High-grade serous ovarian carcinoma organoids as models of chromosomal instability
Curation statements for this article:-
Curated by eLife
eLife assessment
This valuable study reveals that patient-derived organoids recapitulate similar genomic signatures as that of the parental tissue, which could be a useful model to evaluate chromosome instability, drug sensitivity, and intratumoral heterogeneity. However, whereas some of the sequencing data are compelling, the theoretical analysis is incomplete and would benefit from a more rigorous definition. With the theoretical part strengthened, the work will be of interest to medical biologists working on ovarian carcinoma.
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
- Cancer Biology (eLife)
Abstract
High-grade serous ovarian carcinoma (HGSOC) is the most genomically complex cancer, characterized by ubiquitous TP53 mutation, profound chromosomal instability, and heterogeneity. The mutational processes driving chromosomal instability in HGSOC can be distinguished by specific copy number signatures. To develop clinically relevant models of these mutational processes we derived 15 continuous HGSOC patient-derived organoids (PDOs) and characterized them using bulk transcriptomic, bulk genomic, single-cell genomic, and drug sensitivity assays. We show that HGSOC PDOs comprise communities of different clonal populations and represent models of different causes of chromosomal instability including homologous recombination deficiency, chromothripsis, tandem-duplicator phenotype, and whole genome duplication. We also show that these PDOs can be used as exploratory tools to study transcriptional effects of copy number alterations as well as compound-sensitivity tests. In summary, HGSOC PDO cultures provide validated genomic models for studies of specific mutational processes and precision therapeutics.
Article activity feed
-
-
Author Response
Reviewer #2 (Public Review):
In this manuscript, Vias and co-authors develop HGSOC PDOs and characterized their genomes, transcriptomes, drug sensitivity, and intra-tumoural heterogeneity. They show that PDOs represent the high variability in copy number genotypes observed in HGSOC patients. Drug sensitivity was reproducible compared to parental tissues and the ability of these models to grow in vivo.
Overall, the manuscript lacks sufficient novelty. Several pieces of information and a number of conclusions that are presented here have been previously published by other groups (Nina Maenhoudt, Stem cell reports, 2020; Shuang Zhang, Cancer Discov, 2021).
We agree that several important papers on HGSOC organoids have been published. However, we disagree about your assessment of “lacks sufficient novelty”. Our MS …
Author Response
Reviewer #2 (Public Review):
In this manuscript, Vias and co-authors develop HGSOC PDOs and characterized their genomes, transcriptomes, drug sensitivity, and intra-tumoural heterogeneity. They show that PDOs represent the high variability in copy number genotypes observed in HGSOC patients. Drug sensitivity was reproducible compared to parental tissues and the ability of these models to grow in vivo.
Overall, the manuscript lacks sufficient novelty. Several pieces of information and a number of conclusions that are presented here have been previously published by other groups (Nina Maenhoudt, Stem cell reports, 2020; Shuang Zhang, Cancer Discov, 2021).
We agree that several important papers on HGSOC organoids have been published. However, we disagree about your assessment of “lacks sufficient novelty”. Our MS addresses critical questions about conservation of mechanisms of chromosomal instability, how PDOs can be selected as clinical relevant models based on patterns of CIN and their comparative drug response. These questions are vital to using PDOs for therapeutic development and have not been explored before. By contrast, Maenhoudt et al. performed many analyses on several organoids (whole-genome sequencing, whole exome sequencing) but did not analyse the relationships between copy number profiles, mutational signatures or drug sensitivity between donor tissues and derived organoids and did not perform transcriptomic or scDNA analyses. A major novelty of our approach is to provide robust clinical validation of individual HGSOC PDOs by analysing how our PDOs are statistically representative of the various CN subclasses of HGSOC. Maenhoudt et al and Zhang et al classify their models only using infrequent recurrent mutations in driver genes. We do not understand how the Zhang MS overlaps with our MS as it describes the CRISPR-engineering of mouse cells to model HGSOC and investigates drivers of the mouse tumour microenvironment.
Reviewer #3 (Public Review):
- The manuscript adequately demonstrates that genomic instability is maintained in HGSOC tumourspheres. The use of 3-dimensional HGSOC models to more greatly resemble the in vivo environment has been used for more than a decade, but this is the first demonstration using a variety of genomic assessment tools to show genomic instability in the HGSOC tumoursphere model. It is clearly demonstrated that these HGSOC tumourspheres represent copy number variations similar to information in public datasets (TCGA, PAWG, BriTROC-1) and that cellular heterogeneity is present in these tumourspheres. The simple steps outlined to establish and passage tumourspheres will benefit the field to further study mechanisms of genomic instability in HGSOC.
