A Descriptive Study of the Types and Survival Patterns of Saudi Patients with Multiple Primary Solid Malignancies: A 30-Year Tertiary Care Center Experience
, Rakan B. Alanazi 1,2, Sultan T. Algarni 1,2, Ahmed Abdullah R. Alhuntushi 1,2, Mohammed Qasim Alshehri 1,2, Yusra Sajid Chachar 2,3, Mohammad Alkaiyat 2,4 and Fouad Sabatin 2,4
Round 1
Reviewer 1 Report
The manuscript gives a report on descriptive study of the types and survival patterns of Saudi patients with multiple primary solid malignancies. With longer lifespan the cancer burden is increasing and each country should research cancer also on population level, possibly leading to (some sort of) national cancer plan. I have some major suggestions for improvement of the manuscript.
· - Lines 50-51: Sentence is unclear. Probably it is meant that “… has decreased by 29% in 2017 compared to 1991.” Same comment for line 212.
· - Lines 55-57: Please make it more clear how you differentiated recurrences and progressions from second primary cancers.
· - Subchapter 2.3: In case the first primary cancer was solid tumour and the second primary cancer was haematological, the case was excluded from the analysis? It should be written how many of (probably) cancer cases were excluded due to missing data or missing medical record? As many numbers as possible should be written and if possible presented by a chart. Also give (estimated) number of patients who were diagnosed with second primary cancer in some other heath institution and consequently not included into your analysis.
· - Subchapter 2.4: What classification was used to determine anatomical location? ICD-10 or some other? And what classification was used to determine morphology?
· - Subchapter 2.4: What is the size of population covered by the hospital?
· - Subchapter 2.5: If survival was calculated, please report how you defined the starting date (date of diagnosis of first primary cancer) and end date. It is written in lines 157-158 but it should be moved in this subchapter. Furthermore, details are missing on how you followed up for vital status.
· - Line 115: A test is not considered significant but the result is.
· - Line 125-126: 321 out of 450 is 71.3% is quite low. Please explain why 28.7% cases were not eligible and if this introduces any (possible) bias.
· - Line 130: How was the prevalence rate calculated? Was the number of total first primary cancers taken into account? Clarify the calculation of prevalence rate. In case the number of first primary cancers is known in the study period, please report it, because it is important for the interpretation. By first primary cancers I mean all first primary cancers and not only those followed by second primary cancers (MPMs).
· - Lines 138-139: also in connection to the previous comment, the numbers (and percentages) of anatomic sites are interesting only when the comparison is made to the number (and percentages) of this anatomic site in general population (e.i. the total incidence, not only MPMs).
· - Line 133: There is an error with numbers, since SE can’t be bigger than average. If the test yields bigger SE, than the distribution is not normal and the average should not be calculated.
· - Table 1: The last two titles in the table are the same.
· - Table 1: Age should also be reported (median and range).
· - Tables 2 and 3: The prostate cancer is one of the most common male cancers – give explanation in the discussion why numbers in your study are so low.
· - Figures 1, 2 and 3: In the scientific literature 3D pie charts are not appropriate. The graphs could also be modified and compiled so they would take less space. Some could be omitted.
· - Figures 5 and 6: In SPSS one can edit the graphs. The ones in the manuscript are appropriate for publication. For example, the name of the x axis should be only Time in months, number without decimals, clearer text in legend, no title, …
· - Figures 5: not all results need to be presented in graphs. Please also report in the text what are the median survival time and p-values when analysis is done separately for synchronous only and separately for metachronous only.
· - Line 207: If all number are given to one decimal point, also 10.0 should be written instead of 10.
· - Lines 225-226: Due to specifics of childhood and adolescent tumours and late effect of cancer treatment in young population, the whole analysis should be performed separately for patients at age 0-19 years at diagnosis of first primary cancers. In the analysis of they should not be joined with adults. The issue itself should be mentioned in the discussion (there are lots of international publication on this topics).
· - Some second primary cancers are related or caused by first primary cancer, some second primaries are a consequence of treatment of first primary cancer, some are related to common risk factors (at individual patient level) and some are independent. This issue should also be addressed in the discussion.
· - Lines 243-245: It might also depend on the lifespan and accessibility of health services to general population.
· - Lines 299-300: The sentence has awkward wording. The difference is statistical significant (and not median survival time itself).
· - Line 311: The “impact of cancer” is very general.
· - Line 360: Year and publisher are missing.
Author Response
On behalf of the team, I would like to thank you for your thorough review and for trying to improve the quality of our paper.
Response to comment 1:
Yes, we did change the sentence to be more clear. Thank you for this comment.
Response to comment 2:
Mainly by histopathology. At the initial inclusion criteria, we asked the tumor registry to extract the cases of biopsy-proven and confirmed MPMs. Maybe the registry has some cases that are suspected to have MPMs, but we requested the confirmed cases only.
Response to comment 3:
We added a chart showing the details of the excluded patients. 97 patients were excluded from the analysis because they had hematological malignancies (either one or both malignancies were hematological).
Response to comment 4:
Yes, our department always uses ICD-10 classification to determine anatomical location. Before 2000, we were using ICD-O-2, but after 2000 we started using ICD-O-3 for tumor staging. We added this information in the method section. Thank you for bringing this up.
Response to comment 5:
We double-checked with the tumor registry of our department (Department of Medical Oncology), the total number of oncology patients (from 1993-2022) is 25,276. This information was added in the method section, and the prevalence of MPMs has been modified accordingly. We also modified the method and included that the cases were identified by the tumor registry initially, and then they were further screened by the research members.
