Tumor Testing and Genetic Analysis to Identify Lynch Syndrome Patients in an Italian Colorectal Cancer Cohort

Round 1

Reviewer 1 Report

The single center study took place between 2017 and 2023 in Bari, Italy.

841 CRC patients were tested for MSI and BRAF mutation to determine possible patients with LS.

70 were found, of which 30 patients were DNA tested for LS or hereditary cancer genes (40 did not undergo testing at this institute). 19 were found with pathogenic or likely pathogenic variants, 5 had VUS and 6 had no findings.

Methods and analysis processes are fully described. Different sequency strategies were used depending on the strength of suspicion of LS in probands and their families.

 Missense variants were analysed using in silico tools. Detailed genetic counselling was performed including personal and family history for the 30 patients. The clinical and pathological characterisation of patient/families is summarised and tabulated, and the distribution of gene and mutation types is displayed.

A detailed description of patients and families  summarising the average age of onset for different cancers by gene was provided. Investigation of the VUS was described noting specific variants and the patient and family histories, and reasons why the variants were uncertain. Pedigrees of 3 families with VUS are shown, and 2 pedigrees with no detected genetic variants are shown.

This is a comprehensive report on the single-center study of 841 patients that were analysed down to 30 patients using tumour characteristics, which were then followed up by clinical and pathological characterisation, genetic counselling and detailed genetic variant analysis.  The structure and writing is of high quality, and supported by detailed graphs and tables. In summary, the paper adds to the understanding of Lynch Syndrome and it's genetic underpinnings, and is of a high overall standard.

Minor suggestions:

line 402:  "Based on the NCCN guidelines, the age of onset of 402

CRC in LS patients is earlier (4th-5th decade of life) compared to the general population"    The reference citation to NCCN guidelines should be on this line?

Figure 2 would improve if it had the actual numbers, not only the percentages.

 

 

 

 

 

Author Response

Dear Prof. Lyra Yang,

we are pleased to submit the amended version of our work “Tumor testing and genetic analysis to identify Lynch syndrome patients in an Italian colorectal cancer cohort” (Manuscript ID: cancers-2635130) that we would like to have considered for publication in Cancers.

We addressed all the comments raised by the Reviewers by providing detailed explanations and/or making appropriate changes to the manuscript.

Reviewer1

The single center study took place between 2017 and 2023 in Bari, Italy.

841 CRC patients were tested for MSI and BRAF mutation to determine possible patients with LS.

70 were found, of which 30 patients were DNA tested for LS or hereditary cancer genes (40 did not undergo testing at this institute). 19 were found with pathogenic or likely pathogenic variants, 5 had VUS and 6 had no findings.

Methods and analysis processes are fully described. Different sequency strategies were used depending on the strength of suspicion of LS in probands and their families.

 Missense variants were analysed using in silico tools. Detailed genetic counselling was performed including personal and family history for the 30 patients. The clinical and pathological characterisation of patient/families is summarised and tabulated, and the distribution of gene and mutation types is displayed.

A detailed description of patients and families  summarising the average age of onset for different cancers by gene was provided. Investigation of the VUS was described noting specific variants and the patient and family histories, and reasons why the variants were uncertain. Pedigrees of 3 families with VUS are shown, and 2 pedigrees with no detected genetic variants are shown.

This is a comprehensive report on the single-center study of 841 patients that were analysed down to 30 patients using tumour characteristics, which were then followed up by clinical and pathological characterisation, genetic counselling and detailed genetic variant analysis.  The structure and writing is of high quality, and supported by detailed graphs and tables. In summary, the paper adds to the understanding of Lynch Syndrome and it's genetic underpinnings, and is of a high overall standard.

Minor suggestions:

line 402:  "Based on the NCCN guidelines, the age of onset of 402

CRC in LS patients is earlier (4th-5th decade of life) compared to the general population"    The reference citation to NCCN guidelines should be on this line?

Figure 2 would improve if it had the actual numbers, not only the percentages.

 We thank the reviewer for this suggestion. According to the reviewer's suggestion, we have included the reference citation and we reported the absolute numbers in addition to the percentages in Figure 2.

