Paper - Review

DOI: 10.1038/s43018-020-0096-5

Abstract

ICB (Immune Checkpoint Blockade) → provides (clinical benefit)

❗ Existing biomarkers do NOT reliably predict (treatment response) ← across diverse cancer types
(Limited data) show → how serial ctDNA (circulating tumor DNA) testing may perform (← as a predictive biomarker)

To assess ctDNA in 5 distinct cohorts of patients (← with advanced solid tumor treated with pembrolizumab)

Baseline ctDNA concentration correlated with
1⃣ progression-free survival
2⃣ overall survival
3⃣ clinical response
4⃣ clinical benefit

Introduction

ICB (← using inhibitors of PD-1 & PD-L1) → emerged as (an effective therapy)
❗ There is NO robust predictive marker of treatment response

To pursue new non-invasive biomarkers (← of ICB response)
→ can provide (predictive value) & (early determination of clinical benefit)

ctDNA (← within peripheral blood plasma)
→ provides non-invasive access to cancer-specific somatic mutations
→ has NOT ❌ yet been clinically implemented → for patients treated with ICB
→ could assist in (prognostication & response monitoring)

⭐ Hypothesis
1⃣ Baseline ctDNA levels would be prognostic
2⃣ Early changes in ctDNA levels would precede radio-graphic response

Results

A multi-cohort clinical trial of ICB in advanced solid tumors

A single-institution phase II study of pembrolizumab ← in patients with advanced solid tumors

Five parallel cohorts
1⃣ SCCHN: squamous cell cancer of head and neck
2⃣ TNBC: triple negative breast cancer
3⃣ HGSOC: high-grade serous ovarian cancer
4⃣ malignant melanoma
5⃣ MST: mixed solid tumors

⭐ the feasibility (← of using ctDNA) as an indicator of tumor burden
→ with advanced solid tumors being treated with pembrolizumab

Bespoke ctDNA assays performed at baseline are sensitive and prognostic

WES
← protocol-mandated fresh-frozen tumor biopsy tissue
← within 4 weeks of treatment initiation (n = 71)

Mutation burden
← within individual cohorts (← with malignant melanoma)
→ displaying the highest rate

16 selected clonal somatic mutations
← for personalized ctDNA assay design
→ Baseline ctDNA was detected in 98%
→ NO differences (← between fresh-frozen tumor biopsy tissue & archival tumor tissue)

NO statistically significant correlation (← between baseline ctDNA level & baseline response evaluation criteria)

Change in ctDNA levels from baseline are predictive of benefit to ICB across cancer types

❓ Dynamic changes (← in ctDNA levels) were predictive of (benefit to ICB)
Change (← in ctDNA levels) from baseline to C3 → display less ↓ variability (← across cancer types)

❗Correlation of early ctDNA kinetics (← during ICB treatment)

Association of ΔctDNA with clinical outcomes is independent of other putative ICB biomarkers

Compare (ctDNA & ΔctDNA) (← with TMB & PD-L1 staining)
→ to evaluate the value (← of ctDNA as therapeutic biomarker)
∴ (Both ctDNA metrics displayed weak negative correlation with PD-L1) & (No correlation with TMB)
∴ The (prognostic & predictive) value (← of ctDNA quantification) → be applicable to (micro-satellite stable patients)

Combination of cycle 3 RECIST and ΔctDNA refines risk groupings

Patient (← who progress early in ICB treatment) → represent a challenging clinical scenario → may eventually be (effective & extend) survival
∴ Examined → INSPIRE patients (← with serial ctDNA values)

Clearance of ctDNA during treatment identifies a highly favorable risk group

(the predictive value) (← of ΔctDNA) treated with pembrolizumab
→ ❓ dynamic changes (← throughout the treatment)
→ (clinical utility) ← of ctDNA-based monitoring

Patients
← sustained ctDNA clearance & radiographic response
→ ctDNA clearance often preceded (the radiographic response)

Discussion

⭐ Biomarker-directed use of ICB
👎: Only a minority (← with solid cancer) will derived sustained response

Tumor characteristics
← including TMB & PD-L1
→ have shown (variable predictive value) ← depending on cancer type

Monitoring (← of ctDNA dynamics)
→ broader application (← biomarker-directed ICB)
→ ctDNA can be non-invasively accessed during treatment

(Effect size) ←was (modest & partly confounded) by tumor type

Pseudo-progression (← early in the course of ICB treatment)
→ remains difficult → to recognize with (current clinical & radiological methods)

Evaluated (→ the value of ctDNA kinetics)
→ as a (therapeutic biomarker) ← during ICB treatment

👎: Lack (← of earlier on-treatment ctDNA assessment)

---