# MK-3475-975/KEYNOTE-975: HCP commentary report

> 180 verified statements (80 from speakers with a confirmed clinical role) from 16 commentary sources · 11 findings · updated Oct 18, 2026

## Report

- Trial: A Randomized, Double-blind, Placebo-controlled Phase 3 Trial of Pembrolizumab (MK-3475) Versus Placebo in Participants With Esophageal Carcinoma Receiving Concurrent Definitive Chemoradiotherapy (KEYNOTE 975)
- NCT ID: NCT04210115
- Registry: https://clinicaltrials.gov/study/NCT04210115
- Sponsor: Merck Sharp & Dohme LLC
- Phase: Phase 3
- Status: Completed
- Drugs: pembrolizumab, cisplatin, fluorouracil, leucovorin, oxaliplatin
- Version: v1.0.0, edition 1
- Accepted: 2026-10-18
- Last checked: 2026-10-18
- Release hash: 4f7b7e8fa4f3db92ae461ad654665f6ebb48123a7afd2165ee0a42ca692b6504

## Counts

- HCP commentary sources screened: 22
- Sources cited by verified statements: 17
- Verified statements: 180
- From speakers with a confirmed clinical role: 80
- Findings: 11
- People found in screened sources: 25
- People identity-verified: 19
- Voices cited: 24

Counts describe the retained sources. They do not estimate how many HCPs hold a view.

## Abstract

KEYNOTE-975 asks whether adding pembrolizumab to definitive chemoradiation improves outcomes for patients with esophageal cancer treated without planned surgery. Clinicians discuss survival endpoints, residual disease, treatment sequencing, histology, PD-L1 testing, staging and multidisciplinary referral, and immune toxicity. Other topics include chemotherapy choices, radiation dose, perioperative alternatives, organ preservation, response assessment, surveillance and evidence needed to change care. The trial’s main results are expected at ESMO 2026; this report reflects commentary before that presentation.

## Open Medical Affairs question

How should the announced lack of EFS improvement be communicated without implying known OS, subgroup or safety conclusions?

Raised by the leading finding. The report shows what clinicians said and the sources behind it; it does not settle the question.

## Findings (titles)

- F1. Negative EFS announcements are consequential, but are not a complete results assessment
- F2. Routine post-chemoradiation nivolumab without surgery is questioned, with selective exceptions
- F3. Histology drives expectations and challenges interpretation of pooled populations
- F4. Staging errors can change the entire definitive-treatment decision
- F5. PD-L1 scoring concerns complicate confidence near treatment thresholds
- F6. Perioperative advances narrow—but do not eliminate—the perceived role of dCRT
- F7. Organ preservation exposes a substantive dispute over surgery's value and burden
- F8. Nonoperative management depends on uncertain selection and nonuniform surveillance
- F9. Concurrent timing, chemotherapy restrictions and radiation dose shape interpretability
- F10. Research enthusiasm is qualified by endpoint priorities, feasibility and uncertain mechanisms of benefit
- F11. Immune-toxicity vigilance qualifies enthusiasm for treatment intensification

## What sits behind each finding

Counts describe what the report retained, not how many clinicians hold a view. A statement can support more than one finding.

| Finding | Statements | Passages | Sources | Named voices | Counterview | Limitations |
| --- | ---: | ---: | ---: | ---: | --- | ---: |
| F1 | 6 | 6 | 5 | 4 | none recorded | 2 |
| F2 | 17 | 9 | 3 | 2 | none recorded | 2 |
| F3 | 15 | 8 | 7 | 4 | recorded | 3 |
| F4 | 21 | 11 | 3 | 3 | none recorded | 2 |
| F5 | 36 | 17 | 2 | 2 | recorded | 2 |
| F6 | 23 | 13 | 6 | 5 | recorded | 2 |
| F7 | 14 | 14 | 1 | 6 | recorded | 2 |
| F8 | 18 | 16 | 3 | 7 | recorded | 2 |
| F9 | 11 | 9 | 3 | 4 | recorded | 2 |
| F10 | 19 | 16 | 4 | 10 | none recorded | 3 |
| F11 | 8 | 3 | 1 | 0 | none recorded | 2 |

