Khatami & Tavangar (2018) — Circulating Tumor DNA (ctDNA) in the Era of Personalized Cancer Therapy

Bibliographic Reference

Khatami, F., & Tavangar, S. M. (2018). Circulating tumor DNA (ctDNA) in the era of personalized cancer therapy. Journal of Diabetes & Metabolic Disorders, 17, 19–30. https://doi.org/10.1007/s40200-018-0334-x

Core Argument

Tumor heterogeneity is the single greatest obstacle to successful personalized cancer medicine, and ctDNA — as a non-invasive, repeatable liquid biopsy analyte — provides a practical solution. The authors argue that because ctDNA is shed into the bloodstream from both primary and metastatic tumor sites, it captures a systemic “screenshot” of a patient’s cancer (as opposed to the localized “snapshot” of a single tissue biopsy), enabling real-time tracking of intratumor heterogeneity, clonal evolution, and treatment response. This capability is positioned as essential for realizing P4 medicine (predictive, preventive, personalized, and participatory) in oncology.

The review surveys ctDNA applications across four major cancer types — breast, lung, gastrointestinal, and thyroid — showing how ctDNA mutation tracking and quantification can guide targeted therapy selection, detect emergent resistance mutations earlier than conventional imaging, monitor minimal residual disease, and inform adaptive treatment strategies. The central thesis is that integrating ctDNA liquid biopsy with tissue biopsy creates a comprehensive genetic portrait of each patient’s tumor that outperforms either modality alone.

Methods

Narrative review based on a systematic search of PubMed, Scopus, Web of Science, and EMBASE from 1990 to 2017. The search syntax combined terms for cancer (“cancer” OR “neoplasm” OR “tumor”) with cell-free nucleic acid terms (“cfDNA” OR “circulating tumor DNA” OR “ctDNA” OR “cell free DNA” OR “CTC”) and precision medicine terms (“personalized medicine” OR “Precision medicine” OR “P4 medicine” OR “treatments” OR “therapy” OR “Diagnosis”). Inclusion was restricted to English-language articles, yielding 19 final selected articles. The review synthesizes these with additional cited literature across ctDNA biology, detection technologies (ddPCR, NGS, CAPP-Seq), and clinical applications by cancer type. No meta-analysis or quantitative synthesis was performed.

Key Findings

1. “ctDNA is a ‘screenshot’ of the primary and metastatic tumor” — in contrast to tissue biopsy as a “snapshot,” ctDNA captures heterogeneity across multiple tumor sites simultaneously. The authors state: “ctDNA is a repeatable non-invasive biopsy method and contrary to tissue biopsy as a ‘snapshot,’ ctDNA is a ‘screenshot’ of the primary and metastatic tumor.” This systemic view is critical for tracking multifocal clonal evolution, as demonstrated by Murtaza et al. (2015) showing that ctDNA levels from plasma samples “imitate the clonal hierarchy concluded from sequencing of tumor biopsies” in metastatic breast cancer.

2. “ctDNA can be a potential source of tumor DNA alteration pursuing for the documentation of tumor-associated genetic changes in order to real-time tumor monitoring.” ctDNA fragment length (~166 bp, corresponding to nucleosome-protected DNA) enables size-based discrimination from non-tumor cfDNA. Technologies including digital droplet PCR (ddPCR), next-generation sequencing (NGS), and Cancer Personalized Profiling by deep Sequencing (CAPP-Seq) — which “is able to identify one molecule of mutant DNA in ten thousands molecules of normal DNA” — provide the sensitivity needed for clinical ctDNA analysis.

3. “Peripheral ctDNA is a small proportion of DNA in bloodstream, fewer than 100 ng/mL and no more than a fraction of this whole cfDNA (< 1% of total cfDNA) is in certainty tumor-derived.” Despite this low fraction, ctDNA levels correlate with tumor stage and metastatic burden — “the degree of metastatic tumor is joined to the volume of ctDNA” — and the direct presence of metastasis to liver or bone has been “straight attached to the greater levels of ctDNA.” This concentration dynamic makes ctDNA both a challenge (detection sensitivity limits in early-stage disease) and an opportunity (quantitative monitoring of disease progression and treatment response).

4. “Even a minor genetic clone within the tumor if carries a drug-resistant mutation can be developed after treatment” — ctDNA monitoring can identify such resistant subclones before they become clinically dominant. In breast cancer, ESR1 mutations emerging during aromatase inhibitor therapy are “strongly recognized with ctDNA analysis” and predict response to subsequent hormone therapy. In NSCLC, ctDNA profiling via CAPP-Seq after Rociletinib treatment “highlighted frequent intra-patient heterogeneity” and showed that MET proto-oncogene copy number increases “involve in resistance recurrently.” In GIST, mutation detection in cfDNA of patients with metastatic disease can be recruited “for personalized usage of imatinib and monitoring of early treatment adaptations.”

