to lycopene. The certainty rating of “high” is not convincing under standard GRADE reasoning, particularly given the very small evidence base, the dominance of one study, several moderate-risk-of-bias studies, surrogate endpoints, lack of robust sensitivity analyses, and the inappropriate or underpowered assessment of publication bias with only four studies.

In my view, the manuscript requires major revision before it can be considered further. The article may be publishable only if the Authors substantially tone down the conclusions, revise the certainty assessment, perform additional sensitivity analyses, and present the review as limited evidence on lycopene-containing foods/products rather than strong evidence for lycopene as an isolated active compound.

Major comments

1. Novelty and added value are currently overstated.
   The manuscript presents itself as overcoming key limitations of the previous systematic review/meta-analysis. However, the actual added evidence appears limited. Only a small number of additional studies are included, and the main FMD meta-analysis remains based on four studies. The added value should be described more cautiously. The manuscript should explicitly state how many studies and participants are newly added to the previous evidence base and whether the overall conclusion materially changes without the Authors’ own recent study.
2. The conclusion that “lycopene intake improves vascular endothelial function” is stronger than the evidence permits.
   Most included FMD studies used complex food matrices such as tomato paste, tomato puree, tomato juice, or watermelon juice. These products contain other potentially bioactive compounds, including polyphenols, carotenoids other than lycopene, vitamin C, potassium, nitrate, fiber-derived compounds, and matrix-dependent effects. Therefore, the evidence does not isolate lycopene as the causal factor. The title, abstract, conclusions, and discussion should be reframed around “lycopene-containing foods/products” unless evidence from purified lycopene interventions is analyzed separately. The RH-PAT capsule study cannot compensate for this limitation because it is a single study and is not included in the FMD meta-analysis.
3. The evidence certainty rating of “high” is not adequately justified.
   The GRADE-like assessment is too permissive. A total sample size of 182 across four studies should not automatically be considered “precise”, especially when the confidence interval is close to the null and the evidence base is very small. Publication bias cannot be reliably excluded with four studies; funnel plots and Egger’s test are not informative with such a small number of studies. The authors should not state that publication bias is low based on these analyses. The certainty should be downgraded for risk of bias, imprecision, and likely publication/small-study uncertainty. A rating of “high” is not credible in the present form.
4. The pooled result appears highly dependent on one industry-funded, author-led study.
   The Authors are employees of Kagome Co., Ltd., Kagome sponsored the review, and the Authors are also authors of one included study. This is transparently declared, but the methodological implications are not adequately addressed. The Authors’ own trial contributes a very large weight to the meta-analysis. A mandatory sensitivity analysis excluding the authors’ own study should be added and discussed prominently. If the significance is lost or markedly weakened, the abstract and conclusions must be revised accordingly.
5. Sensitivity and influence analyses are essential.
   With only four FMD studies, each individual study can substantially influence the pooled effect. The authors should perform and report leave-one-out analyses, analyses excluding the authors’ own trial, analyses excluding non-placebo/no-intervention controls, and analyses separating food-matrix interventions from purified lycopene supplementation. If these analyses are not statistically meaningful due to the small number of studies, that limitation should be explicitly stated and the conclusions reduced accordingly.
6. The handling of crossover studies requires clarification and possibly reanalysis.
   Four included studies used randomized crossover designs. The meta-analysis appears to treat intervention and control periods as if they were independent groups. This may not appropriately account for within-participant correlation, period effects, carryover, and washout. The authors should clarify exactly how crossover data were handled and whether paired analyses were used when possible. If paired data were unavailable, this should be acknowledged as a limitation and the potential impact on precision should be discussed.
7. Endpoint values versus change scores are not consistently handled or explained.
   The Methods state that endpoint values were used, while Table 1 reports mean differences in increases from baseline, and Supplementary File 6 provides a meta-analysis of MDs. The authors should clearly distinguish endpoint-based analyses from change-score analyses, justify the primary choice, and explain how baseline imbalance was addressed. In interventions assessing FMD, change from baseline or paired treatment differences may often be more appropriate than endpoint values alone, especially in small crossover studies.
8. The risk-of-bias tool is non-standard and insufficiently transparent.
   The authors use a modified 13-item checklist and sum “risk” or “unclear” ratings into total scores. This approach is less informative than domain-based tools such as RoB 2 for randomized trials. The authors should either use a standard risk-of-bias tool or provide stronger justification for the modified scoring approach. Each judgment should be supported with study-specific explanations. In particular, blinding, allocation concealment, compliance, selective reporting, and outcome assessor blinding need clearer documentation.
9. Publication bias assessment is not valid with four studies.
   Funnel plots and Egger’s test should not be used to infer absence of publication bias in such a small meta-analysis. The text should be revised to state that publication bias could not be meaningfully assessed. The authors should avoid concluding that publication bias is low.
10. The search strategy is useful but should be strengthened or better justified.
    The review searches several databases and registries, including Japanese sources, which is positive. However, the absence of Embase, Web of Science, and Scopus should be justified. The search strategy also focuses heavily on lycopene/tomato/watermelon and may not fully capture all lycopene-rich sources mentioned in the Introduction, such as pink grapefruit or papaya. The authors should verify whether the search strategy is sufficiently comprehensive for the stated scope. The search date should also be updated close to publication, given that the manuscript cites sources accessed in 2026 while the literature search was conducted in September 2025.
11. The list of excluded studies should provide more specific reasons.
    Supplementary File 5 often states only “PICOS did not meet the eligibility criteria”. This is insufficiently transparent. For each excluded full-text study, the exact reason should be specified, for example wrong population, wrong intervention, wrong comparator, wrong outcome, uncontrolled design, acute-only design, or inadequate control.
12. Health-claim and prevention language should be moderated.
    The manuscript refers to cardiovascular disease prevention and suggests functional benefits of lycopene. However, the review evaluates surrogate endothelial-function outcomes, not cardiovascular events, disease prevention, or long-term clinical outcomes. Statements implying cardiovascular prevention or substantiated functional benefit should be toned down.
13. Dose-duration conclusions are too speculative.
    The statement that beneficial effects are achieved by taking at least 15 mg lycopene for 12 weeks is based on very limited evidence and appears strongly dependent on one study. Dose-response conclusions should be presented as hypotheses, not as evidence-based thresholds.
14. Conflict-of-interest management should be strengthened.
    Disclosure is present, but disclosure alone is not sufficient here. The authors should state whether data extraction, risk-of-bias assessment, and certainty assessment for their own included study were independently checked by someone external to Kagome. If not, this limitation should be explicitly acknowledged.