Response to comment 6:
The starting date was defined as the date of the first primary malignancy diagnosis (the date of the final pathology report confirming the diagnosis of 1st cancer). The end date was March 2022 (the date we started collecting data in). We reviewed all the charts using the hospital electronic system (BESTCARE) to see the notes and follow-up of patients from 2016-2022 to determine their vital status. The vital status of patients before 2016 was obtained by the research and data management of the Medical Oncology Department in King Abdullah Specialized Children Hospital, KAMC. They follow the patients who do not have regular follow-ups in KAMC with annual phone calls (after the active surveillance period ends) to determine their vital status, and so they have data for all oncology patients.
Response to comment 7:
Thank you for this comment. Noted and changed.
Response to comment 8:
We have added a chart that displays the number of patients who were excluded with the reason for exclusion. Patients with hematological malignancies (97), non-Saudi (29), lack of pathology reports, or lack of data related cancer’s diagnosis (3) were excluded. These patients were excluded based on predetermined criteria, and because of that, they should not introduce a bias.
Response to comment 9:
In the modified version of the manuscript, the prevalence of MPMs was calculated by dividing the number of MPMs cases (450) over the total sample (25276) and then multiplying by 100 (1.78%). This represents the prevalence of MPMs in general, including hematological malignancies, as it was calculated prior to applying the exclusion criteria. If you recommend deleting the sentence regarding the prevalence, please let us know.
Response to comment 10:
Thank you for suggesting this. We did not do this because it is not a part of our objective, but if you think that this is essential to make the paper publishable, we will do it. The problem is that the journal gives 10 days for resubmission of the revised version of the manuscript, and to do this, we will probably need more than 10 days as we have to request the diagnoses of all oncology patients to make a comparison.
Response to comment 11:
We changed the diagnostic interval (time between the 1st and 2nd cancers into median and percentiles).
Response to comment 12:
Thank you for noticing this. We modified it.
Response to comment 13:
We added the median and percentiles for age at the time of 1st, 2nd, and 3rd diagnoses.
Response to comment 14:
Although I don’t have a specific answer, it might be related to the fact that in the Kingdom of Saudi Arabia, colorectal cancer is the most frequent cancer in males (based on WHO statistics “2020”). Another cause would be the fact that, unlike colorectal, breast, and gynecological cancers, prostate cancer is not a part of syndromic malignancies such as familial adenomatous polyposis, Lynch syndrome, or Li-Fraumeni syndrome. It is also important to notice that our study showed a higher number of female patients and this might be a cause as well. Again I apologize for not having a specific answer for this comment.
Response to comment 15:
We changed all 3D pie charts by 1 alluvial plot and 2 bar plots. Thank you for this great suggestion, it is easier to read and understand now.
Response to comment 16:
We modified the survival graphs as requested.
Response to comment 17:
To make the manuscript shorter and easier to read, we tried to minimize the written description of the graphs and represent the information in a visual presentation (via graphs and tables). If you think that additional description is necessary, please let us know so we can expand our result section further. Thank you.
Response to comment 18:
We unified all the numbers as requested.
Response to comment 19:
In our hospital, patients aged 18 and more are considered adults. Only 4 cases were under 18, and we think that keeping them will not change the data. If you think that we should exclude these 4 cases, just let us know and we will.
Response to comment 20:
We further emphasized on this issue in the discussion. Thank you.
Response to comment 21:
We have added what you recommended, thank you.
Response to comment 22:
We tried to make the sentence more specific.
Response to comment 23:
Noted and changed. Thank you
Response to comment 24:
If you mean references 1 and 2, these are websites, and, unfortunately, we could not find the name of the publishers.
We did our best to address your suggestions and recommendations. Please let us know if there is anything else you would like us to modify/improve
Regards and have a nice day,
Moustafa S. Alhamadh
A modified version of the manuscript will be submitted tomorrow.
Reviewer 2 Report
Alluvial plot will be more useful to replace Table 3.
Fig 1A-B we can't trace the association of staging i the first with the second malignancy. So dot plot connected with line or alluvial plot will show if e.g patients with advanced stage in the first malignancy will have advanced stage for the second cancer. OR frequencies shown on a human body and presenting two figures together or show 2 frequencies on same figure.
Fig. 2A-B suggest to be replaced by bar plot of both first and second for each cancer, e.g 2 bars for CRC, 2 bars for TC, .....etc, so we can identify the difference visually. Legend color is so small, could not trace the difference.
The aim mentioned and conclusion at the end of discussion do not match the results. There was no geographic distribution in a table or in a map showing difference in different countries.
Author Response
On behalf of the team, I would like to thank you for your honest review and for trying to improve the quality of our paper.
Response to comment 1:
We did Alluvial plot as requested. Thank you for this great suggestion. It is definitely a nice plot that will improve the paper’s quality.
Response to comment 2:
Bar plot was created instead of pie chart.
Response to comment 3:
We actually agree with you. We deleted the last sentence of the conclusion. We mention this because, by observation, our results are different from the previously published results in India, Turkey, and Chaina.
We did our best to address your suggestions and recommendation. Please let us know if there is anything else you would like us to modify/improve
Regards and have a nice day,
Moustafa S. Alhamadh
The revised version of the manuscript will be submitted tomorrow.
Round 2
Reviewer 1 Report
I do not have any additional comment.
Reviewer 2 Report
Line 124: typo 'prevalence'
The colors and font used in Figures 2 and 5 could be improved.
