The impact of metastatic sites in advanced pancreatic adenocarcinoma, systematic review and meta-analysis of prospective randomized studies.

The real impact of specific sites of metastasis on prognosis of metastatic pancreatic cancer (MPC) is unknown. To evaluate the association of specific metastatic sites and survival outcomes in MPC a systematic literature review was performed including prospective randomized trials of systemic treatments in metastatic pancreatic cancer indexed in PubMed, Embase and Web of Science. Data regarding systemic treatment regimens, progression free survival and overall survival were extracted. The outcomes were compared using a random effects model. The index I2 and the graphs of funnel plot were used for the interpretation of the data. Of 1,052 abstracts, 7 randomized trials were considered eligible with a combined sample size of 2,975 MPC patients. Combining the studies with meta-analysis, we could see that patients with liver metastasis had a HR for death of 1.53 with 95% CI of 1.15 to 2.02 (p-value 0.003) and HR for risk of progression of 1.96 with 95% CI of 1.28 to 2.99 (p-value 0.002), without significant heterogeneity. Having two or more sites of metastasis comparing to one site did not have impact on overall survival; RR of 1.05 with 95% CI 0.91 to 1.23 (p-value 0.493). In conclusion, liver metastasis confers worse outcomes among patients with MPC. Apparently, multiple metastatic sites do not present worse prognosis when compared with only one organ involved, therefore, demonstrating the severity of this disease. Prospective studies evaluating other treatments are necessary to address the impact of local treatments in liver metastasis in MPC.

Introduction

Pancreatic cancer (PC) is a deadly cancer, being among the five most lethal malignant tumors in the last years [1]. Most patients are diagnosed with advanced disease and few survival gains were obtained in the past decades [2].

Efforts have been made to obtain better outcomes. Lately, numerous studies have described molecular advances to improve and develop guided therapies [3-5]. However, the applicability of molecular classifications in treatment decisions is still a topic to be evaluated [6].

A deep understanding about the disease and its particularities is fundamental, in terms of outcomes and prognosis. Based on the current literature, the prognostic relationship with some factors—such as pre and postoperative CA19-9 serum levels, lymph node involvement and tumor size—are already well established [7,8].

Criteria such as volume of disease and number of metastatic sites are related to outcome in other tumor subtypes, such as prostate cancer [9]. In metastatic pancreatic cancer (MPC), there also appears to be an association between outcomes and specific metastasis. There are research groups reviewing this question and suggesting stratification of metastasis by different sites, however, few studies have approached this subject in the published literature [10].

To elucidate these questions, this study aims to perform a systematic review and meta-analysis of randomized trials that collected data from metastasis sites and evaluate the association of specific metastatic sites and survival outcomes in MPC.

Materials and methods

Search strategy

This study was designed in conformity with the 2009 Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement guidelines [11]. We searched PubMed (1950–2019) on July 2019. We used the keywords (((((((((((pancreatic neoplasms [MeSH Terms]) OR (pancreatic neoplasms[Title/Abstract] OR Pancreatic cancer[Title/Abstract] OR pancreatic tumor*[Title/Abstract] OR pancreatic tumour* [Title/Abstract] OR pancreas neoplasms[Title/Abstract] OR pancreas cancer[Title/Abstract] OR pancreas tumor*[Title/Abstract] OR pancreas tumour* [Title/Abstract])) OR (pancreatic neoplasms[Text Word] OR Pancreatic cancer[Text Word] OR pancreatic tumor* [Text Word] OR pancreatic tumour* [Text Word] OR pancreas neoplasms[Text Word] OR pancreas cancer[Text Word] OR pancreas tumor* [Text Word] OR pancreas tumour* [Text Word])) OR pancreatic ductal adenocarcinoma [Title/Abstract]) OR pancreatic ductal adenocarcinoma [Text Word]) OR "cancer of the pancreas" [Text Word]) and Best Match Filters: Clinical Trial, Phase III; with 196 abstracts. We searched Embase and Web of Science using the following keywords (((('pancreas'/mj AND cancer* OR pancreatic) AND cancer* OR pancreatic) AND neoplasm* OR pancreatic) AND adenocarcinoma* OR pancreatic) AND tumor* AND [randomized controlled trial]/lim. Results retrieved more 926 abstracts. After excluding duplications and adding 5 more records identified through other sources (google search), the final database sample was made up of 1,052 records ( The protocol (number 3493–18) was registered in the system for research project management (SGPP) of Hospital Israelita Albert Einstein. The protocol is available for consultation upon request.