Minor comments

• The abstract should report that the meta-analysis includes only four FMD studies and 182 participants, and that conclusions are limited by the small evidence base.
• The phrase “with some scientific support” in the abstract is vague and should be replaced by a clearer statement such as “limited evidence suggests…” or “low-to-moderate certainty evidence suggests…”, depending on the revised certainty assessment.
• The statement that the certainty of evidence is high should be removed unless the certainty assessment is substantially revised and convincingly justified.
• The discussion comparing the effect size of lycopene with potassium, CoQ10, and resveratrol should be toned down because the evidence bases differ substantially in population, design, number of studies, and risk of bias.
• The safety discussion is not central to the review question and should be shortened or clearly separated from efficacy. This review does not establish safety across doses, long-term intake, or vulnerable populations.
• Line/reference issue: the statement “Participants ingested test foods containing 6–46.2 mg of lycopene…” appears to cite references [30–33,35], but references 30–31 are methodological references, not intervention studies. Please check and correct this reference range.
• The manuscript title and conclusions should distinguish “lycopene-containing foods/products” from “lycopene” as an isolated compound.
• The PRISMA checklist reports some items as “N.a.”, including sensitivity analyses. Given the small dataset and potential dominance of one study, sensitivity analyses are not optional and should be performed.

Comments on the Quality of English Language

Please correct minor wording issues, for example “the previous meta-analysis” rather than “previous meta-analysis”, and “are employed by Kagome Co., Ltd.” rather than “are employed Kagome Co., Ltd.”

Author Response

Major comments

[Comment 1]

Novelty and added value are currently overstated.

The manuscript presents itself as overcoming key limitations of the previous systematic review/meta-analysis. However, the actual added evidence appears limited. Only a small number of additional studies are included, and the main FMD meta-analysis remains based on four studies. The added value should be described more cautiously. The manuscript should explicitly state how many studies and participants are newly added to the previous evidence base and whether the overall conclusion materially changes without the Authors’ own recent study.

[Response 1]

We added information on the number of newly added studies and participants.

“Compared with the studies included in the previous meta-analysis, one study [40] was excluded owing to an unsuitable control, while two newly published studies [37,38] were included. Additionally, the total sample size of the meta-analysis increased from 107 to 182 participants.” (L. 277 – L. 280)

 

Also, we conducted a leave-one-out analysis to evaluate the influence of individual studies on the meta-analysis results, and accordingly, added or changed the following descriptions in the Methods, Results, Discussion, and Conclusions sections.

“A leave-one-out analysis, a sensitivity analysis that repeatedly recalculates the overall pooled effect by systematically removing one study at a time, was conducted to evaluate the robustness of the results.” (L. 204 – L. 207)

“Leave-one-out analysis showed that the meta-analysis results lost significance when Yoshida et al. [38] was excluded. Significance was maintained when any of the other three studies [35–37] were excluded (Supplementary File 7).” (L. 284 – L. 287)

“A leave-one-out analysis showed that exclusion of Yoshida et al. [38] resulted in a non-significant effect of the lycopene-containing food intervention on FMD in the me-ta-analysis, whereas exclusion of any of the other three studies [35–37] did not alter the statistical significance of the pooled effect. These findings suggest that the results of Yoshida et al. exert a substantial influence on the overall conclusions of the present SR, and therefore the findings should be interpreted considering this dependency. However, Yoshida et al. was the only study assessed as having a low risk of bias and enrolled the largest number of participants among the included studies. Consequently, the meta-analysis performed after excluding this study is based solely on studies with greater methodological limitations, which may substantially reduce the quality of the evidence and potentially lead to misleading interpretations. Although the considerable influence of Yoshida et al. on the pooled results should be acknowledged, further accumulation of high-quality RCTs with a low risk of bias is required to more accurately evaluate the contribution and influence of individual studies on the overall evidence base.” (L. 337 – L. 350)

“It suggested the beneficial effects of continuous intake of lycopene-containing products intake on vascular endothelial function in healthy adults, although the findings came from a small-scale meta-analysis in which the results of a single study exerted a substantial influence on the overall conclusions.” (L. 464 – L. 467)

We added the following Supplementary File. (Supplementary File 7, L. 475 – L. 476)

“Supplementary File 7: Meta-analysis of the effects of continuous lycopene intake on FMD (Leave-one-out analysis)”

 

Based on the description above, we corrected the conclusion of Abstract section as follows.

“This systematic review/meta-analysis with moderate certainty evidence suggests continuous intake of lycopene-containing products may have beneficial effects on vascular endothelial function in healthy adults, although the findings come from a small-scale meta-analysis in which the results of a single study exert a substantial influence on the overall conclusions.” (L. 28 – L. 31)

 

[Comment 2]

The conclusion that “lycopene intake improves vascular endothelial function” is stronger than the evidence permits.

Most included FMD studies used complex food matrices such as tomato paste, tomato puree, tomato juice, or watermelon juice. These products contain other potentially bioactive compounds, including polyphenols, carotenoids other than lycopene, vitamin C, potassium, nitrate, fiber-derived compounds, and matrix-dependent effects. Therefore, the evidence does not isolate lycopene as the causal factor. The title, abstract, conclusions, and discussion should be reframed around “lycopene-containing foods/products” unless evidence from purified lycopene interventions is analyzed separately. The RH-PAT capsule study cannot compensate for this limitation because it is a single study and is not included in the FMD meta-analysis.