  

 

Author Response File: Author Response.pdf

Reviewer 2 Report

The manuscript by Antonino Pantaleo and co-author focused on a workflow implementation to identify patients (families) with a Lynch Syndrome condition, on a consecutive cohort of CRC patients attended at a Medical Center in Italy. They studied well-stablished CRC molecular markers and run a genetic test analysis using two approaches: gene specific directed sequencing or by a gene panel; looking to achieve at the end the goal of stablishing the genetic predisposing cause and to implement a genetic counselling.

 

First, I wanted to congratulate authors for their work, the number of patients in the studied cohort is significant and so to accomplish this work, it was required a well stablished and coordinated work between many sections of the health system. So, congrat on that.

 

Regarding the manuscript, although all the clinical and genetic data findings for such a relevant number of families are really interested, I suggest that the whole text need to be improved in regards of the way such data are presented. Many times, along the manuscript it´s difficult to understand what data exactly author want to detail in each part. Same happens with tables and figures, I think their presentation can be improved to make lecture easier.

For example,

1, Figure 1. Looking at the figure, it´s does not clear what the “40 no data” means; and when we go to the figure legend, it become even more confusing “Patients with MSI-H and BRAFV600 wild-type CRC were genetically tested for germline variants in genes associated with major hereditary cancer predisposition syndromes”. This sentence denote that all 70 CRC were genetically tested, while only 30 were studied. Needs to be changed.

2, Figure 2. First, the order of the figures must be changed: on the right the figure denoting the identified mutation-carrying genes (the authors must also include all the identified genes and not only MMR); and on the left show the figure with the data of the type of variant identified.

3, Table 1, in the “” you should denote names for all the identified mutation carrier genes and not only MMR and EPCAM; for example, modify “Other genes” by Other genes: ATM, NBN, APC and BMPR1A. Also, here in Table 1 and main text, change “Wild-Type” by, for example, “Negative result” (clarifying this in legend as: no identified variant in studied genes or something like that).

4, Table 4. Authors should change the Title. "...CRC not carrying the mismatch repair (MMR) gene variant." The title is not correct, here the authors show data from the six families with negative results, both those studied only for MMR genes (FAM 26-29-30) and those studied by the panel of 25 genes (FAM-25-27 - 28). So should the title just say negative cases or something, and keep the details in the legend?

In fact, the title of section 3.4, which refers to Table 4, clearly says “Patients with negative results…”.

5, The results section should be reorganized to make it easier to read, as in many places the authors move from data directly related to a specific family to describing data related to the entire cohort studied. For example,

-        Section 3.2: “Patients with germline genetic variants in MMR genes”

Lines 221 to 231. Authors describes results for number of patients carrying a MMR variant (only class5+4, Pathogenic + Likely Path).

Line 231 to 237. Authors here describes genetic finding and clinical characteristics specifically related to family 18.

Line 237 till end of section. Author went back to describe MMR mutation carriers for the entire series of genetic tested cases.

May be, move text related to Family 18 to the end of section?

6, Please check and rewrite section for the PMS2 VUS (c.184G>A) variant identified in patient from FAM-21; both in Table 3 as well as discussion.

Table 3 shows, Clinvar assertion: VUS; which is right.

Here in the link you can find further details for this variant (which you should add to the text): https://clinvarminer.genetics.utah.edu/submissions-by-variant/NM_000535.7%28PMS2%29%3Ac.184G%3EA%20%28p.Gly62Arg%29

There are 2 reported cases, 1 is an HNPCC or LS case, and the other is a case with Hereditary cancer predisposing syndrome.

However, in discussion Line433: “To date, this variant has not been reported in the literature in patients with LS and/or other hereditary tumor predisposition syndromes”.

Although this is technically correct (the variant is not reported in any article to date), authors should mention report from clinvar.

.

Author Response

Dear Prof. Lyra Yang,

we are pleased to submit the amended version of our work “Tumor testing and genetic analysis to identify Lynch syndrome patients in an Italian colorectal cancer cohort” (Manuscript ID: cancers-2635130) that we would like to have considered for publication in Cancers.

We addressed all the comments raised by the Reviewers by providing detailed explanations and/or making appropriate changes to the manuscript.