5. “Mutation tracking of ctDNA through sequencing could outline the genetic events of MRD in order to projected the genetic background of the subsequent metastatic relapse extra precisely than sequencing of the primary tumor.” ctDNA enables minimal residual disease monitoring that can detect relapse months earlier than conventional follow-up. In colorectal cancer, Reinert et al. demonstrated that “relapses could be detected months in advance compared to conventional follow-up.” In breast cancer, ctDNA-based MRD monitoring following adjuvant therapeutic interventions “possibly will be personalized with the genetic profile existing in the MRD, a therapeutic approach that could solve the problem of intra-tumor genetic heterogeneity.”

Concepts Introduced or Used

• ctDNA (circulating tumor DNA) — Short tumor-derived DNA fragments (~166 bp) freely circulating in serum and plasma, released via apoptosis, necrosis, and active secretion. Provides a real-time molecular portrait of the tumor genome. See ctdna.

• Liquid biopsy — Non-invasive blood-based sampling that captures tumor-derived material (ctDNA, CTCs, exosomes) without requiring tissue biopsy. Enables serial monitoring. See liquid-biopsy.

• Intratumor heterogeneity — Genetic and epigenetic diversity among cancer cells within a single tumor, arising from clonal evolution. The paper identifies this as the central obstacle to personalized medicine. See intratumor-heterogeneity.

• CAPP-Seq (Cancer Personalized Profiling by deep Sequencing) — Ultrasensitive NGS method for ctDNA quantification capable of detecting one mutant molecule in 10,000 normal DNA molecules. Applicable across cancer types. See capp-seq.

• Minimal residual disease (MRD) — Residual tumor cells below imaging detection threshold after treatment. ctDNA monitoring can track MRD and predict relapse earlier than conventional methods. See minimal-residual-disease.

• P4 medicine — Predictive, preventive, personalized, and participatory approach to cancer management. The paper positions ctDNA as a key enabling technology for P4 medicine.

• cfDNA (cell-free DNA) — Total DNA fragments circulating in blood, predominantly from hematopoietic cell turnover. ctDNA is the tumor-derived subset. See cell-free-dna.

• Nucleosome positioning — The wrapping pattern of DNA around histone octamers (~147 bp per nucleosome). ctDNA fragment length reflects nucleosome spacing, and ctDNA deep sequencing “have the path of transcription factors,” enabling inference of cell type of origin.

Entities Referenced

• EGFR — Epidermal growth factor receptor; EGFR mutation status (particularly L858R point mutation in exon 21 and deletion mutation in exon 19) is a predictive biomarker for EGFR-TKI therapy in NSCLC. The FDA-approved EGFR Mutation Test v2 (2016) is a blood-based companion diagnostic for Erlotinib. See egfr.

• ESR1 — Estrogen receptor alpha; D538G mutation detected in ctDNA predicts response to endocrine therapies and resistance to aromatase inhibitors in breast cancer. ESR1 mutations are “infrequently developed during adjuvant AI therapy, but are frequently designated by therapy for metastatic disease.” See esr1.

• PIK3CA / PIK3C — Phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha; common genomic alterations in ER-positive breast cancer. ctDNA PIK3CA status predicts efficacy of Buparlisib plus Fulvestrant in the phase III BELLE-2 trial. See pik3ca.

• KRAS — Kirsten Rat Sarcoma Viral Oncogene Homolog; mutations in plasma ctDNA may indicate poor prognosis and have predictive value in NSCLC. Serial ddPCR of KRAS ctDNA serves as a pharmacodynamic biomarker in early-phase clinical trials. See kras.

• BRAF V600E — BRAF V600E ctDNA can be detected pre-operatively as a discriminative tool between benign and malignant thyroid nodules. Vemurafenib shows antitumor activity in BRAFV600E-positive papillary thyroid cancer. See braf.

• RET M918T — Circulating RET M918T mutation in ctDNA is predictive of overall survival in advanced medullary thyroid carcinoma and can monitor treatment response. See ret.

• KIT / PDGFRA — Mutations in these receptor tyrosine kinase genes define gastrointestinal stromal tumors (GISTs). KIT exon 11 deletion predicts beneficial response to adjuvant imatinib with “considerably extended progression free survival compared with placebo.” See kit, pdgfra.

• CAPP-Seq — Cancer Personalized Profiling by deep Sequencing; ultrasensitive ctDNA detection method developed by Diehn and Alizadeh groups at Stanford. See capp-seq.

• ddPCR — Droplet digital PCR; enables absolute quantification of mutant ctDNA molecules. Used for serial pharmacodynamic monitoring and mutation detection. See ddpcr.

• TCGA — The Cancer Genome Atlas; referenced for comprehensive molecular characterization of gastric adenocarcinoma subtypes. See tcga.

