Skip to main content

Table 3 Summary of included articles in the systematic review

From: Evolutionary perspectives, heterogeneity and ovarian cancer: a complicated tale from past to present

First author

Year

Cohort

Method

Conclusion

Ref

Taylor HC

1959

550 samples from 36 patients and 12 temporal paired samples

Light microscopy

Consistent morphology across tumor areas and at relapse

[25]

Knoerr-Gaertner H

1977

4 cases of bilateral cystadenocarcinomas

Chromosomal banding

Identical chromosomal markers found in bilateral cystadenocarcinomas without definite proof for clonal or multicentric evolution

[26]

Roberts CG

1990

3 samples from 1 case

Chromosomal banding and inferred clonal cytogenetic evolution

First report of clonal evolution based on chromosomal breakpoints analysis

[28]

Pejovic T

1991

34 samples from 15 cases

Chromosomal banding

No evolution seen in spatially separated samples. Possible explanations given: metastatic spread is a late event in evolution, identical evolution in paired samples, clonal dominance of cells with growth advantage

[27]

Vandamme B

1992

17 samples from 6 cases and 4 temporal samples from 1 case

Restriction fragment length polymorphism

11p deletion is a late event in disease progression

[30]

Abeln ECA

1994

3 cases

Flow cytometry and PCR on microsatellite markers

Proof of principle PCR-based LOH detection on flow sorted tumor cells can determine their monoclonal origin and can be used for the analysis of tumor heterogeneity

[32]

Deger RB

1997

21 samples from 8 cases

Chromosomal banding

50% of cases had less complex chromosomal aberrations in metastatic lesions

[34]

Ioakim-Liossi A

1999

9 patients with paired temporal samples

Chromosomal banding

Identical chromosomal aberrations after disease progression or treatment administration

[29]

Zborovskaya I

1999

18 samples from 9 cases

PCR on microsatellite markers

In 5 out of 9 cases the LOH patterns were different in the primary tumor and the metastatic nodes

[33]

Fishman A

2005

14 samples from 7 cases

Chromosomal comparative genomic hybridization

6 out of 7 cases presented less chromosomal aberrations or normal genomes in metastatic sites than in paired primary lesions

[35]

Khalique L

2007

110 samples from 16 cases

PCR on microsatellite markers, multiplex SNP analysis, maximum parsimony tree analysis

All cases are monoclonal in origin and display extensive intra-tumor heterogeneity

[41]

Khalique L

2009

186 samples from 22 cases

PCR on microsatellite markers, maximum parsimony tree analysis

Monoclonal origin but become polyclonal with different clones gaining metastatic potential during early and late stages of evolution. There is extensive heterogeneity between metastatic samples with no significant differences from primary tumor

[42]

Cooke SL

2010

3 platinum sensitive and 4 platinum resistant cell lines derived from 3 patients, 6 paired pre/post neoadjuvant chemotherapy samples

Array CGH, M-FISH, Array SNP

Platinum resistance develops from pre-existing minor clones but no enrichment is observed after short-term chemotherapy

[36]

Micci F

2010

70 samples from 32 cases

Chromosomal banding, HR-CGH

Metastatic spread to the contralateral ovary is a late event in evolution

[31]

Ng CK

2012

2 platinum sensitive and 2 platinum resistant cell lines from 2 patients

NGS

Tandem duplicator mutator phenotype is an ongoing phenotype that arose early before divergence of platinum sensible/resistant lineages and might represent a novel DNA repair defect independent of HR

[37]

Bashashati A

2013

29 samples from 5 cases and 2 paired temporal samples from 1 case

NGS, SNP array, RNA array

Extensive heterogeneity at all levels with early divergence and highly individual evolutionary trajectories. TP53 is the only consistent mutation across samples. Cell free DNA offers a narrow and heterogeneous picture of common ancestral mutations. Stable clonal population at relapse might be a feature of long survivorship

[43]

Castellarin M

2013

6 paired temporal samples from 2 cases

NGS

89% of mutations found in relapsed tumors are present in primary tumors. Predominant TP53 mutations in all samples

[38]

Tone AA

2014

19 primary and 18 recurrent samples from 11 cases

Cancer hotspot NGS

Almost half of patients with low grade SOC had RAS/RAF pathway mutations. Among them, 20% exhibit spatial and temporal heterogeneity

[48]

Lee JY

2015

11 samples from 1 case

NGS, SNP array

Early divergence of primary clusters, metastatic cluster accumulated few mutations. Only 6% shared somatic mutations across all samples except TP53

[44]

Mota A

2015

6 primary and 6 recurrent samples from 1 case

NGS, CGH

Significant mutational divergence, with 41% shared somatic variants between 1 primary and 2 recurrent samples

[39]

Schwarz RF

2015

135 samples from 14 cases

SNP array, NGS, Digital PCR, MEDICC algorithm

Potentially unknown mutator phenotypes lead to non-clock-like evolutionary trajectories. There are only minor genomic changes (in comparison with the overall changes) with neoadjuvant chemotherapy, with an average of 46 new genomic events. Patients with high clonal expansion have a shorter survival. Marked early divergent evolution of relapsed samples

[49]

Eckert MA

2016

32 samples from 8 patients

NGS

Various anatomic sites have similar mutation burden indicative of early establishment of genomic instability. STIC can represent primary or metastatic lesions

[45]

Lambrechts S

2016

31 cases with paired primary and recurrent samples

SNP array, NGS

Huge variability between primary and recurrent tumors with 58% shared mutations between primary and recurrent samples. TP53 was the most frequently observed (18 out of 23 pairs, 78%) with few other recurrent mutations

[40]

McPherson A

2016

68 samples from 7 cases

NGS, Single cell sequencing

Metastatic sites usually comprise oligoclonal populations with at least one polyclonal site in each patient. Two distinct modes of peritoneal spreading: monoclonal, unidirectional and polyclonal with reseeding

[50]

Choi YJ

2017

25 samples and ascitic cells from 4 cases

NGS, microarray CGH, methylation profiling

Ascitic cancer cells diverge early through a polyseeding mechanism and accurately reflect common somatic mutations, CNAs and DNA methylations in advanced ovarian cancer

[53]

Yin X

2017

22 bilateral samples from 11 cases

NGS

Bilateral ovarian cancer lesions have a common ancestor but diverge early and display significant heterogeneity

[46]

Arend RC

2018

Paired pre/post NACT samples from 19 patients, 14 paired plasma samples

NGS 50 genes, Gene expression panel

Targeted sequencing of cell free DNA was not informative for NACT tumor induced changes

[52]

Chien J

2018

9 samples for each of the 4 patients

RNA seq

Peritoneal metastases arise early and are composed of multiple subclones, with shared SNV at all regions of one patient 2.13–25%

[47]

Zhang AW

2018

212 samples from 38 cases

NGS, Gene expression, IHC

Within patient spatial immune variation influences the dissemination of malignant clonal populations

[51]