Two authors (P.L.S.U.J. and V.M.S.) reviewed all abstracts. Inclusion criteria were: (1) randomized prospective cohorts investigating first-line systemic treatment; (2) patients with metastatic pancreatic adenocarcinoma; (3) available information of metastatic sites. Exclusion criteria were: (1) language other than English; (2) duplicate publication; (3) review articles, and case reports; (4) clinical trial protocols.

After a preliminary review, 133 full-text articles were selected for evaluation ( Reasons for exclusion were: non-randomized study (n = 43), review articles and meta-analysis (n = 215), experimental research (n = 118), non-first-line therapy (n = 34), non-pancreatic adenocarcinoma (n = 224), resectable disease/ locally advanced (n = 237) and clinical trial protocol (n = 48).

After review of full-text articles, we excluded 126 articles. Reasons for exclusions were no data about metastatic sites (n = 88) and duplicates (n = 38). Seven articles were selected for meta-analysis.

Definitions and outcomes of interest

Two authors (P.L.S.U.J. and F.T.) extracted data of all included studies using a standardized data collection form. Primary outcome was overall survival (OS) related to metastatic sites. Secondary outcome was evaluation of progression free survival (PFS). Data were collected as published in the studies. Nonrandomized retrospective and phase II studies were not included in this analysis because they have a higher risk of bias than randomized studies and the resultant larger sample size with just randomized data provides greater reliability.

Data extraction, data synthesis and analysis

For meta-analysis we considered articles that met the inclusion criteria and presented a hazard ratio (HR) followed by standard errors / confidence intervals or medians of survival of the groups accompanied by the number of events in each group (to allow the estimation of variability). For studies that presented HR and CI, we estimated the log (HR) by log HR transformation and the standard error (SE) was estimated as (log (upper limit of the CI)—log (Lower limit of the CI)) / 3.92. For the studies that we don’t have CI, the confidence interval limits will then be given by HR—1.96 x SE (HR) and HR + 1.96 x SE (HR), the comparison of patients with liver metastasis was performed with patients with metastasis elsewhere.

For studies that presented the median survival of two or more groups, and the number of events in each group, we estimated the risk ratio (RR) for each group compared with a group defined as a reference [12].

We combined data using a random-effects-model meta-analysis, because we assumed that results of different studies depended not only on the sample variation and the covariables investigated, but also on other factors. The I2 index was used to measure the heterogeneity of the results of the different studies. Publication bias was evaluated by funnel plot and analyses were performed using the R [13] and Meta [14] packages, considering a significance level of 5%.

Quality assessment

Methodological quality assessments of the included studies were accessed using the Revised Cochrane risk-of-bias tool for randomized trials. The scale is constituted by five domains, namely (1) bias arising from the randomization process; (2) bias due to deviations from intended interventions; (3) bias due to missing outcome data; (4) bias in measurement of the outcome; (5) bias in selection of the reported result. Each domain is judged as low risk, some concerns or high risk of bias, two authors (P.L.S.U.J. and V.M.S.) have done the quality assessment [15].

Results

All seven studies were published after 2009. Sample size ranged from 125 to 607 patients, with a combined sample of 2,975 patients. Almost all included studies evaluated gemcitabine-based chemotherapy regimens [16-21]. Only the study from Conroy et al. evaluated FOLFIRINOX [22]. Two randomized phase 2 [16,19] and five randomized phase 3 studies [17,18,20-22] presented data related with the presence or absence of liver metastasis only, in metastatic patients, two studies also reported data (number of events) on lung metastasis and comparative outcomes of 1 or more metastatic sites [21, 22], one study reported results of lymph node metastasis [17], and one about peritoneal metastasis [21] ( The studies were evaluated using the Revised Cochrane risk-of-bias tool for randomized trials (, details about the assessment in each trial and other information’s can be found in .