### What the statements are

| Finding | Views | Questions and uncertainty | Clinical experience | Comparisons | Restated results and context | Passages supporting | Passages giving context | Sources by kind | Feeds the Medical Affairs question |
| --- | ---: | ---: | ---: | ---: | ---: | ---: | ---: | --- | --- |
| F1 | 2 | 0 | 0 | 0 | 4 | 7 | 0 | 3 posts and articles, 1 podcast or video, 1 registry | yes |
| F2 | 5 | 5 | 2 | 0 | 5 | 23 | 0 | 3 podcasts and video | yes |
| F3 | 3 | 5 | 0 | 2 | 5 | 21 | 0 | 7 podcasts and video | yes |
| F4 | 6 | 3 | 4 | 1 | 7 | 23 | 3 | 3 podcasts and video | yes |
| F5 | 3 | 2 | 3 | 2 | 26 | 78 | 1 | 1 podcast or video, 1 registry | yes |
| F6 | 8 | 5 | 1 | 3 | 6 | 26 | 0 | 1 post or article, 5 podcasts and video | yes |
| F7 | 8 | 5 | 0 | 1 | 0 | 15 | 0 | 1 post or article | yes |
| F8 | 4 | 6 | 6 | 0 | 2 | 19 | 1 | 2 posts and articles, 1 podcast or video | yes |
| F9 | 4 | 4 | 0 | 1 | 2 | 12 | 0 | 1 post or article, 1 podcast or video, 1 registry | yes |
| F10 | 7 | 5 | 2 | 0 | 5 | 20 | 0 | 2 posts and articles, 2 podcasts and video | yes |
| F11 | 1 | 0 | 0 | 1 | 6 | 8 | 1 | 1 podcast or video | yes |

### Trial dimensions

- Primary endpoint: Event-free Survival (EFS): F1 (3 shared claims)
- Primary endpoint: Overall Survival (OS): F3 (2 shared claims), F10 (1 shared claims)
- Secondary endpoint: Number of participants with an adverse event (AE): no finding draws on it directly
- Secondary endpoint: Number of participants discontinuing study treatment due to an adverse event (AE): no finding draws on it directly
- Population and setting: F3 (10 shared claims), F4 (5 shared claims), F6 (6 shared claims)
- The comparison: F3 (3 shared claims), F9 (2 shared claims)
- Safety and tolerability: F9 (1 shared claims), F11 (8 shared claims)
- Design and conduct: F3 (1 shared claims), F9 (2 shared claims)

## Sample findings

Paraphrased; the full report links each statement to its source.

### F2. Routine post-chemoradiation nivolumab without surgery is questioned, with selective exceptions

A podcast moderator asked whether adjuvant nivolumab should be offered after neoadjuvant chemoradiation when a patient with initially resectable disease declines surgery or is judged surgically ineligible.

The drug and timing differ from KEYNOTE-975: this question concerns nivolumab after neoadjuvant chemoradiation, rather than pembrolizumab given with definitive chemoradiation.

Source: podcast, 2024-02-07.

### F5. PD-L1 scoring concerns complicate confidence near treatment thresholds

A speaker in a YouTube discussion asked which tissue to use for PD-L1 CPS testing and whether testing should precede or follow therapy. Further questions addressed study-specific versus routine scorer training and when treatment could proceed without PD-L1 staining.

This was a general gastroesophageal biomarker discussion from 2022, not an evaluation of assay performance in KEYNOTE-975.

Local reporting thresholds discussed reflected practice at that time, not KEYNOTE-975 selection criteria.

Source: YouTube video, 2022-06-22.