[Response 2]

Regarding expressions that appear to require correction, we changed the wording from "lycopene" to "lycopene-containing products" in the Title, Abstract, Discussion, and Conclusions sections.

 

[Comment 3]

The evidence certainty rating of “high” is not adequately justified.

The GRADE-like assessment is too permissive. A total sample size of 182 across four studies should not automatically be considered “precise”, especially when the confidence interval is close to the null and the evidence base is very small. Publication bias cannot be reliably excluded with four studies; funnel plots and Egger’s test are not informative with such a small number of studies. The authors should not state that publication bias is low based on these analyses. The certainty should be downgraded for risk of bias, imprecision, and likely publication/small-study uncertainty. A rating of “high” is not credible in the present form.

[Response 3]

Regarding the assessment of risk of bias, as the currently presented method follows the prospectively registered protocol and cannot be omitted, a post-hoc assessment using the Cochrane Risk of Bias version 2.0 (RoB2 tool) was conducted. The assessment revealed similar results for both methods. The following details were added in the Methods and Results sections.

“In addition, we assessed the risk of bias using the Cochrane Risk of Bias version 2.0 (RoB2 tool [29]) as a post-hoc assessment.” (L. 187 – L. 188)

“The results of the risk of bias assessed using the RoB2 tool are shown in Supplementary File 6. One study [38] was assessed as having the highest rating on a three-point scale, while the others [34–37,39] were assessed as having the second-highest rating; these results were consistent with those shown in Table 2.” (L. 253 – L. 257)

We added the following reference in the Reference section.

29. Sterne, J.A.C.; Savović, J.; Page, M.J.; Elbers, R.G.; Blencowe, N.S.; Boutron, I.; Cates, C.J.; Cheng, H.Y.; Corbett, M.S.; Eldridge, S.M.; et al. RoB 2: A revised tool for assessing risk of bias in randomised trials. BMJ 2019, 366, l4898. DOI:10.1136/bmj.l4898. (L. 574 – 576)

We added the following Supplementary File. (Supplementary File 6, L. 474 – L. 475)

“Supplementary File 6: The results of the risk of bias evaluation using RoB2 tool”

 

Regarding the assessment of the risk of bias, we concluded that publication bias was not evaluated appropriately because of the low number of included studies. Therefore, we corrected the Results section as follows.

“Although no obvious asymmetry was observed in the funnel plots (Supplementary File 9) and there were no unpublished studies, publication bias was not evaluated appropriately because of the low number of included studies.” (L.306 – L. 308)

 

Regarding the assessment of certainty of evidence across the studies, as the currently presented method follows the prospectively registered protocol and cannot be omitted, a post-hoc assessment using the GRADEpro was conducted. As a result, the certainty of the evidence across the studies was comprehensively assessed as “moderate”. The following details were added in the Methods and Results sections.

“Additionally, as a post-hoc supplementary analysis that was not prespecified in the prospectively registered protocol, we assessed the certainty of evidence across the studies using the GRADEpro GDT tool [33] based on the criteria described in the Cochrane Handbook [28].” (L. 218 – L. 221)

“The certainty of the evidence assessed by the registered protocol was "high" and that by GRADEpro (Supplementary File 10) was “moderate” with the limitation that publication bias cannot be assessed. Considering the insufficient assessment of publication bias, the certainty of the evidence across the study was comprehensively assessed as “moderate”.” (L. 308– L. 312)

We added the following Supplementary File. (Supplementary File 10, L. 478 – L. 479)

“Supplementary File 10: The results of the certainty of the evidence assessed using GRADEpro.”

We added the following reference in the Reference section.

33. GRADEpro GDT [Computer program]; Hamilton: ON: McMaster University (developed by Evidence Prime). Available online: https://www.gradepro.org (accessed on 2 July, 2026). (L. 584 – L. 585)

 

[Comment 4]

The pooled result appears highly dependent on one industry-funded, author-led study.

The Authors are employees of Kagome Co., Ltd., Kagome sponsored the review, and the Authors are also authors of one included study. This is transparently declared, but the methodological implications are not adequately addressed. The Authors’ own trial contributes a very large weight to the meta-analysis. A mandatory sensitivity analysis excluding the authors’ own study should be added and discussed prominently. If the significance is lost or markedly weakened, the abstract and conclusions must be revised accordingly.

[Response 4]

As written in the response to Comment 1, we conducted a leave-one-out analysis to evaluate the influence of individual studies on the meta-analysis results, and accordingly added or changed the descriptions in the Abstract, Methods, Results, Discussion, and Conclusions sections.

 

[Comment 5]

Sensitivity and influence analyses are essential.

With only four FMD studies, each individual study can substantially influence the pooled effect. The authors should perform and report leave-one-out analyses, analyses excluding the authors’ own trial, analyses excluding non-placebo/no-intervention controls, and analyses separating food-matrix interventions from purified lycopene supplementation. If these analyses are not statistically meaningful due to the small number of studies, that limitation should be explicitly stated and the conclusions reduced accordingly.

[Response 5]

As written in the response to Comment 1, we conducted a leave-one-out analysis to evaluate the influence of individual studies on the meta-analysis results. Other sensitivity analyses considering factors such as study design, the form and type of the intervention, lycopene intake amounts or duration, and ethnicity could not be conducted because of the limited number of included studies. Based on this, we added the following description in the Discussion section as a limitation.

“First, the number of interventional studies included in the meta-analysis was limited. We conducted leave-one-out analysis as a sensitivity analysis, but sensitivity analyses considering factors such as study design, the form and type of the intervention, lycopene intake amounts or duration, and ethnicity could not be conducted because of the limited number of included studies. Insufficient information from sensitivity analysis makes it difficult to discuss the dependence of results on a specific study design, the effect of ingredients other than lycopene, the relationship between the duration and amount of lycopene intake and its effect, ethnic differences in the effects of lycopene. Regarding the effect of ingredients other than lycopene, a recent meta-analysis showed that intake of L-citrulline, which is a major ingredient in watermelon, significantly improved FMD [51]. A broader evidence base is needed, accounting for the source of lycopene intake, the duration/amount of lycopene intake, and regional differences. As additional RCTs accumulate, it will become possible to perform sensitivity analyses and clarify the effects of lycopene on vascular endothelial function.” (L. 424 – L. 437)

We added the following reference in the reference section.