Liver metastasis

The HR for death in the presence of liver metastasis ranged in the studies between 0.83 (0.68; 1.02) -2.36 (1.63; 3.42), only one study did not have the hazard for death [22]. Comparing the six studies (those that presented outcome data related to each medication arm were included as individual data in meta-analysis, i.e. Fuchs et al. 2015 and Fuchs et al. 2015.1. refers to study arms 1 and 2) including 2,633 patients, we had a global estimate for HR of 1.53 with 95% CI of 1.15 to 2.02 (p-value 0.003), without significant heterogeneity (I2 = 5.03%) (. Regarding publication bias, the funnel plot () shows a lack of studies with more patients than the polled average. Unfortunately, only seven of randomized trials had objective data on metastasis outcomes. However, in a sample of more than 2,500 patients, the inclusion of more studies would have a little effect in the outcome of this analysis.

Forest Plot for risk of death and liver metastasis based on hazard ratios (N = 2633).

Overall the presence of hepatic metastasis increases the risk of death around 50% (HR1,53 CI 95% [1,15; 2,02] (p 0,003).

Funnel plot for HR of OS and liver metastasis.

Funnel plot shows the absence of studies with a greater number of patients than the mean analyzed.

Three studies reported the presence of liver metastasis and PFS [16,18,19]. With a combined sample size of 817 patients, a global estimation by meta-analysis obtained a HR for relapse risk (progression) of 1,96 with 95% CI of 1.28 to 2.99 (p-value 0.002). The I2 index was 0.00% (.

Lymph nodes, lung and peritoneum

The analysis of other metastatic sites such as lymph nodes, lungs and peritoneum were not performed because just one study included HR for OS for lymph nodes [17], and one study reported data of mean OS for lung metastasis [21]. It is important to mention that the HR for OS with lung metastasis in the Conroy et al. [22] study was calculated by authors (by number of events) with a HR of 0.89; CI 95% of 0.78 to 1.02 (p-value 0,089), but no statistical significance was found.

Number of metastatic sites

For the presence of two, three or more metastasis sites in relation to one regarding death risk, two studies reported numbers of events [21, 22]. Interestingly, these studies considered regimens that are currently used in first line setting (FOLFIRINOX and Gemcitabine-Nab-Paclitaxel). None of the combined analyzes of the two studies compared two, three or more metastasis sites with one site only presented by statistically significant results, i.e., RR of 1.10 with 95% CI of 1.00 to 1.21 (0.061) I2 index 0% and RR of 1.05 with 95% CI 0.91 to 1.23 (p-value 0.493) I2 index 43.10% (. These results suggested that number of different metastasis sites have no impact on the overall survival outcome for patient with metastatic pancreatic cancer.

Forest plot for relative risk of death with three or more sites of metastasis compared to one site.

The forest plot based on number of events, shows absence of statistical significance of the sites of metastasis in relative risk of death.

Discussion

This meta-analysis of prospective randomized studies showed that patients with liver metastasis in MPC have worse outcomes (PFS and OS). The analysis of number of metastatic sites did not show any association.

Several tumors have different behavior and outcomes depending on the involved metastatic sites. In breast cancer cases, visceral involvement is associated with worse outcomes compared with bone involvement [23]. The same is seen in prostate cancer in which randomized trials are designed to define low- or high-volume of metastatic disease and outcomes [24,25]. Most of the international series evaluating MPC have demonstrated that OS for most patients is less than one year, and classifications with respect to metastatic sites are not currently performed, probably because there are poor outcomes [1]. However, trends in reduced mortality are being seen [26].

An extensive systematic literature review was performed by Cannistra and colleagues to determine the impact and outcomes of metastasis in pancreatic cancer [10]. Most studies included in the analysis were case series and reports, and few randomized data were included. They concluded that liver metastasis provides better outcomes than metastasis from other sites, and particularly better outcomes with isolated liver metastasis. Of note is that, most of the included studies evaluated local treatments as resection and given that journals often favor positive-outcomes findings the risk of publication bias should be also considered.