Response to Reviewer 2:

Comments and Suggestions for Authors

The manuscript by Antonino Pantaleo and co-author focused on a workflow implementation to identify patients (families) with a Lynch Syndrome condition, on a consecutive cohort of CRC patients attended at a Medical Center in Italy. They studied well-stablished CRC molecular markers and run a genetic test analysis using two approaches: gene specific directed sequencing or by a gene panel; looking to achieve at the end the goal of stablishing the genetic predisposing cause and to implement a genetic counselling.

 First, I wanted to congratulate authors for their work, the number of patients in the studied cohort is significant and so to accomplish this work, it was required a well stablished and coordinated work between many sections of the health system. So, congrat on that.

 Regarding the manuscript, although all the clinical and genetic data findings for such a relevant number of families are really interested, I suggest that the whole text need to be improved in regards of the way such data are presented. Many times, along the manuscript it´s difficult to understand what data exactly author want to detail in each part. Same happens with tables and figures, I think their presentation can be improved to make lecture easier.

 

 

For example,

1, Figure 1. Looking at the figure, it´s does not clear what the “40 no data” means; and when we go to the figure legend, it become even more confusing “Patients with MSI-H and BRAFV600 wild-type CRC were genetically tested for germline variants in genes associated with major hereditary cancer predisposition syndromes”. This sentence denote that all 70 CRC were genetically tested, while only 30 were studied. Needs to be changed.

We thank the reviewer for this suggestion. In this amended version, we modified the Figure 1 in order to better illustrate the Flowchart describing the study design. More specifically, we changed the footnote within the hexagons as follow: “40 no data” has been changed with “40 no genetic counseling” and “30 genetic testing” has been changed with “30 genetic counseling and testing”. Accordingly, we modified the Figure 1 legend as follow: “Flow chart describing the study design. Colorectal (CRC) specimens (n=841) collected from January 2017 to July 2023 were analyzed for their microsatellite instability (MSI) status. Next, MSI-high (MSI-H) specimens (n=100) were analyzed for BRAF^V600 status. Of these, 70/100 CRC specimens resulted WT for MSI-H and BRAF^V600. Patients with MSI-H and BRAF^V600 wild-type CRC (n=30) were genetically tested for germline variants in genes associated with major hereditary cancer predisposition syndromes. The remaining patients with MSI-H and BRAF^V600 wild-type CRC ( n=40) did not undergo genetic counseling.” 

2, Figure 2. First, the order of the figures must be changed: on the right the figure denoting the identified mutation-carrying genes (the authors must also include all the identified genes and not only MMR); and on the left show the figure with the data of the type of variant identified.

We thank the reviewer for this suggestion. In this amended version, we modified the Figure 2 in order to better illustrate the percentage distribution of gene variants and the type of variant identified in patients with MMR loss and BRAF^V600 wild-type CRC. Specifically, we modified the pie charts to better display the contribution of each type of genetic variant (pathogenic variants (PVs), likely pathogenic variants (LPVs), Variant of unknown significance (VUS) in MMR genes, VUS in other genes (BMPR1A, NBN, ATM, APC) identified in patients with MMR loss and BRAF^V600 wild-type CRC.

3, Table 1, in the “” you should denote names for all the identified mutation carrier genes and not only MMR and EPCAM; for example, modify “Other genes” by Other genes: ATM, NBN, APC and BMPR1A. Also, here in Table 1 and main text, change “Wild-Type” by, for example, “Negative result” (clarifying this in legend as: no identified variant in studied genes or something like that).

We thank the reviewer for this suggestion. In this amended version, we revised the Table 1 accordingly to reviewer’s suggestions. Specifically, we replaced “Other genes” with “Other genes: APC, ATM, BMPR1A, NBN”. Moreover, we replaced “Wild-type” with “Negative Result” and we added the following sentence “Negative Results: No LPV/PV identified in the genes analyzed.” in the corresponding Table’s legend

 

 

4, Table 4. Authors should change the Title. "...CRC not carrying the mismatch repair (MMR) gene variant." The title is not correct, here the authors show data from the six families with negative results, both those studied only for MMR genes (FAM 26-29-30) and those studied by the panel of 25 genes (FAM-25-27 - 28). So should the title just say negative cases or something, and keep the details in the legend?

In fact, the title of section 3.4, which refers to Table 4, clearly says “Patients with negative results…”.