51. Luo, P.; Li, Z.; Liu, K.; Gao, W. Effects of L-citrulline supplementation and watermelon intake on arterial stiffness and endothelial function in middle-aged and older adults: a systematic review and meta-analysis of randomized controlled trials. Front Nutr 2025, 12, 1632952. DOI: 10.3389/fnut.2025.1632952. (L. 635 – L. 637)

 

[Comment 6]

The handling of crossover studies requires clarification and possibly reanalysis.

Four included studies used randomized crossover designs. The meta-analysis appears to treat intervention and control periods as if they were independent groups. This may not appropriately account for within-participant correlation, period effects, carryover, and washout. The authors should clarify exactly how crossover data were handled and whether paired analyses were used when possible. If paired data were unavailable, this should be acknowledged as a limitation and the potential impact on precision should be discussed.

[Response 6]

Paired analyses should be used in this case, but we could not obtain the paired data. We added the following description in the Discussion section as the limitation.

“Third, crossover design studies were pooled in the meta-analysis using the same method as for parallel-group studies. This may not appropriately account for with-in-participant correlation, period effects, carryover, and washout. Paired analyses should be used in this case, but we could not obtain the paired data. Although leave-one-out analysis suggested that excluding any of the three crossover studies in-dividually does not substantially affect the meta-analysis results, the possibility that the lack of paired data is influencing the results cannot be ruled out.” (L. 447 – L. 454)

 

[Comment 7]

Endpoint values versus change scores are not consistently handled or explained.

The Methods state that endpoint values were used, while Table 1 reports mean differences in increases from baseline, and Supplementary File 6 provides a meta-analysis of MDs. The authors should clearly distinguish endpoint-based analyses from change-score analyses, justify the primary choice, and explain how baseline imbalance was addressed. In interventions assessing FMD, change from baseline or paired treatment differences may often be more appropriate than endpoint values alone, especially in small crossover studies.

[Response 7]

Endpoint values were mainly pooled because information on mean difference values and corresponding SD for FMD was not reported in the published articles for three studies. A meta-analysis of MDs was presented as a supplementary file because it was conducted using estimated SDs calculated by the formula. We added the following description in the Results section.

“Endpoint values were mainly pooled because information on mean difference values and corresponding SD for FMD were not reported in the published articles for three studies [35–37].” (L. 280 – L. 282)

“A meta-analysis of MDs conducted using estimated SDs calculated by the formula is shown in Supplementary File 8.” (L. 287 – L. 288)

Furthermore, individual studies included in the meta-analysis showed no significant imbalances in baseline values between the groups. To demonstrate this, the pre- and post-intervention FMD values ​​are listed in Table 1.

 

[Comment 8]

The risk-of-bias tool is non-standard and insufficiently transparent.

The authors use a modified 13-item checklist and sum “risk” or “unclear” ratings into total scores. This approach is less informative than domain-based tools such as RoB 2 for randomized trials. The authors should either use a standard risk-of-bias tool or provide stronger justification for the modified scoring approach. Each judgment should be supported with study-specific explanations. In particular, blinding, allocation concealment, compliance, selective reporting, and outcome assessor blinding need clearer documentation.

[Response 8]

As written in response to Comment 3, a post-hoc assessment using the Cochrane Risk of Bias version 2.0 (RoB2 tool) was conducted in addition to the currently presented method in the prospectively registered protocol. The assessment revealed similar results for both methods. The details were added in the Methods and Results sections.

 

[Comment 9]

Publication bias assessment is not valid with four studies.

Funnel plots and Egger’s test should not be used to infer absence of publication bias in such a small meta-analysis. The text should be revised to state that publication bias could not be meaningfully assessed. The authors should avoid concluding that publication bias is low.

[Response 9]

As written in response to Comment 3, we concluded that publication bias was not evaluated appropriately due to the low number of studies included. Therefore, we corrected the Results section.

 

[Comment 10]

The search strategy is useful but should be strengthened or better justified.

The review searches several databases and registries, including Japanese sources, which is positive. However, the absence of Embase, Web of Science, and Scopus should be justified. The search strategy also focuses heavily on lycopene/tomato/watermelon and may not fully capture all lycopene-rich sources mentioned in the Introduction, such as pink grapefruit or papaya. The authors should verify whether the search strategy is sufficiently comprehensive for the stated scope. The search date should also be updated close to publication, given that the manuscript cites sources accessed in 2026 while the literature search was conducted in September 2025.

[Response 10]

Regarding the limitation of search strategies imposed by databases or search term constraints, the following description was added in the Discussion section as the limitation.

“Second, the search strategy might not be sufficiently comprehensive. We searched four major literature databases and three clinical trial registries to maximize the comprehensiveness of the literature search. Embase, Web of Science, and Scopus were not included because reports in the medical and life sciences indexed in these databases were expected to substantially overlap with those retrieved from PubMed and the Cochrane Library. Furthermore, we used “lycopene” as a primary search term based on the assumption that reports evaluating lycopene-containing foods would be extracted through this approach. Consequently, pink-fleshed Citrus and papaya, which contain significant amounts of lycopene [52,53], were not included as separate search terms. These search strategies might limit the number of reports subject to screening.” (L. 438 – L. 447)

We added the following references in the Reference section. (L. 638 – L. 643)

52. Kumar, A.; Bhawsar, N.; Manekar, S.; Pendram, B.; Pal, P.; Ali, D.; Alarifi, S.; Wanale, S.G.; Singh, S.; Katare, S.; et al. Extraction and optimization of lycopene from selected fruits and their assessment as an ultraviolet ray protectant for Escherichia coli. Food Sci Nutr 2025, 13, e70090. DOI:10.1002/fsn3.70090.
53. Zacarías-García, J.; Karp, D.; Roberts, J.; Tornero-Raga, C.; Zacarías, L.; Jesús Rodrigo, M. Lycopene in citrus fruit: biochemical mechanisms, molecular insights and health implications. Crit Rev Food Sci Nutr 2026, 1–19. DOI:10.1080/10408398.2026.2625334.