Liver is one of the major metastatic sites in gastrointestinal cancers. Randomized trials have addressed resection combined with perioperative chemotherapy in colorectal cancer and have reported better outcomes using these strategies. Several case series have pointed out benefit of locoregional treatments in liver metastasis in MPC, but randomized data on the subject are scarce [10,27]. Our results show that liver metastasis confer a HR for death of 1.53 with 95% CI of 1.15 to 2.02 (p-value 0.003), without significant heterogeneity (I2 = 5.03%) and a HR for risk of relapse of 1.96 with 95% CI of 1.28 to 2.99 (p-value 0.002). Based on these poor outcomes among patients with metastatic disease and the studies previously discussed, locoregional strategies including resection may be an option for future trials.

In this systematic review, 133 randomized trials were included, but only seven had individual metastatic sites related with data outcomes. Despite the exclusion of most of studies, the analysis could be performed with a sample of more than 2,900 randomized patients, strengthening the results that were found. Other relevant fact is the inclusion in the analysis of two positive chemotherapy randomized phase 3 trials with regimens that are used in clinical practice [21,22]. These findings demonstrate a worse outcome of liver metastasis compared with other sites in patients treated with modern chemotherapy regimens.

A consensus statement from a group of experts from the pancreatic cancer field provided mandatory and recommended baseline measurements and prognostic characteristics to be included in trials investigating palliative systemic therapy for MPC. Sites and number of metastatic sites in MPC were considered mandatory variables, and these mandatory prognostic factors should be included in regression analyses to adjust their effect on treatment outcomes [28]. This recommendation provides strength to our findings and reinforces the need for a better characterization of the impact of sites and number of metastatic sites in MPC in outcomes of randomized trials. The adequacy of future clinical trials evaluating MPC could include analyses of non-conventional metastatic sites such lungs, bones and central nervous system. The analyses of the impact of these sites were not possible to perform in our study because few studies considered regression data from these sites.

Publication bias in this study was evaluated by the funnel plot. We can observe that largest studies with small standard errors, placed at the top of the graph were absent. This bias could be overcome by the inclusion of more randomized trials, but as previously discussed most of studies did not present regression models regarding metastatic sites. It is worth to emphasize that funnel plot is more appropriate for reviews with more than ten studies, but the inclusion of more patients would hardly change the results found.

This meta-analysis has some limitations, publication bias and lack of patient-level data. Considering that we included only published data, we clearly observed in the analysis of funnel plot the lack of studies, which lead to the pattern found. Publication bias is intrinsic to this type of study, but such bias becomes less relevant because of the number of patients included in the analysis and the quality of the studies included. Some trials also included locally advanced disease, but we used for hazard ratio comparison liver metastasis versus overall metastasis. The randomization of the original studies included in this analysis was performed regarding the treatment. We are not considering two treatments comparisons in this meta-analysis, but the real impact of different systemic treatments would hardly change the results found since no systemic treatment confers more than 10–12 months of overall survival in any randomized trial for metastatic pancreatic adenocarcinoma. Finally, individual data of patients could have helped to identify other sources of heterogeneity and the inclusion of more studies in the analysis, particularly in analysis of other metastatic sites individually, volume of liver disease and performance status (most of the randomized trials included just patients with good performance status). Analysis of other metastatic sites such as lymph nodes, lungs and peritoneum was not possible because all the studies evaluated did not include individual outcomes of these sites to draw exact conclusions of their real impact on patient outcomes, in addition, lymph node and peritoneal metastasis usually is accompanied by other sites of metastasis and multiple site analysis was possible, with no statistically significant difference found. An analysis of large survival series of pancreatic cancer patients may help to stratify the impact of performance status and other sites of metastasis in MPC. The strength of our analysis is related to the inclusion of a fair number of prospective well-designed studies with a combined sample size of more than 2,900 patients, and report of consistent findings.