Thank you for these positive comments.
- A weakness of the manuscript is the lack of operational definitions for what constitutes an organoid and an appropriate definition to distinguish genomic instability from chromosomal instability (a distinct type of genomic instability). Line 147 states "As PDOs consist of 100% tumour cells...", although this does not appear to have been established by any assessment. This limited characterization of the 3D model is a weakness since no data is provided on whether the tumourspheres constitute only a single cell type (as indicated on line 147) or multiple cell types (e.g., HGSOC cell, mesothelial cells) using markers beyond p53 expression. Based on this information, this model cannot be called a PDO, rather it should be referred to as a tumoursphere.
We define continuous PDO models on page 3 stating our criteria based on passage > 5 and successful reculture after thawing (previous publications have not defined whether their models are continuous or finite). As shown in our targeted-gene mutation analysis, all our PDOs contain a TP53 mutation allele fraction between 80–95%. Moreover, in our single cell DNA-Seq data we do not observe any normal copy number profiles that would indicate normal cells. This information is now included in the text for clarification. Our reasons not to use the term spheroids or tumourspheres are:
The word spheroid comes from the in vitro spheroid formation assay which was originally designed to overcome the difficulties found in functional in vivo serial transplantations. This method generates colony-forming units in suspension. Our patient-derived cells are not growing in suspension but within an extra-cellular matrix.
Spheroids are clonally expanded from a single-cell as part of the colony-forming assay; our patient-derived organoids were not clonally expanded in any way.
Organoids derived from patient-tumours have been named PDOs in multiple publications where pure tumour cellularity was stated for the PDOs [Vlachofiannis et al. Science (2018) 359, 920; Li et al. Nat. Comm.(2018) 9, 2983; Lee et al. Cell (2018)173, 515; Kopper et al. Nat Med (2019) 25, 838]. Use of other terms will cause confusion for readers and prevent important comparisons between PDO from different researchers.
- Chromosome instability (CIN) is a type of genomic instability that is broadly defined as an increased rate of chromosome gains or losses and is best identified through analysis of single cells (e.g., karyotype analysis), something that bulk whole genome sequencing cannot determine since it is a reflection of cell populations and not individual cells. While the data demonstrate genomic instability is retained in the tumourspheres, and chromosome losses or copy-number amplifications were observed using single-cell whole genome sequencing, evaluation of samples from the same patient over time was not evaluated. While there is evidence to support CIN in these samples, in agreement with other published work that has demonstrated CIN in >95% of HGSOC samples analyzed at the single-cell level, this work is not conclusive. The title of the manuscript should be modified to more accurately represent what the evidence supports.
We have discussed the ambiguity of CIN in our recent publication “A pan-cancer compendium of chromosomal instability” Drews et al Nature 2022.
“CIN has complex consequences, including loss or amplification of driver genes, focal rearrangements, extrachromosomal DNA, micronuclei formation and activation of innate immune signalling. This leads to associations with disease stage, metastasis, poor prognosis and therapeutic resistance. The causes of CIN are also diverse and include mitotic errors, replication stress, homologous recombination deficiency (HRD), telomere crisis and breakage fusion bridge cycles, among others.
Because of the diversity of these causes and consequences, CIN is generally used as an umbrella term. Measures of CIN either divide tumours into broad categories of high or low CIN, are restricted to a single aetiology such as HRD, are limited to a particular genomic feature such as whole-chromosome-arm changes, or can only be quantified in specific cancer types. As a result, there is no systematic framework to comprehensively characterize the diversity, extent and origins of CIN pan-cancer, or to define how different types of CIN within a tumour relate to clinical phenotypes. Here we present a robust analysis framework to quantitatively measure different types of CIN across cancer types.”
Many authors use CIN to include the consequences of CIN and other specifically use CIN to indicate ongoing numerical and structural change. We do not think our usage of CIN in the title and text is controversial and is consistent with previous peer reviewed publications, including our own.
- An additional weakness is missing information (e.g., Figure 1d, Supplementary Figure 3b, and Supplementary Table 4 were not included in the manuscript; the 13 anticancer compounds used to test drug sensitivity are not indicated) making an assessment of the data impossible, and assessment of some conclusions difficult.
We apologise for this misunderstanding as a typo suggested that there was a Figure 1d (it should have referred to Figure 1c) or Figure 1-Figure supplement 3B (the label of which was missing); we also apologise for the omission of Supplementary Table 4. These errors have been corrected and the list of compounds is now included in the Methods section.