 

Also, we again searched the four literature databases and three clinical study registries on June 30, 2026, and the search date and results were updated in the Results section, Figure 1, and Supplementary File 2. No additional studies that met the predefined eligibility criteria were identified in the updated search.

 

[Comment 11]

The list of excluded studies should provide more specific reasons.

Supplementary File 5 often states only “PICOS did not meet the eligibility criteria”. This is insufficiently transparent. For each excluded full-text study, the exact reason should be specified, for example wrong population, wrong intervention, wrong comparator, wrong outcome, uncontrolled design, acute-only design, or inadequate control.

[Response 11]

We added the exact reason for exclusion to Supplementary File 5.

 

[Comment 12]

Health-claim and prevention language should be moderated.

The manuscript refers to cardiovascular disease prevention and suggests functional benefits of lycopene. However, the review evaluates surrogate endothelial-function outcomes, not cardiovascular events, disease prevention, or long-term clinical outcomes. Statements implying cardiovascular prevention or substantiated functional benefit should be toned down.

[Response 12]

Statements implying cardiovascular prevention or substantiated functional benefit were toned down. For example, “improve vascular endothelial function” was corrected to “may have beneficial effects on vascular endothelial function”.

 

[Comment 13]

Dose-duration conclusions are too speculative.

The statement that beneficial effects are achieved by taking at least 15 mg lycopene for 12 weeks is based on very limited evidence and appears strongly dependent on one study. Dose-response conclusions should be presented as hypotheses, not as evidence-based thresholds.

[Response 13]

The statement related to the dose-duration conclusion in the Discussion section was corrected as follows.

“Based on the available evidence, it can be hypothesized that the beneficial effect of lycopene intake on vascular endothelial function may require a daily intake of at least 15 mg for 12 weeks, although this hypothesis warrants further investigation, and the proposed dose and duration thresholds may change as further evidence accumulates.” (L. 372 – L. 375)

Also, we deleted the following statement that was originally included in the Discussion section.

“The guidelines issued by the European Food Safety Authority (EFSA) state that “an increase in fasting FMD caused by the continuous consumption of food ingredients for >4 weeks has a beneficial physiological effect on vascular endothelial function.” Increasing FMD with lycopene intake for 12 weeks, as shown in this SR, strongly sup-ports lycopene's beneficial effect on vascular endothelial function.”

 

[Comment 14]

Conflict-of-interest management should be strengthened.

Disclosure is present, but disclosure alone is not sufficient here. The authors should state whether data extraction, risk-of-bias assessment, and certainty assessment for their own included study were independently checked by someone external to Kagome. If not, this limitation should be explicitly acknowledged.

[Response 14]

In light of your comment, we added the following description in the Discussion section as the limitation.

“Fifth, the authors’ own recent study exerted a substantial influence on the overall conclusions of this SR. Although the study was the only study assessed as low risk of bias and enrolled the largest number of participants among the included studies, the possibility of unintended reviewer bias cannot be ruled out because the review authors were in-volved in the conduct of that study.” (L. 456 – L. 460)

 

Minor comments

[Comment 15]

The abstract should report that the meta-analysis includes only four FMD studies and 182 participants, and that conclusions are limited by the small evidence base.

[Response 15]

As written in response to Comment 1, we corrected the description in the Abstract as follows.

“This systematic review/meta-analysis with moderate certainty evidence suggests continuous intake of lycopene-containing products may have beneficial effects on vascular endothelial function in healthy adults, although the findings come from a small-scale meta-analysis in which the results of a single study exert a substantial influence on the overall conclusions.” (L. 28 – L. 31)

 

[Comment 16]

The phrase “with some scientific support” in the abstract is vague and should be replaced by a clearer statement such as “limited evidence suggests…” or “low-to-moderate certainty evidence suggests…”, depending on the revised certainty assessment.

[Response 16]

As written in response to Comment 3, the certainty of the evidence across the study was comprehensively assessed as “moderate”. We corrected the description in the Abstract as follows.

“This systematic review/meta-analysis with moderate certainty evidence suggests continuous intake of lycopene-containing products may have beneficial effects on vascular endothelial function in healthy adults, although the findings come from a small-scale meta-analysis in which the results of a single study exert a substantial influence on the overall conclusions.” (L. 28 – L. 31)

 

[Comment 17]

The statement that the certainty of evidence is high should be removed unless the certainty assessment is substantially revised and convincingly justified.

[Response 17]

As written in response to Comment 3, the certainty of the evidence across the study was comprehensively assessed as “moderate”.

 

[Comment 18]

The discussion comparing the effect size of lycopene with potassium, CoQ10, and resveratrol should be toned down because the evidence bases differ substantially in population, design, number of studies, and risk of bias.

[Response 18]

We corrected the related discussion as follows.

“Although the number of intervention studies evaluating the effects of lycopene on FMD was limited (four studies), the effect size observed in this SR appeared to be within the range of those reported for other representative food ingredients. Furthermore, the findings may be noteworthy because the effect was observed in healthy participants; however, this interpretation should be considered with caution given the limited evidence base.” (L. 395– L. 400)

 

[Comment 19]

The safety discussion is not central to the review question and should be shortened or clearly separated from efficacy. This review does not establish safety across doses, long-term intake, or vulnerable populations.

[Response 19]

We deleted the following statement that was originally included in the Discussion section.

　　　“Participants ingested test foods containing 6–46.2 mg of lycopene for 7 days to 12 weeks in the continuous-intake studies included in this SR [30–33,35], and none of the adverse events that occurred during the study period could be causally attributed to the intake of the test foods. Lycopene extracted from tomatoes has been certified Generally Recognized as Safe by the U.S. Food and Drug Administration [43], and EFSA also reported the safety of lycopene oleoresin from tomatoes [44]. Given that lycopene is consumed daily from various natural foods, such as tomatoes [13], it is a safe food ingredient with a sufficient history of safety and is considered to have the functional benefit of improving vascular endothelial function.”