The absence of association of more than two metastatic sites compared with one site outcomes shows that MPC is an aggressive disease that requires incorporation of more strategies for better stratification.

Conclusion

Liver metastasis confers worse outcomes (PFS and OS) in patients with MPC. Apparently multiple sites of metastasis do not provide worse prognosis when compared with only one organ. Prospective randomized studies evaluating other treatments such as liver-directed therapies are warranted.

Forest plot for relative risk of progression due to liver metastasis.

(TIF)

Click here for additional data file.

(XLSX)

Click here for additional data file.

Summary of quality assessment.

(XLSX)

Click here for additional data file.

PRISMA checklist.

(DOC)

Click here for additional data file.

6 Jan 2020

PONE-D-19-32606

The impact of metastatic sites in advanced pancreatic adenocarcinoma, systematic review and meta-analysis of prospective randomized studies

PLOS ONE

Dear Dr. Pedro Uson Junior,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

==============================

ACADEMIC EDITOR: I have comments for this article as follows:

1. Please consider that the outcome from first-line and second-line studies might be different.

2. In figure 2, all included studies show no impact of survival, but the final HR was a positive impact of liver metastasis to the overall survival? How come? Please explain. Also, why the authors have Fuchs et al. 2015 and Fuchs et al. 2015.1. There is no explanation in the text and analysis. Do they refer to study arms 1 and 2? Please clearly describe them in the manuscript.

3. The legend of Figure 2 illustrates an HR 1.53 (1.15-2.02, p=0.003), which seems different from the figure itself (0.42[0.14-0.70]). Please explain.

4. There were no clear abbreviations in the legend for some short words in the figures. Please correct them.

5. Liver metastasis includes huge liver (higher %) metastasis and tiny liver metastasis (lower % of occupying lesions). The study failed to exclude the factor, resulting in a not scientific enough conclusion. Please at least declare the limitations of the study.

Please respond to these questions as well.

==============================

We would appreciate receiving your revised manuscript by Feb 20 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Jason Chia-Hsun Hsieh, M.D. Ph.D

Academic Editor

PLOS ONE

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. We noticed you have some minor occurrence(s) of overlapping text with the following previous publication(s), which needs to be addressed:

https://doi.org/10.1038/s41598-019-52334-y

https://doi.org/10.1016/j.clcc.2018.03.007

In your revision ensure you cite all your sources (including your own works), and quote or rephrase any duplicated text outside the Methods section. Further consideration is dependent on these concerns being addressed."

3. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

Additional Editor Comments (if provided):

I have comments for this article as follows:

1. Please consider that the outcome from first-line and second-line studies might be different.

2. In figure 2, all included studies show no impact of survival, but the final HR was a positive impact of liver metastasis to the overall survival? How come? Please explain. Also, why the authors have Fuchs et al. 2015 and Fuchs et al. 2015.1. There is no explanation in the text and analysis. Do they refer to study arms 1 and 2? Please clearly describe them in the manuscript.

3. The legend of Figure 2 illustrates an HR 1.53 (1.15-2.02, p=0.003), which seems different from the figure itself (0.42[0.14-0.70]). Please explain.

4. There were no clear abbreviations in the legend for some short words in the figures. Please correct them.

5. Liver metastasis includes huge liver (higher %) metastasis and tiny liver metastasis (lower % of occupying lesions). The study failed to exclude the factor, resulting in a not scientific enough conclusion. Please at least declare the limitations of the study.

Please respond to them as well.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: No

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: I Don't Know

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Comment 1: Statistically, this is a well executed meta-analysis. Moreover, the data from this manuscript may help in designing future trials and adjusting randomization by metastatic site in accrual (knowing survival biases in patients assignment), indeed PC with lung metastasis & bone metastasis are at lower death risk than those with liver metastasis.

Comment 2: Is the HR of "liver metastasis" versus the "overall metastasis" (any kind considered) and/or "locally advanced" as comparator? This is not well stated in the manuscript. Could you please clearly state the comparator to the "liver metastasis". Also, considering that "locally advanced" patients are included, it probably would be fair to mention this in the limitations.