## What's inside

| Section | Items | Access |
| --- | ---: | --- |
| Findings with interpretation, counterviews and limits | 11 | Titles free |
| Detailed analysis | 12 sections | Paid |
| Verified statements with short source quotes and links | 180 | Paid |
| Trial aspects mapped | 16 | Free |
| Sources with links | 17 | Counts free |
| Speaker attribution for cited voices | 24 | Paid |

## Trial aspects

14 discussed · 2 no commentary found · 0 assessment incomplete.

| Aspect | Coverage | Statements | Sources |
| --- | --- | ---: | ---: |
| Primary endpoint: Event-free Survival (EFS) | discussed | 3 | 2 |
| Primary endpoint: Overall Survival (OS) | discussed | 3 | 2 |
| Secondary endpoint: Number of participants with an adverse event (AE) | no commentary found | 0 | 0 |
| Secondary endpoint: Number of participants discontinuing study treatment due to an adverse event (AE) | no commentary found | 0 | 0 |
| Population and setting | discussed | 21 | 9 |
| The comparison | discussed | 6 | 5 |
| Safety and tolerability | discussed | 9 | 2 |
| Design and conduct | discussed | 3 | 2 |
| Residual disease without surgery: additional systemic therapy, withholding and selective exceptions | discussed | 17 | 3 |
| Staging, airway invasion, referral and multidisciplinary suitability | discussed | 21 | 3 |
| CPS measurement, reporting and confidence near treatment thresholds | discussed | 35 | 1 |
| Who still receives dCRT amid perioperative and surgical alternatives? | discussed | 23 | 6 |
| Surgery omission, organ preservation and salvage-surgery trade-offs | discussed | 15 | 2 |
| Complete-response assessment and surveillance for nonoperative management | discussed | 18 | 3 |
| Research priorities, meaningful endpoints, feasibility and expectations | discussed | 27 | 8 |
| Radiation-dose debate and alternative chemoradiation backbones | discussed | 11 | 2 |

An aspect marked incomplete is a question the public commentary did not let us answer. We show it rather than hide it.

## Sources

| Channel | Screened | Cited |
| --- | ---: | ---: |
| X threads | 13 | 7 |
| YouTube | 5 | 5 |
| Podcasts | 4 | 4 |
| Registry (primary source, cited for context) | not screened | 1 |

- Screened window: 2021-03-31 to 2026-10-16
- Cited window: 2022-06-22 to 2026-10-16
- Retrieved: 2026-10-18

Screened sources were retrieved for this trial. Cited sources back at least one verified statement and include primary sources used for context.

## The trial

From ClinicalTrials.gov.

- Title: A Randomized, Double-blind, Placebo-controlled Phase 3 Trial of Pembrolizumab (MK-3475) Versus Placebo in Participants With Esophageal Carcinoma Receiving Concurrent Definitive Chemoradiotherapy (KEYNOTE 975)
- Registry: NCT04210115 (https://clinicaltrials.gov/study/NCT04210115)
- Study IDs: 3475-975 · MK-3475-975 (MSD) · KEYNOTE-975 (MSD) · PHRR200210-002490 (PHRR) · 205261 (Japic-CTI) · 2022-501531-16-00 (EU CT) · U1111-1281-0822 (UTN) · 2019-002006-51
- Sponsor: Merck Sharp & Dohme LLC
- Design: Phase 3 · interventional, randomized, parallel, phase3
- Status: Completed
- Enrollment: 703 (actual)
- Start: 2020-02-28 (actual)
- Primary completion: 2026-01-21 (actual)
- Primary endpoint: Event-free Survival (EFS) (Up to ~60 months)
- Primary endpoint: Overall Survival (OS) (Up to ~72 months)
- Intervention: pembrolizumab (biological)
- Intervention: placebo (drug)
- Intervention: cisplatin (drug)
- Intervention: 5-FU (drug)
- Intervention: radiotherapy (radiation)
- Intervention: leucovorin (drug)
- Intervention: levoleucovorin (drug)
- Intervention: oxaliplatin (drug)

## Scope and limits

This report does not judge efficacy, estimate HCP prevalence or consensus or give treatment advice.