 

[Comment 20]

Line/reference issue: the statement “Participants ingested test foods containing 6–46.2 mg of lycopene…” appears to cite references [30–33,35], but references 30–31 are methodological references, not intervention studies. Please check and correct this reference range.

[Response 20]

We deleted the parts as mentioned above.

 

[Comment 21]

The manuscript title and conclusions should distinguish “lycopene-containing foods/products” from “lycopene” as an isolated compound.

[Response 21]

As written in response to Comment 2, we changed the wording from "lycopene" to "lycopene-containing products" in the Title, Abstract, Discussion, and Conclusions sections as needed.

 

[Comment 22]

The PRISMA checklist reports some items as “N.a.”, including sensitivity analyses. Given the small dataset and potential dominance of one study, sensitivity analyses are not optional and should be performed.

[Response 22]

As written in response to Comment 1, we conducted a leave-one-out analysis as a sensitivity analysis. Therefore, we corrected the relevant parts of the PRISMA checklist.

 

Comments on the Quality of English Language

Please correct minor wording issues, for example “the previous meta-analysis” rather than “previous meta-analysis”, and “are employed by Kagome Co., Ltd.” rather than “are employed Kagome Co., Ltd.”

[Response]

We corrected the relevant expressions. Also, the manuscript was corrected by an English editing service.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The systematic review and meta-analysis by Kazutaka Yoshida et al. examined lycopene’s effects on vascular endothelial function in healthy adults for cardiovascular disease prevention by studying various articles from four literature databases (Ichushi-Web, PubMed, Cochrane Central Register of Controlled Trials, and Global Index Medicus), three clinical trial registries (InternationalClinical Trial Registry Platform, ClinicalTrials.gov, and University Hospital Medical Information Network Clinical Trials Registry), and notification information for Foods with Function Claims, using peer-reviewed search strategies. Overall, the manuscript is very well written. This systematic review/meta-analysis suggests continuous lycopene intake improves vascular endothelial function in healthy adults, with some scientific support as claimed by the author. The conclusion section is too short. The author should improve it emphasing on such a study and identifying significant results and future prospects. The English language requires more clarity. The author cited a reference in "conflict of interest", needs to remove it and provide all details without citations. The author should cite more recent references in the manuscript and discuss them in the introduction and discussion parts.

Comments on the Quality of English Language

The English could be improved to more clearly express the research. Extensive language editing is required.

Author Response

Comments and Suggestions for Authors

The systematic review and meta-analysis by Kazutaka Yoshida et al. examined lycopene’s effects on vascular endothelial function in healthy adults for cardiovascular disease prevention by studying various articles from four literature databases (Ichushi-Web, PubMed, Cochrane Central Register of Controlled Trials, and Global Index Medicus), three clinical trial registries (InternationalClinical Trial Registry Platform, ClinicalTrials.gov, and University Hospital Medical Information Network Clinical Trials Registry), and notification information for Foods with Function Claims, using peer-reviewed search strategies. Overall, the manuscript is very well written. This systematic review/meta-analysis suggests continuous lycopene intake improves vascular endothelial function in healthy adults, with some scientific support as claimed by the author. The conclusion section is too short. The author should improve it emphasing on such a study and identifying significant results and future prospects. The English language requires more clarity. The author cited a reference in "conflict of interest", needs to remove it and provide all details without citations. The author should cite more recent references in the manuscript and discuss them in the introduction and discussion parts.

【Response】

Regarding the conclusion section, we revised it as follows, considering the significant limitations of this study.

“This SR with meta-analysis included more recent studies and a broader range of interventions, outcome measures, intake amounts, and intervention durations than the previous SR. It suggested the beneficial effects of continuous intake of lycopene-containing products on vascular endothelial function in healthy adults, although the findings came from a small-scale meta-analysis in which the results of a single study exerted a substantial influence on the overall conclusions. Additional clinical trials are required to evaluate the effects of lycopene on vascular endothelial function.” (L. 462 – L. 468)

 

Regarding the quality of English, we corrected some expressions and the manuscript was corrected by an English editing service.

 

Regarding the conflict of interest, the inclusion of the reference in the COI section was my mistake. We corrected the COI section as follows.

“Three authors (K.Y., Y.N., and S.T.) were employed by Kagome Co., Ltd., which sponsored this study. The authors (K.Y., Y.N., and S.T.) were also investigators and co-authors of one of the studies included in this systematic review. A reward was paid to Prof. Hiroharu Kamioka for guidance and supervision in conducting the SR and to an external librarian for peer review of the literature search formula.” (L. 496 – L. 500)

 

Regarding citing more recent references, I added the following content in the Discussion section as limitations, including citations of research from last year and this year.

“First, the number of interventional studies included in the meta-analysis was limited. We conducted leave-one-out analysis as a sensitivity analysis, but sensitivity analyses considering factors such as study design, the form and type of the intervention, lycopene intake amounts or duration, and ethnicity could not be conducted because of the limited number of included studies. Insufficient information from sensitivity analysis makes it difficult to discuss the dependence of results on a specific study design, the effect of ingredients other than lycopene, the relationship between the duration and amount of lycopene intake and its effect, ethnic differences in the effects of lycopene. Regarding the effect of ingredients other than lycopene, a recent meta-analysis showed that intake of L-citrulline, which is a major ingredient in watermelon, significantly improved FMD [51]. A broader evidence base is needed, accounting for the source of lycopene intake, the duration/amount of lycopene intake, and regional differences. As additional RCTs accumulate, it will become possible to perform sensitivity analyses and clarify the effects of lycopene on vascular endothelial function. Second, the search strategy might not be sufficiently comprehensive. We searched four major literature databases and three clinical trial registries to maximize the comprehensiveness of the literature search. Embase, Web of Science, and Scopus were not included because reports in the medical and life sciences indexed in these databases were expected to substantially overlap with those retrieved from PubMed and the Cochrane Library. Furthermore, we used “lycopene” as a primary search term based on the assumption that reports evaluating lycopene-containing foods would be extracted through this approach. Consequently, pink-fleshed Citrus and papaya, which contain significant amounts of lycopene [52,53], were not included as separate search terms. These search strategies might limit the number of reports subject to screening.” (L. 424 – L. 447)