• Erlotinib / Tarceva — EGFR-TKI with FDA-approved blood-based companion diagnostic for ctDNA EGFR mutation testing in NSCLC.

• Imatinib — Targeted therapy for KIT/PDGFRA-mutant GIST; ctDNA monitoring can track early treatment adaptations. See imatinib.

• Vemurafenib — BRAF V600E inhibitor; phase II trial in BRAFV600E-positive papillary thyroid cancer showed antitumor activity.

Limitations (as stated by authors)

• “The precise mechanism of ctDNA release has not been cleared yet” — while roles for tissue necrosis, apoptosis, and dynamic secretion from tumor cells are suggested, the exact biological mechanisms by which DNA is released into peripheral blood remain not well understood. This limits the ability to predict which tumors will shed detectable ctDNA.

• Small ctDNA fraction in early-stage disease — ctDNA constitutes “fewer than 100 ng/mL” of total plasma and tumor-derived ctDNA is generally ”< 1% of total cfDNA,” creating a significant detection sensitivity challenge, particularly for early-stage cancers.

• “PCR as an amplification tool can launch some known errors which will pass to the sequencing step” — the amplification methods required to detect low-abundance ctDNA introduce polymerase errors that propagate to sequencing results, potentially affecting mutation calling accuracy.

• The review itself is a narrative synthesis — based on 19 selected articles from a literature search (1990–2017), not a systematic review or meta-analysis. The scope is broad (multiple cancer types, multiple technologies) rather than deep on any single application, and the search likely missed relevant studies published after 2017.

Relevance to Clonal Evolution

This review positions ctDNA as a direct measurement tool for clonal evolution in cancer, complementing the more biologically focused stejskal2023-ctdna-biology-review with a stronger clinical application emphasis. Several connections are critical:

ctDNA captures systemic clonal heterogeneity. The paper’s core distinction — ctDNA as a “screenshot” vs. tissue biopsy as a “snapshot” — directly addresses the central problem in clonal evolution research: spatial heterogeneity. Because ctDNA is shed from all tumor subclones across primary and metastatic sites, it provides a more complete picture of the clonal architecture than single-site biopsies. The authors cite Murtaza et al. (2015) showing that ctDNA “imitate[s] the clonal hierarchy concluded from sequencing of tumor biopsies,” confirming that ctDNA-based clonality inference recapitulates tissue-based results.

ctDNA enables tracking of resistance-conferring subclones. The paper highlights that “even a minor genetic clone within the tumor if carries a drug-resistant mutation can be developed after treatment” — a fundamental principle of clonal evolution under selective pressure. ctDNA monitoring can identify such minor clones before clinical progression, enabling adaptive therapy strategies. This connects to therapy-resistance and the clinical implications of intratumor-heterogeneity.

ctDNA as a MRD monitoring tool for evolutionary dynamics. The finding that ctDNA can detect relapse “months in advance compared to conventional follow-up” and can “outline the genetic events of MRD” (minimal-residual-disease) means that ctDNA can capture the earliest stages of clonal expansion from residual disease — a phase of clonal evolution that is invisible to imaging and inaccessible to tissue biopsy. The paper’s emphasis on post-surgical ctDNA surveillance connects to the TruncalClonePersistence framework: residual truncal clones can be tracked through ctDNA as they seed metastasis.

Cancer-type-specific shedding heterogeneity. The paper implicitly documents that ctDNA detectability varies by cancer type (breast, lung, GI, thyroid) and stage, mirroring the observation in Stejskal 2023 cited in that source-summary that brain, renal, prostate, and thyroid cancers show ctDNA detection rates under 50% even in advanced disease. This shedding heterogeneity has implications for which clonal evolution questions can be addressed with ctDNA in which cancer types.

Technology as an enabler of clonal tracking. The three main technologies surveyed — ddPCR (targeted mutation quantification), NGS (broad mutational profiling), and CAPP-Seq (ultrasensitive detection) — represent different trade-offs between sensitivity and breadth for clonal evolution studies. CAPP-Seq’s ability to detect “one molecule of mutant DNA in ten thousands molecules of normal DNA” is particularly relevant for detecting rare resistant subclones early in their expansion.

Figure Descriptions

Figure 1 — “Beads-on-a-string” nucleosome structure. A schematic diagram (attributed to a slide share source) showing the chromatin organization of DNA wrapped around histone octamers. The figure illustrates that linker DNA between nucleosomes is approximately 150 base pairs, which directly corresponds to ctDNA fragment length. The caption explains that this nucleosome-based fragmentation pattern supports the hypothesis that ctDNA originates through apoptosis or necrosis, where caspase-activated DNase cleaves DNA at internucleosomal linker regions. This figure provides the structural rationale for ctDNA size profiling as a diagnostic tool.