-
eLife assessment
This valuable study reveals that patient-derived organoids recapitulate similar genomic signatures as that of the parental tissue, which could be a useful model to evaluate chromosome instability, drug sensitivity, and intratumoral heterogeneity. However, whereas some of the sequencing data are compelling, the theoretical analysis is incomplete and would benefit from a more rigorous definition. With the theoretical part strengthened, the work will be of interest to medical biologists working on ovarian carcinoma.
-
Reviewer #1 (Public Review):
In this manuscript, the authors have elegantly demonstrated the significance of asking fundamental questions in patient-derived models of patient-derived organoids (PDOs). This is especially relevant for studying complex cancers such as High-Grade Serous Ovarian Carcinoma (HSCOG). In addition to developing patient-derived organoids, this study has comprehensively examined transcriptomic, genomic, and single-cell data. In addition, based on this data, the authors have performed a complex drug sensitivity assay that further stratifies the PDOs into drug-sensitive and resistant categories. This approach would be central to identifying therapeutic regimens for difficult-to-treat cancers in the future.
-
Reviewer #2 (Public Review):
In this manuscript, Vias and co-authors develop HGSOC PDOs and characterized their genomes, transcriptomes, drug sensitivity, and intra-tumoural heterogeneity. They show that PDOs represent the high variability in copy number genotypes observed in HGSOC patients. Drug sensitivity was reproducible compared to parental tissues and the ability of these models to grow in vivo.
Overall, the manuscript lacks sufficient novelty. Several pieces of information and a number of conclusions that are presented here have been previously published by other groups (Nina Maenhoudt, Stem cell reports, 2020; Shuang Zhang, Cancer Discov, 2021).
-
Reviewer #3 (Public Review):
The manuscript adequately demonstrates that genomic instability is maintained in HGSOC tumourspheres. The use of 3-dimensional HGSOC models to more greatly resemble the in vivo environment has been used for more than a decade, but this is the first demonstration using a variety of genomic assessment tools to show genomic instability in the HGSOC tumoursphere model. It is clearly demonstrated that these HGSOC tumourspheres represent copy number variations similar to information in public datasets (TCGA, PAWG, BriTROC-1) and that cellular heterogeneity is present in these tumourspheres. The simple steps outlined to establish and passage tumourspheres will benefit the field to further study mechanisms of genomic instability in HGSOC.
A weakness of the manuscript is the lack of operational definitions for what …
Reviewer #3 (Public Review):
The manuscript adequately demonstrates that genomic instability is maintained in HGSOC tumourspheres. The use of 3-dimensional HGSOC models to more greatly resemble the in vivo environment has been used for more than a decade, but this is the first demonstration using a variety of genomic assessment tools to show genomic instability in the HGSOC tumoursphere model. It is clearly demonstrated that these HGSOC tumourspheres represent copy number variations similar to information in public datasets (TCGA, PAWG, BriTROC-1) and that cellular heterogeneity is present in these tumourspheres. The simple steps outlined to establish and passage tumourspheres will benefit the field to further study mechanisms of genomic instability in HGSOC.
A weakness of the manuscript is the lack of operational definitions for what constitutes an organoid and an appropriate definition to distinguish genomic instability from chromosomal instability (a distinct type of genomic instability). Line 147 states "As PDOs consist of 100% tumour cells...", although this does not appear to have been established by any assessment. This limited characterization of the 3D model is a weakness since no data is provided on whether the tumourspheres constitute only a single cell type (as indicated on line 147) or multiple cell types (e.g., HGSOC cell, mesothelial cells) using markers beyond p53 expression. Based on this information, this model cannot be called a PDO, rather it should be referred to as a tumoursphere.
Chromosome instability (CIN) is a type of genomic instability that is broadly defined as an increased rate of chromosome gains or losses and is best identified through analysis of single cells (e.g., karyotype analysis), something that bulk whole genome sequencing cannot determine since it is a reflection of cell populations and not individual cells. While the data demonstrate genomic instability is retained in the tumourspheres, and chromosome losses or copy-number amplifications were observed using single-cell whole genome sequencing, evaluation of samples from the same patient over time was not evaluated. While there is evidence to support CIN in these samples, in agreement with other published work that has demonstrated CIN in >95% of HGSOC samples analyzed at the single-cell level, this work is not conclusive. The title of the manuscript should be modified to more accurately represent what the evidence supports.
An additional weakness is missing information (e.g., Figure 1d, Supplementary Figure 3b, and Supplementary Table 4 were not included in the manuscript; the 13 anticancer compounds used to test drug sensitivity are not indicated) making an assessment of the data impossible, and assessment of some conclusions difficult.
-