- Discussion that does not name the trial may be missing from this edition.
- For 1 recording, a transcript separated by speaker was not available, so quotes come from the original automatic transcript and may lack speaker labels.
- Some passages in the sources used in this report were not fully assessed for clinician views, so a view there may be missing from this edition.
- Out of scope: 18 passages in the reviewed sources discuss other trials or topics outside this report's scope; they are not part of this edition.
- 79 statements in the sources reviewed for this edition discuss other treatment settings or topics unrelated to this trial's question and are not included.
- Further source limits apply, so some matching discussion may be missing.
- The sources reviewed are not a representative sample of HCP opinion, and multiple formats of one discussion do not represent independent agreement.
- Some potentially relevant posts and a video were unavailable to read. Parts of several long discussions were not completely assessed, so additional relevant opinions may be missing.
- Several treatment, histology and safety comments cannot be reliably assigned to a named speaker. A program roster, institutional reference or professional identifier does not resolve that uncertainty.
- Professional information is incomplete for some named contributors. Daisuke Kotani's and Megan Lenihan's clinical roles are not confirmed for this edition; names and source attribution do not verify credentials.
- The negative EFS statements relay sponsor announcements. No KEYNOTE-975 results publication, quantitative efficacy analysis, detailed subgroup analysis or comparative safety analysis is among the sources reviewed.
- Earlier anticipatory comments and later announcement reactions come from different dates and contexts. Their sequence does not establish a change of mind or direct disagreement over detailed results.
- Some transcript wording is unclear, including an overexpression qualifier in the adenocarcinoma discussion; the initial organ-preservation proposal is truncated.
- Cross-disease analogies, other-study response figures, institutional experience and anecdotal tolerability observations are not KEYNOTE-975 outcomes.
- No full protocol document or complete regulatory evidence is available here. Descriptions of prescribing labels and local treatment standards remain the contributors' accounts.
- General CPS-scoring concerns do not establish inaccurate testing in KEYNOTE-975, and local historical reporting thresholds must not be substituted for trial criteria.

How reports are built and checked: https://www.erudio.com/methods

## Editions and corrections

- v1.0.0, accepted 2026-10-18: current edition

Last checked for new commentary 2026-10-18. No corrections. Versions follow semver: a new or changed finding is a minor version, more evidence without a material change is a patch.

## Offer

- Size: Standard (180 verified statements)
- Current edition: $1,250. A purchase starts a check for new commentary and any newer edition follows at no charge
- 3 months of weekly updates: $2,200
- 6 months of weekly updates: $3,100
- 12 months of weekly updates: $4,400
- Optional identity linkages (12-month updates only, product updates-12-identities; speakers are named and described in every report): NPI and other verified IDs at $15 per identity-verified person in the edition at purchase (19 now), at most $1,250; people verified later in the term at no charge
- Pricing: https://www.erudio.com/pricing
- Buy now (card, through Stripe): https://www.erudio.com/reports/mk-3475-975-keynote-975-nct04210115
- Agents: POST https://www.erudio.com/api/orders with {"report": "mk-3475-975-keynote-975-nct04210115", "product": "edition" | "updates-3" | "updates-6" | "updates-12" | "updates-12-identities", "buyer_email": "...", "licence_key": "<your key, if you hold one>"}; it returns a payment link for your user (and a pay_url your own MPP or x402 payment client can pay over HTTP 402, when offered), and once paid the status_url returns your license key. Send the same key with every later order so all reports sit on one key. The buyer gets their account key by email, which reads every report bought with their address; recover_key (POST https://www.erudio.com/api/keys/recover) sends it again
- Talk to Audun Utengen and Tom Lee (invoice, PO, questions): https://www.erudio.com/contact?report=mk-3475-975-keynote-975-04210115

## Cite

Erudio Health. MK-3475-975/KEYNOTE-975 (NCT04210115): HCP commentary report. Edition 1, v1.0.0. Accepted Oct 18, 2026. https://www.erudio.com/reports/mk-3475-975-keynote-975-nct04210115

## Machine access

- JSON: https://www.erudio.com/reports/mk-3475-975-keynote-975-nct04210115.json
- Markdown: https://www.erudio.com/reports/mk-3475-975-keynote-975-nct04210115.md
- MCP: https://mcp.erudio.com/mcp, tool `get_report_card` with `{"slug": "mk-3475-975-keynote-975-04210115"}`