Comment 3: line 153; What do you mean by "PE" (is it Standard Error)? Also, on line 153 you mention that you estimate the "Confidence Interval limits", but you mention on line 148-149 that confidence intervals are supplied by the original articles. Why do say, that you you re-estimate the Confidence Intervals, could you please explain?

Comment 4: Why do you use different scales to report the "HRs & 95%CI", between the "main text" and the Forest Plots (Figure 2, Figure 4 & Supplementary 3). For the Forest Plots (Figure 2, Figure 4 & Supplementary 3) you use a "logarithmic scale" which is not stated for for Figure 2 & Supplementary 3. Providing the data on a different scale between the main text and the figures is confusing; could you please provide also the data on the Figures on a normal scale?

Comment 5: Forest Plots (Figure 2, Figure 4 & Supplementary 3); could you please provide the "group at favor" on the "x_axis", on each side between the "vertical line of no-effect". For example, Forest Plot Figure_2: providing the side that favors "metastasis on the liver" vs the side that favors "no metastasis on the liver" on the "x_axis", could make the Figure easier to understand.

Comment 6: line 212, you report a "HR", but for the Figure (Supplementary data 3), you mention "relative risk". Could you please explain?

Comment 7: Despite the fact that data were from RCTs, the level of evidence seems still to be low. This meta-analysis reports outcomes with respect to the site(s) of metastasis, the treatment arms are not considered to be the comparisons in this particular case.The randomization of the original studies was performed with regard to the treatment. We are not considering two treatments comparisons in this meta-analysis, so randomization does not seem to provide security for the robustness of the outcome when considering metastatic site(s) as comparison. Could you please elaborate?

Comment 8: Patients enrolled in RCTs are strongly selected "PS 0-1". So the conclusions pertain more properly to patients in good PS and receiving the investigated chemo regimens; making it hard to make proper conclusions for the "real life" overall metastatic PC population. Could you please elaborate in the discussion? (and probably provide solutions, for example would a separate section analysis from large survival series of data from non randomized trials be of value?)

Reviewer #2: The authors evaluated the impact of metastatic site of pancreatic adenocarcinoma based on meta-analysis of prospective randomized studies. The liver was well- known to be most common metastatic site of pancreatic adenocarcinoma, and patients with disease progression usually died of hepatic failure. Other metastasis sites including the lung, lymph nodes or peritoneum should be included into analysis. Different treatment regimens were also found in the trials which may cause bias in evaluating the impact of metastatic sites on survival. The patient’s general performance and comorbidity will also influence the prognosis that cannot be assessed in this study. The results of this study did not provide additional information to our current knowledge to this disease.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

29 Jan 2020

Thank you very much for the review and suggestions.

Response to reviewers:

Reviewer#1:

Comment 2: Is the HR of "liver metastasis" versus the "overall metastasis" (any kind considered) and/or "locally advanced" as comparator? This is not well stated in the manuscript. Could you please clearly state the comparator to the "liver metastasis". Also, considering that "locally advanced" patients are included, it probably would be fair to mention this in the limitations.

Response: Included the comparator in the methods (overall metastasis). Some trials included locally advanced disease, but we used as comparator just metastatic patients, we have included this statement in the limitations.

Comment 3: line 153; What do you mean by "PE" (is it Standard Error)? Also, on line 153 you mention that you estimate the "Confidence Interval limits", but you mention on line 148-149 that confidence intervals are supplied by the original articles. Why do say, that you you re-estimate the Confidence Intervals, could you please explain?

Response: We are sorry about this typo error, corrected in the line 153 (Standard error=SE). For the articles that the CI was available we used the value of the article, in the articles that CI was not available we calculated using the available data.

Comment 4: Why do you use different scales to report the "HRs & 95%CI", between the "main text" and the Forest Plots (Figure 2, Figure 4 & Supplementary 3). For the Forest Plots (Figure 2, Figure 4 & Supplementary 3) you use a "logarithmic scale" which is not stated for for Figure 2 & Supplementary 3. Providing the data on a different scale between the main text and the figures is confusing; could you please provide also the data on the Figures on a normal scale?