We thank the reviewer for this helpful suggestion. In this amended version, we changed the title of the indicated table by replacing “Clinical features of patients with high microsatellite instability (MSI-H) and BRAF^V600 wild-type CRC.” with “Clinical features of patients with negative results on genetic testing.” Moreover, we moved the following detail “with high microsatellite instability (MSI-H) and BRAF^V600 wild-type CRC.” from the title to the table’s legend. Moreover, as suggested by reviewer#3 in this amended version of the manuscript we shifted the Table 4 in Supplementary Data as Table S4. According to reviewers’s suggestions, in this amended version of the manuscript all Tables and Supplementary Table were renamed in the text according to the order in which they were inserted.

 

 

5, The results section should be reorganized to make it easier to read, as in many places the authors move from data directly related to a specific family to describing data related to the entire cohort studied. For example,

-        Section 3.2: “Patients with germline genetic variants in MMR genes”

Lines 221 to 231. Authors describes results for number of patients carrying a MMR variant (only class5+4, Pathogenic + Likely Path).

Line 231 to 237. Authors here describes genetic finding and clinical characteristics specifically related to family 18.

Line 237 till end of section. Author went back to describe MMR mutation carriers for the entire series of genetic tested cases.

May be, move text related to Family 18 to the end of section?

We thanks the reviewer for this suggestion. In the amended version of this manuscript we reorganized the results section. Specifically, we moved the text related to Family 18 to the end of the paragraph “3.2 Patients with germline genetic variants in MMR genes”. Moreover, we moved the Table 2 and Figure 3 in order to insert them into the main text close to their first citation according to Cancers Journal format.

6, Please check and rewrite section for the PMS2 VUS (c.184G>A) variant identified in patient from FAM-21; both in Table 3 as well as discussion.

Table 3 shows, Clinvar assertion: VUS; which is right.

Here in the link you can find further details for this variant (which you should add to the text): https://clinvarminer.genetics.utah.edu/submissions-by-variant/NM_000535.7%28PMS2%29%3Ac.184G%3EA%20%28p.Gly62Arg%29

There are 2 reported cases, 1 is an HNPCC or LS case, and the other is a case with Hereditary cancer predisposing syndrome.

However, in discussion Line433: “To date, this variant has not been reported in the literature in patients with LS and/or other hereditary tumor predisposition syndromes”.

Although this is technically correct (the variant is not reported in any article to date), authors should mention report from clinvar.

We thanks the reviewer for this suggestion. In the amended version of this manuscript, we added the findings reported in ClinVar (https://clinvarminer.genetics.utah.edu/submissions-by-variant/NM_000535.7%28PMS2%29%3Ac.184G%3EA%20%28p.Gly62Arg%29). Specifically we added the following phrase “Nevertheless, in the ClinVar database, two different patients harboring the PMS2 (c.184G>A; p.Gly62Ser) VUS have been reported in association to LS and other hereditary tumor predisposition syndromes.” in the Discussion section.

Author Response File: Author Response.pdf

Reviewer 3 Report

In this manuscript, the authors study colorectal cancer patients by the assessment of MSI status, clinical evaluation of patients and their families, and genetic analysis to identify mutations in MMR genes. This is a well described study which could be of great interest for patients with this type of cancer. Lynch syndrome appears in small number of CRC patients which makes the reports about this and similar types of CRCs very important for patient management for the families with LS and other hereditary cancers.

I have some minor concerns about this study that are numbered below:

1.       Could the authors describe the patient inclusion and exclusion criteria more specifically. The way it is described now it appears they included all the patients with colorectal cancer without any specific exclusion criteria. If that is so, please clarify.

2.       When describing the recruited patient group, could the authors describe the patients with more details like cancer stage, histological diagnosis, percentage of male and female patients, age range/average, etc. It might be convenient to present it as a table.

3.       The authors describe several different mutations in their patient group. Did the authors register those mutations in any database like ClinVar? It is very important to publish new suspected variants and to confirm the already published ones , especially when the aim of the manuscript is to help patient management.

Author Response

Dear Prof. Lyra Yang,

we are pleased to submit the amended version of our work “Tumor testing and genetic analysis to identify Lynch syndrome patients in an Italian colorectal cancer cohort” (Manuscript ID: cancers-2635130) that we would like to have considered for publication in Cancers.

We addressed all the comments raised by the Reviewers by providing detailed explanations and/or making appropriate changes to the manuscript.