Additional references are as follows. (L. 635 – L. 643)

51. Luo, P.; Li, Z.; Liu, K.; Gao, W. Effects of L-citrulline supplementation and watermelon intake on arterial stiffness and endothelial function in middle-aged and older adults: a systematic review and meta-analysis of randomized controlled trials. Front Nutr 2025, 12, 1632952. DOI: 10.3389/fnut.2025.1632952.
52. Kumar, A.; Bhawsar, N.; Manekar, S.; Pendram, B.; Pal, P.; Ali, D.; Alarifi, S.; Wanale, S.G.; Singh, S.; Katare, S.; et al. Extraction and optimization of lycopene from selected fruits and their assessment as an ultraviolet ray protectant for Escherichia coli. Food Sci Nutr 2025, 13, e70090. DOI:10.1002/fsn3.70090.
53. Zacarías-García, J.; Karp, D.; Roberts, J.; Tornero-Raga, C.; Zacarías, L.; Jesús Rodrigo, M. Lycopene in citrus fruit: biochemical mechanisms, molecular insights and health implications. Crit Rev Food Sci Nutr 2026, 1–19. DOI:10.1080/10408398.2026.2625334.

 

Comments on the Quality of English Language

The English could be improved to more clearly express the research. Extensive language editing is required.

【Response】

We reviewed the English expression again, and the entire manuscript was corrected by an English editing service.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

I thank the Authors for their detailed response and for the substantial revision of the manuscript. The revised version is considerably improved compared with the previous submission. The authors have addressed several key methodological concerns by reframing the title around “lycopene-containing products”, adding a leave-one-out analysis, updating the literature search, adding post-hoc RoB2 and GRADEpro assessments, expanding the limitations section, providing more specific reasons for excluded studies, and toning down several efficacy and health-claim statements.

Overall, the manuscript is now much more balanced and transparent. In particular, the explicit acknowledgement that the pooled FMD result loses statistical significance when the authors’ own recent study is excluded is important and substantially improves the credibility of the review. The revised conclusion is also more cautious than in the original submission.

However, a few residual issues should be addressed before acceptance. These are mostly minor but important for consistency and for avoiding overinterpretation.

Remaining comments

1. Some wording still attributes the effect to lycopene rather than to lycopene-containing products.
   Although the title and several conclusions were revised appropriately, several parts of the abstract, research question, Methods, Results, and Discussion still refer to “lycopene intake” or “lycopene’s effects” as if the evidence isolated lycopene as the active causal factor. Because most FMD studies used complex food matrices, the authors should consistently use “lycopene-containing products” or “test foods containing lycopene” when referring to the meta-analyzed evidence. References to “lycopene” alone are acceptable only when discussing purified lycopene supplementation or mechanistic/nonclinical evidence.
2. The abstract should remove or further soften “for cardiovascular disease prevention”.
   The review evaluates vascular endothelial function, mainly FMD, not cardiovascular events, cardiovascular disease prevention, or long-term clinical outcomes. The phrase “for cardiovascular disease prevention” remains too strong for an abstract. I suggest replacing it with “in the context of cardiovascular risk assessment” or deleting it.
3. The certainty rating remains somewhat optimistic.
   The change from “high” to “moderate” is a major improvement. Nevertheless, the evidence base remains very small, the lower confidence limit is close to no effect, publication bias cannot be meaningfully assessed, and the pooled result depends strongly on a single study. The authors should either further downgrade the wording to “limited/moderate-certainty evidence” or provide a clearer justification for not downgrading for imprecision and small-study/publication-bias uncertainty. The statement that imprecision is “not serious” should be justified beyond the total sample size alone.
4. Avoid prominently reporting the protocol-based “high” certainty result in the main Results.
   The current text states that the registered protocol assessment yielded “high” certainty, whereas GRADEpro yielded “moderate”. This may confuse readers and may weaken the revised cautious interpretation. The protocol-based result can be retained in supplementary material for transparency, but the main text should emphasize the final comprehensive assessment only.
5. The dependence on the Yoshida et al. study should be stated as directly as possible.
   The current revision acknowledges this issue, but the wording could be even clearer. A sentence such as “the statistically significant pooled effect was not robust to exclusion of Yoshida et al.” would be preferable to more general wording such as “substantial influence”.
6. The dose-duration discussion remains too speculative.
   The revised phrasing is better because it presents the 15 mg/12-week threshold as a hypothesis. However, this hypothesis is still mainly driven by one study and should be framed as exploratory. The authors should avoid any wording implying a practical intake threshold.
7. Post-hoc analyses and protocol deviations should be transparently reflected in the PRISMA checklist.
   The revised manuscript now includes post-hoc RoB2, GRADEpro, leave-one-out analysis, and outcome grouping not fully prespecified in the protocol. The PRISMA checklist item on amendments to protocol/registration should not remain “N.a.” unless the authors explicitly justify why these additions are not considered amendments or deviations.

Comments on the Quality of English Language

Minor language and consistency edits are still needed.
Examples include “may contribute to improve” which should be “may contribute to improving”, and remaining instances of “lycopene intake” where “intake of lycopene-containing products” would be more accurate.

Author Response

【Comment 1】

Some wording still attributes the effect to lycopene rather than to lycopene-containing products.
Although the title and several conclusions were revised appropriately, several parts of the abstract, research question, Methods, Results, and Discussion still refer to “lycopene intake” or “lycopene’s effects” as if the evidence isolated lycopene as the active causal factor. Because most FMD studies used complex food matrices, the authors should consistently use “lycopene-containing products” or “test foods containing lycopene” when referring to the meta-analyzed evidence. References to “lycopene” alone are acceptable only when discussing purified lycopene supplementation or mechanistic/nonclinical evidence.