Comment 5: Forest Plots (Figure 2, Figure 4 & Supplementary 3); could you please provide the "group at favor" on the "x_axis", on each side between the "vertical line of no-effect". For example, Forest Plot Figure_2: providing the side that favors "metastasis on the liver" vs the side that favors "no metastasis on the liver" on the "x_axis", could make the Figure easier to understand.

Comment 6: line 212, you report a "HR", but for the Figure (Supplementary data 3), you mention "relative risk". Could you please explain?

Response: Sorry about the confuse forest plots, the graphs were remade on a non-logarithmic scale. We also included the information suggested to make it clearer.

The altered graphics:

- Comparison of metastasis in liver for overall survival - studies with HR

- Comparison of metastasis in liver for progression-free survival - studies with HR

- Comparison of studies with three sites of metastasis in relation to one site - studies with number of events.

The first two have a measure of HR effect because we were able to extract this information from the studies and use it in the meta-analysis. In the case of comparing the number of metastatic sites, the studies presented only the number of cases and not the hazard ratio, so we did the meta-analysis considering the proportion of events, the summary measure is relative risk.

Comment 7: Despite the fact that data were from RCTs, the level of evidence seems still to be low. This meta-analysis reports outcomes with respect to the site(s) of metastasis, the treatment arms are not considered to be the comparisons in this particular case. The randomization of the original studies was performed with regard to the treatment. We are not considering two treatments comparisons in this meta-analysis, so randomization does not seem to provide security for the robustness of the outcome when considering metastatic site(s) as comparison. Could you please elaborate?

Response: This was certainly one of the biggest limitations in the development of this project since there are no randomized data for metastasis in advanced pancreatic adenocarcinoma, our goal was to generate some solid evidence for the development of future studies based on this aspect. Regarding the treatment-related outcome, there is no impression that the treatments would objectively influence the results found, since no type of systematic treatment for advanced pancreatic cancer resulted in more than 10-12 months of median survival in any randomized trial. We included this information of randomization in the limitations of the study.

Comment 8: Patients enrolled in RCTs are strongly selected "PS 0-1". So the conclusions pertain more properly to patients in good PS and receiving the investigated chemo regimens; making it hard to make proper conclusions for the "real life" overall metastatic PC population. Could you please elaborate in the discussion? (and probably provide solutions, for example would a separate section analysis from large survival series of data from non-randomized trials be of value?)

Response: Included in the limitations of the study the PS in the RCT included and the need to analyze patient level data and large survival series data to draw better conclusions.

Reviewer #2:

The authors evaluated the impact of metastatic site of pancreatic adenocarcinoma based on meta-analysis of prospective randomized studies. The liver was well- known to be most common metastatic site of pancreatic adenocarcinoma, and patients with disease progression usually died of hepatic failure. Other metastasis sites including the lung, lymph nodes or peritoneum should be included into analysis. Different treatment regimens were also found in the trials which may cause bias in evaluating the impact of metastatic sites on survival. The patient’s general performance and comorbidity will also influence the prognosis that cannot be assessed in this study. The results of this study did not provide additional information to our current knowledge to this disease.

Response: We cannot agree with the reviewer. The COMM-PACT consensus suggested stratification of metastatic pancreatic cancer in specific metastatic sites for all randomized studies to be developed in the future, our meta-analysis provides an objective analysis of the impact of metastatic site on disease outcomes, prospective studies that specifically assess this aspect do not exist.

ACADEMIC EDITOR:

Thank you very much for the review.

1. Please consider that the outcome from first-line and second-line studies might be different.

Response: We included just first-line trials in this meta-analysis. Line 113: Inclusion criteria were: (1) randomized prospective cohorts investigating first-line systemic treatment;

2. In figure 2, all included studies show no impact of survival, but the final HR was a positive impact of liver metastasis to the overall survival? How come? Please explain. Also, why the authors have Fuchs et al. 2015 and Fuchs et al. 2015.1. There is no explanation in the text and analysis. Do they refer to study arms 1 and 2? Please clearly describe them in the manuscript.