Response to Reviewer 3:

Open Review

In this manuscript, the authors study colorectal cancer patients by the assessment of MSI status, clinical evaluation of patients and their families, and genetic analysis to identify mutations in MMR genes. This is a well described study which could be of great interest for patients with this type of cancer. Lynch syndrome appears in small number of CRC patients which makes the reports about this and similar types of CRCs very important for patient management for the families with LS and other hereditary cancers.

I have some minor concerns about this study that are numbered below:

1. Could the authors describe the patient inclusion and exclusion criteria more specifically. The way it is described now it appears they included all the patients with colorectal cancer without any specific exclusion criteria. If that is so, please clarify.

 

We thank the reviewer for this suggestion. In this amended version, we added the following phrase “All 841 patients with CRC were included in this study without inclusion and exclusion criteria.” in the paragraph “2.1 Patient recruitment”.

 

2. When describing the recruited patient group, could the authors describe the patients with more details like cancer stage, histological diagnosis, percentage of male and female patients, age range/average, etc. It might be convenient to present it as a table.

 

As suggested by the Reviewer, in this amended version of the manuscript we added the Table S1 which contains the clinicopathological features (Sex, Age at CRC onset, CRC Stage, and histological diagnosis) of 30 CRC patients recruited for genetic testing. In addition, in this amended version of the manuscript all Tables and Supplementary Table were renamed in the text according to the order in which they were inserted.

 

3. The authors describe several different mutations in their patient group. Did the authors register those mutations in any database like ClinVar? It is very important to publish new suspected variants and to confirm the already published ones , especially when the aim of the manuscript is to help patient management.

 

We thank the reviewer for this helpful suggestion. We have deposited the genetic variant identified in patients of our study in the “InSight database”, which is  the most comprehensive database of DNA variants in the genes that contribute to gastrointestinal cancer.

Author Response File: Author Response.pdf

Reviewer 4 Report

This paper describes an experience of Lynch syndrome testing. To my opinion, the value of this manuscript is the comparison of clinical features of the patients with and without germline mutations. Is there an age threshold, which allows excluding MSI-positive patients from germline DNA analysis? What is the value of family history: were there some patients, who were old and did not have cancer-affected relatives, but nevertheless turned out to have Lynch syndrome? I suggest to consider only pathogenic (and perhaps, likely pathogenic) alleles, while variants of unknown significance would compromise the suggested above analysis. Please comment on differences in the disease presentation between MSH6/PMS2 vs. MLH1/MSH2 mutations carriers. I also suggest to critically shorten the paper: some tables and figures can be moved to the Supplement. It will be helpful to list all 19 pathogenic alleles identified in the study.     

Author Response

Dear Prof. Lyra Yang,

we are pleased to submit the amended version of our work “Tumor testing and genetic analysis to identify Lynch syndrome patients in an Italian colorectal cancer cohort” (Manuscript ID: cancers-2635130) that we would like to have considered for publication in Cancers.

We addressed all the comments raised by the Reviewers by providing detailed explanations and/or making appropriate changes to the manuscript.

Response to Reviewer 4:

This paper describes an experience of Lynch syndrome testing. To my opinion, the value of this manuscript is the comparison of clinical features of the patients with and without germline mutations.

Is there an age threshold, which allows excluding MSI-positive patients from germline DNA analysis?

