Immune cell topography predicts response to PD-1 blockade in cutaneous T cell lymphoma

Darci J. Phillips; Magdalena Matusiak; Belén Rivero‐Gutiérrez; Salil S. Bhate; Graham L. Barlow

doi:10.1038/s41467-021-26974-6

Immune cell topography predicts response to PD-1 blockade in cutaneous T cell lymphoma

Darci J. Phillips, Magdalena Matusiak, Belén Rivero‐Gutiérrez, Salil S. Bhate, Graham L. Barlow, Sizun Jiang, János Demeter, Kimberly S. Smythe, Robert H. Pierce, Steven P. Fling, Nirasha Ramchurren, Martin A. Cheever, Yury Goltsev, Robert B. West, Michael S. Khodadoust...

Nature Portfolio (2021) • Volume 12, Issue 1, Pages 6726-6726

Empirical-ClinicalObservational-CohortPDF AvailableGrade Eligible⚠️ Moderate Risk Flags

Overall Assessment

Adequate Methodological Quality

Assessment created by PaperScores Medical AI v0 on Dec 14, 2025

C+

66/100

Quick Actions

Read Full Paper

Quality Dimensions

Bias & Integrity

COI handling, outcome switching, selective reporting

C-58

Methods Rigour

Protocol clarity, sampling, controls

C62

External Validity

Generalisability, setting, population

D-35

Transparency & Reproducibility

Data/code/materials availability, preregistration

A-86

Significance & Novelty

Contribution vs prior art; context if known

B+82

Statistical Validity

Models, assumptions, multiplicity, CI priority

B-74

Integrity & Transparency

Integrity checks

P-hacking risk?

?

Review recommended

Signs of selective hypothesis testing or analysis choices that could inflate false positive rates.

Outcome switching?

?

Review recommended

Evidence that primary outcomes were changed during or after the study, potentially distorting results.

Conflict of interest?

?

Review recommended

Authors have financial or other relationships that could bias research findings.

Data integrity issues?

?

None suspected

Concerns about the accuracy, completeness, or authenticity of the reported data.

Open science signals

Research data

?

Open access

Raw data used in the analyses is publicly available or accessible upon request.

Analysis code

?

Open access

Code or scripts used to analyze the data are shared for reproducibility.

Study materials

?

Open access

Protocols, questionnaires, and other materials are publicly available.

Premise

Primary Research Question

Does spatial organisation of immune/tumour cells predict pembrolizumab response in advanced CTCL?

Hypothesis

A SpatialScore based on distances between PD‑1+ CD4+ T cells, tumour cells, and Tregs predicts response.

Hypothesis is Falsifiable

✓Yes

Score-performance vs response with pre/post data enables refutation.

PICO Framework

Population✓Yes

Advanced CTCL (mycosis fungoides/Sézary), n=14 from CITN‑10.

Intervention✓Yes

Pembrolizumab administered per trial schedule.

Comparison✓Yes

Responders vs non‑responders; pre vs post biopsies.

Outcome✓Yes

SpatialScore association with response; survival shown.

Literature Positioning

Literature Review Balanced

Well covered

Summarises prior CTCL biomarker failures and spatial TME literature.

Evidence: Intro p.1–3; refs 13–15.

Research Gap Clearly Stated

Well covered

Need predictive biomarkers for PD‑1 response in CTCL stated clearly.

Evidence: Abstract; Intro p.1–2.

Stated Contribution Clear

Well covered

Derives and validates SpatialScore predicting pembrolizumab response.

Evidence: Abstract; Results ‘Spatial signature’ section; Fig.4.

Study Provenance

Peer reviewed venueGrant funding disclosedConflict disclosure provided4 affiliations listed

Authors & Affiliations

Stanford University; Fred Hutchinson Cancer Research Center; University Hospital Tübingen; others.

Funding Statement

Multiple NIH grants; DoD; FDA; CRUK; Gates Foundation; Parker Institute; other foundations and institutional funds.

Conflicts of Interest

Authors report company co-founders (Akoya) and related patents; one author funded by Merck; others declare none.

Peer Review

✓Nature Communications (2021) peer-reviewed article.

Evidence: journal page

Methodological Assessment

Observational StudySTROBEObservational

11 applicable checklist items.

Yes 6 · No 2 · Unclear 3

Study Design & Definitions

✓

Participant Selection Described

Inclusion/exclusion and reason for 10/24 exclusions described; TMA core selection detailed.