Response: The articles included have negative results between comparison among treatments in the randomized trial, but the site of metastasis (liver versus other) confers worse outcomes among them. We included this limitation about the treatment arm in the discussion, after the reviewer 1 response. We are sorry about the explanation, Fuchs et al. 2015 and Fuchs et al. 2015.1. are each arm of the trial, we included an explanation in the results: 3.1 Liver Metastasis.

3. The legend of Figure 2 illustrates an HR 1.53 (1.15-2.02, p=0.003), which seems different from the figure itself (0.42[0.14-0.70]). Please explain.

Response: Sorry about the confuse forest plots, the graphs were remade on a non-logarithmic scale. We also included the information suggested by the reviewer 1 to make it clearer.

4. There were no clear abbreviations in the legend for some short words in the figures. Please correct them.

Response: Corrected as directed.

5. Liver metastasis includes huge liver (higher %) metastasis and tiny liver metastasis (lower % of occupying lesions). The study failed to exclude the factor, resulting in a not scientific enough conclusion. Please at least declare the limitations of the study.

Response: Unfortunately, no randomized trial in metastatic pancreatic cancer considers volume of liver disease but only presence or absence, included this limitation in the discussion.

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

Response: Corrected as directed.

2. We noticed you have some minor occurrence(s) of overlapping text with the following previous publication(s), which needs to be addressed:

https://doi.org/10.1038/s41598-019-52334-y

https://doi.org/10.1016/j.clcc.2018.03.007

In your revision ensure you cite all your sources (including your own works), and quote or rephrase any duplicated text outside the Methods section. Further consideration is dependent on these concerns being addressed.

Response: Thanks for the guidance, corrected as directed (introduction).

3. Please include captions for your Supporting Information files at the end of your manuscript, and update any in-text citations to match accordingly. Please see our Supporting Information guidelines for more information: http://journals.plos.org/plosone/s/supporting-information.

Response: Corrected as directed.

Submitted filename: Response_to_Reviewers.docx

Click here for additional data file.

21 Feb 2020

The impact of metastatic sites in advanced pancreatic adenocarcinoma, systematic review and meta-analysis of prospective randomized studies

PONE-D-19-32606R1

Dear Dr. Uson Junior,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Jason Chia-Hsun Hsieh, M.D. Ph.D

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

All the questions were answered adequately.

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

26 Feb 2020

PONE-D-19-32606R1

The impact of metastatic sites in advanced pancreatic adenocarcinoma, systematic review and meta-analysis of prospective randomized studies

Dear Dr. Uson Junior:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Jason Chia-Hsun Hsieh

Academic Editor

PLOS ONE

Table 1

Summary of included studies.

Study	Type	Treatment	N°	Data on metastasis
				Liver	Lung	Sites
Borad et al. (2015) [16]	Phase II	Gemcitabine vs Gemcitabine plus TH-302	214	Y	N	N
Conroy et al. (2011) [22]	Phase III	Gemcitabine	171	Y	Y	Y
		FOLFIRINOX	171	Y	Y	Y
Deplanque et al. (2015) [17]	Phase III	Gemcitabine vs Gemcitabine plus masitinib	348	Y	N	N
Fuchs et al. (2015) [18]	Phase III	Ganitumab 12 mg/kg plus Gemcitabine	318	Y	N	N
		Ganitumab 20 mg/kg plus Gemcitabine	160	Y	N	N
Kindler et al. (2012) [19]	Phase II	Gemcitabine plus Ganitumab 12 mg/kg or Conatumumab 10mg/kg or placebo	125	Y	N	N
Van Cutsem et al. (2009) [20]	Phase III	Gemcitabine plus erlotinib vs Gemcitabine plus erlotinib plus bevacizumab	607	Y	N	N
Von Hoff et al. (2013) [21]	Phase III	Gemcitabine	430	Y	Y	Y
		Gemcitabine plus Nab-paclitaxel	431	Y	Y	Y

N°: Number of patients; TH-302: Evofosfamide; FOLFIRINOX: fluorouracil, irinotecan and oxaliplatin; Y: Yes; N: No