【Response 1】

　We revised the word “lycopene” to "lycopene-containing products" primarily in the Abstract, Results, and Discussion sections, where such changes were deemed necessary. However, as the original objective of this SR was to evaluate the effects of lycopene intake on vascular endothelial function, we partially retained the word "lycopene" in the sections describing the objective, background, and methodology.

 

【Comment 2】

The abstract should remove or further soften “for cardiovascular disease prevention”.
The review evaluates vascular endothelial function, mainly FMD, not cardiovascular events, cardiovascular disease prevention, or long-term clinical outcomes. The phrase “for cardiovascular disease prevention” remains too strong for an abstract. I suggest replacing it with “in the context of cardiovascular risk assessment” or deleting it.

【Response 2】

　I have corrected the expression you pointed out. (L. 13- L. 14)

 

【Comment 3】

The certainty rating remains somewhat optimistic.
The change from “high” to “moderate” is a major improvement. Nevertheless, the evidence base remains very small, the lower confidence limit is close to no effect, publication bias cannot be meaningfully assessed, and the pooled result depends strongly on a single study. The authors should either further downgrade the wording to “limited/moderate-certainty evidence” or provide a clearer justification for not downgrading for imprecision and small-study/publication-bias uncertainty. The statement that imprecision is “not serious” should be justified beyond the total sample size alone.

【Response 3】

　We presented and compared the results of the certainty of evidence assessment based on the registered protocol and the GRADE approach. No serious concerns were identified regarding indirectness or inconsistency, although some concerns were noted with respect to risk of bias. While no clear concerns regarding imprecision and publication bias were identified based on the protocol and GRADE assessments, the evaluation was limited by the small number of included studies. Consequently, the certainty of the evidence was judged to be "moderate" under limited conditions. Based on the above, the description regarding the certainty of the evidence was revised as follows.

　“The certainty of evidence across the four studies evaluating the effects of continuous intake of lycopene-containing products on FMD was assessed according to the registered protocol (Supplementary File 10) and GRADEpro software (Supplementary File 11). No serious concerns were identified regarding indirectness or inconsistency, although some concerns were noted with respect to risk of bias. While no clear concerns regarding imprecision and publication bias were identified based on the protocol and GRADE assessments, the evaluation was limited by the small number of included studies. Taking these considerations into account, the overall certainty of the evidence was judged to be moderate, although the available evidence remains limited." (L. 300 – L. 308)

　

【Comment 4】

Avoid prominently reporting the protocol-based “high” certainty result in the main Results.
The current text states that the registered protocol assessment yielded “high” certainty, whereas GRADEpro yielded “moderate”. This may confuse readers and may weaken the revised cautious interpretation. The protocol-based result can be retained in supplementary material for transparency, but the main text should emphasize the final comprehensive assessment only.

【Response 4】

　The results of the certainty of the evidence assessed by the registered protocol were presented as Supplementary File 10, rather than within the main text. Only the overall assessment of the certainty of the evidence was stated in the main text.

　

【Comment 5】

The dependence on the Yoshida et al. study should be stated as directly as possible.
The current revision acknowledges this issue, but the wording could be even clearer. A sentence such as “the statistically significant pooled effect was not robust to exclusion of Yoshida et al.” would be preferable to more general wording such as “substantial influence”.

【Response 5】

　I have corrected the expressions you pointed out in the Abstract, Discussion (limitation), and Conclusion section. (L. 31 - L. 32) (L. 456 – L. 458) (L. 467 – L. 468)

 

【Comment 6】

The dose-duration discussion remains too speculative.
The revised phrasing is better because it presents the 15 mg/12-week threshold as a hypothesis. However, this hypothesis is still mainly driven by one study and should be framed as exploratory. The authors should avoid any wording implying a practical intake threshold.

【Response 6】

　We interpreted the considerations regarding intake amount and duration as based on facts within an exploratory observation and revised the description in the Discussion section as follows.

“An exploratory evaluation of the included studies suggested that improvements in FMD were observed with lycopene supplementation at 15 mg/day for 12 weeks or 33 mg/day for 15 days. However, the available evidence remains insufficient to draw conclusions regarding the effectiveness of other doses or intervention durations. Furthermore, one study that was not included in the meta-analysis reported a significant improvement in RH-PAT following supplementation with 15 mg/day of lycopene for 8 weeks [39]. Further well-designed clinical trials across a wider range of doses and intervention durations are needed to clarify the relationship between lycopene intake and vascular endothelial function, as well as the potential influence of dose and intervention duration on this association.” (L. 368 – 377)

 

【Comment 7】

Post-hoc analyses and protocol deviations should be transparently reflected in the PRISMA checklist.
The revised manuscript now includes post-hoc RoB2, GRADEpro, leave-one-out analysis, and outcome grouping not fully prespecified in the protocol. The PRISMA checklist item on amendments to protocol/registration should not remain “N.a.” unless the authors explicitly justify why these additions are not considered amendments or deviations.

【Response 7】

　We added following contents as limitations in the Discussion section.

　“In addition, to aid interpretation of the findings using a broader body of evidence, post hoc analyses were conducted, including risk of bias assessment using RoB 2, certainty of evidence assessment using the GRADE approach, and leave-one-out sensitivity analyses. Because these analyses were exploratory and conducted post hoc, no amendments were made to the registered protocol.” (L. 452 – L. 456)

　I revised the relevant sections of the PRISMA checklist. (Supplementary File 1)

 

【Comments on the Quality of English Language】

Minor language and consistency edits are still needed.
Examples include “may contribute to improve” which should be “may contribute to improving”, and remaining instances of “lycopene intake” where “intake of lycopene-containing products” would be more accurate.

【Response】

Regarding the first point, I revised the expression as pointed out.  (L. 316)

Regarding the second point, I made revisions as stated in Response 1.

Author Response File: Author Response.pdf