Evidence: Results ‘Patient cohort’; Methods ‘Sample collection’

✓

Variable Definitions Clear

Cell types and SpatialScore formula explicitly defined; markers listed.

Evidence: Fig.1e–h; Fig.4a; Supp. Table 2

?

Confounding Addressed

Baseline comparability stated; limited covariate adjustment beyond patient random effects.

Evidence: Supp. Fig.1b; Methods statistics

✓

Temporality Established

Baseline biomarker measured before clinical response.

Evidence: Design; Fig.4b–c (pretreatment).

✓

Exposure Definition Anchored

Exposure/anchors (pembrolizumab, pre/post) well-defined.

Evidence: Methods ‘Human subjects’.

Bias Mitigation

✓

Balance Diagnostics Acceptable

Demographics/outcomes compared at baseline; no significant differences.

Evidence: Supp. Fig.1b.

✗

Selection Bias Addressed

10/24 excluded for insufficient FFPE; no bias analysis of excluded vs included.

Evidence: Results ‘Patient cohort’.

✓

Misclassification Assessed

Malignant vs reactive T cells discriminated with multi-marker and morphology checks.

Evidence: Fig.1f–i; Supp. Fig.2.

?

Time-varying Confounding Handled

Repeated measures handled by mixed models; no formal time-varying confounder control.

Evidence: Methods ‘Statistics’.

Robustness

✗

Negative Controls or E-value

No negative controls or E-values reported.

?

Model Diagnostics Reported

Model diagnostics not reported.

Evidence: Methods ‘Statistics’.

Modules without applicable checklist items (8)

These designs were fully marked as not applicable but remain available for reference.

Quick Actions

Read Full Paper

Study Overview

Study Design

Observational Cohort

Primary classification used to evaluate this research.

Population

Advanced CTCL patients in CITN-10 pembrolizumab trial; pre/post skin biopsies analysed for spatial biomarkers.

Brief summary of who or what was studied.

Sample Size

14

Number of participants included in the study analyses.

Study Arms

2 arms

Total number of intervention or comparison groups.

Pre-registration

Not registered

Indicates whether the protocol or trial was registered before data collection.

Blinding

Unclear

Who was unaware of group assignments: None (everyone knew), Single blind (participants didn't know their group), Double blind (participants and researchers didn't know), Triple blind (also outcome assessors didn't know).

Analysis Set

Other

Which participants were included in analysis: ITT (all randomized, regardless of adherence), mITT (modified ITT, excluding some for valid reasons), Per protocol (only those who completed as planned), As treated (grouped by actual treatment received), Other (alternative approach).

Follow-up

Median follow-up 142 weeks; NR median survival 109 weeks.

Duration of observation or follow-up for outcomes.

Primary Outcome

Unclear whether the primary outcome was pre-specified.

Effect Size

HR: HR 0.0969; p=0.0168

Reviewer Notes

Exploratory cohort study of 14 CTCL patients correlating spatial biomarkers (CODEX/mIHC, RNA-seq) with pembrolizumab response; no allocation.

Publication Details

DOI

10.1038/s41467-021-26974-6

Published

November 18, 2021

Citations

199 citations

FWCI

41.88

External Resources

DOI Link OpenAlex

Disclaimer: This assessment is generated by AI and should not be the sole basis for clinical or research decisions. Always review the original paper and consult with domain experts.

Suggested Papers

D-

Studies on Serological Cross‐Reaction in Sequential Flavivirus Infections

Yoshihiro Makino et al.•1994

D+

A comparison of the efficacy of sulphadoxine-pyrimethamine with azithromycin versus sulphadoxine-pyrimethamine alone for malaria chemoprophylaxis in pregnancy: a triple-blind randomized controlled trial

R O Ogwu et al.•2025

No Report

The Urban Environment and Childhood Asthma (URECA) birth cohort study: design, methods, and study population

James E. Gern et al.•2009

Suggested Papers

From Our Blog

Lead Time Bias: The Illusion of Survival

Screening finds cancer earlier. It does not always make you live longer. It just makes you sick longer.

The Law of Small Numbers: Don't Trust the Trend

Small sample sizes yield extreme results. This is not a pattern. It is noise.

Base Rate Neglect: Why 99% Accuracy is Meaningless

A test can be 99% accurate and still be wrong most of the time. You are ignoring the base rate.