We thank the reviewer for this interesting observation. Lynch syndrome program reduces cancer morbidity and mortality in mutation carriers. As described in the main text, the universal LS program refers to routine tumor testing for MSI in combination with the BRAF test among all patients with CRC. Patients with absent MMR protein expression or MSI-H in addition to wild-type BRAF^V600 are referred for genetic testing. In our cohort of patients a total of probands with 30 CRC MSI-H BRAF^V600 cases underwent genetic testing. Among these, 19 out of 30 analyzed patients were found to harbor a germline mutation in one of MMR gene. Considering the probands with a molecular diagnosis of LS in addition with their mutation carrier relatives with cancer (n=13), 28 subjects were diagnosed with CRC with an average age of onset of 47,5 years (range 29-69). Of these, 19 subjects carrying the MLH1 (n=10), MSH2 (n=7), MSH6 (n=1), or PMS2 (n=1) gene disease causing alteration developed the CRC before the age of 50 years (range 29-50), while 9 subjects carrying MSH2 (n=3), MSH6 (n=5), or PMS2 (n=1) developed CRC after the age of 50 years (range 52-69). Among 30 patients genetically analyzed, five patients were found to harbor a variant of unknown significance (VUS) in MSH6 (n=1), PMS2 (n=1), NBN (n=1), ATM (n=1), APC and BMPR1A (n=1). These five carriers of VUS developed CRC with an average age of 66,8 years (range 59-78). The remaining six patients, resulted negative on genetic testing, developed CRC with an average age of 52,6 years (range 47-63). Although the average age of developing colon cancer in patients with a mutation in one of the MMR genes is lower than in patients with a VUS or negative genetic test, it is not possible delineate a specific age-threshold to select or exclude patients with CRC MSI-H BRAF^V600 for genetic testing. In fact, in our study, some patients who carried the mutation developed cancer after age 50, while others who did not carry the mutation developed cancer before age 50. Studies including a larger number of patients are needed in order to establish a more accurate LS screening program combining MSI analysis, evaluation of  family history and age threshold of CRC onset.

 

What is the value of family history: were there some patients, who were old and did not have cancer-affected relatives, but nevertheless turned out to have Lynch syndrome? I suggest to consider only pathogenic (and perhaps, likely pathogenic) alleles, while variants of unknown significance would compromise the suggested above analysis.

We thank the reviewer for this observation. In our cohort of CRC patients, the genetic testing of 30 LS suspected patients, with MSI-H and wild-type for BRAF^V600 CRC, identified 19 index LS cases with LPVs/PVs in MMR genes. According to the reviewer's suggestion and considering the 19 LS index cases having PV/LPVs in MMR genes, eight patients developed colorectal cancer after the age of 50 years. Although some of these patients not had a strong family history of cancer potentially correlated to LS, all of these patients developed an extra-colonic LS related tumor before the age of 50. Among these, only two suspected LS patients were analyzed through the NGS multigene panel since had a family history of cancer potentially correlated to other major hereditary tumor predisposition syndromes. In these patients the NGS analysis excluded the presence of LPVs/PVs in genes correlated to other genes correlated to other tumor syndromes and confirmed the molecular diagnosis of LS.

Please comment on differences in the disease presentation between MSH6/PMS2 vs. MLH1/MSH2 mutations carriers.

We thank the reviewer for this important suggestion. In this amended version, we added information about cancer risks associated with constitutional defects of MMR genes. More specifically the following sentences “Furthermore, it is now well established that the cumulative risk of developing cancer varies among LS patients based on specific genetic alterations in MMR genes (Kastrinos F. et al, Gastroenterology. 2021 Aug;161(2):453-462.e15. doi:10.1053/j.gastro.2021.04.010.). More specifically, germline LPVs/PVs in MLH1 and MSH2 genes are associated with a higher lifetime CRC risk of 58%-82%, while LPVs/PVs in MSH6 and PMS2 genes are associated with a lower lifetime CRC risk of 10-22% (Kastrinos F. et al, Gastroenterology. 2021 Aug;161(2):453-462.e15. doi:10.1053/j.gastro.2021.04.010.). Much of the clinical heterogeneity observed in LS patients may be explained by the molecular profiles of individual carcinomas, which partly are dependent on which MMR gene is affected (Valle L. Gastroenterology. 2023 Jul;165(1):20-23. doi: 10.1053/j.gastro.2023.04.027).” has been added in the discussion section.

 I also suggest to critically shorten the paper: some tables and figures can be moved to the Supplement.

We thank the reviewer for this suggestion. According to the reviewer's suggestion, in this amended version we moved the Table 3 and the Table 4 in Supplementary Data. Specifically, Table 3 has moved in Table S3 and Table 4 has moved in Table S4. Moreover, in this amended version of the manuscript all Tables and Supplementary Table were renamed in the text according to the order in which they were inserted.

It will be helpful to list all 19 pathogenic alleles identified in the study.    

We thank the reviewer for this suggestion. Accordingly, in this amended version we listed all 19 pathogenic alleles together with VUS in MMR and other genes in paragraph “3.1 Genetic testing of CRC patients for LS”.

Author Response File: Author Response.pdf

Round 2

Reviewer 4 Report

The comments were addressed in a proper way