Integration map — legacy PMS conclusions → Plus PMS / PMCF Plan

Five named integration points that route the legacy-device post-market conclusions into the Plus PMS Plan (R-TF-007-001), PMCF Plan (R-TF-007-002) and CER (R-TF-015-003). Every Plus-side element below names the legacy input by document ID + section and states the follow-up or confirmation activity it triggers. No dangling inputs (every legacy conclusion is routed); no dangling outputs (every new Plus subsection is grounded in a named legacy input).

Point 1 — Benefit-confirmation monitoring

Legacy input. R-TF-015-012 Appendix D (Study Report) — co-primary endpoints B2 (diagnostic-assessment change rate), C4 (treatment decisions informed by severity), D4 (referral-adequacy improvement), each MCID-positive under Holm-Bonferroni family-wise α = 0.05; supportive quantitative endpoints B4, B6, C5, D2, D6, D7 all MCID-positive. Consolidated in R-TF-007-003 §4.7 (umbrella legacy PMS Report, Report of the cross-sectional observational study).

Plus output. R-TF-007-001 new subsection § Benefit confirmation — continuity with legacy evidence (lives inside § PMCF activities, before the generic PMCF bullet list). Cites the three co-primary endpoints and their MCID deltas per benefit (7GH, 5RB, 3KX); commits the Plus post-market programme to continuing benefit-confirmation monitoring at least at the legacy cadence through the PMCF activities in R-TF-007-002 §Specific PMCF Methods A–D. Cross-reference into R-TF-015-003 §Legacy-device post-market evidence and §Post-market clinical study subsection.

Point 2 — Safety-signal pattern as baseline

Legacy input. R-TF-015-012 Appendix D §Data-quality exclusions and §Safety (Section F). Analysis set N = 56 after the pre-specified Section 10.7 evidence-quality substantiation principle. F1 (potentially-misleading output) 26.8 %, below the protocol's 30 % follow-up threshold after data-quality exclusions; F2 (usability-affecting-clinical-use) 30.4 %; F3 (overall perceived safety, Likert 1–5) mean 4.14 / 5; F4 (formal institutional incident) 7.1 %. Aggregate 21-contract / ≈ 250,000-diagnostic-report denominator with 0 MDR Article 87 serious incidents, 0 Article 88 trend reports, 0 FSCAs, 7 non-serious complaints. Consolidated in R-TF-007-003 §4.7.5 and §2 / §3 (passive surveillance).

Plus output. R-TF-007-001 new subsection § Carried-forward safety signals from legacy post-market (lives after § PMCF activities, before § Indicators and threshold values). Names F1 26.8 %, F2 30.4 %, F3 mean 4.14 / 5, F4 7.1 % as baseline rates. Adds trigger conditions for an unscheduled CER / PMS review under GP-015 when observed Plus rates exceed the legacy baselines by the pre-specified margins. Augments § Indicators and threshold values for benefit-risk reassessment with a new "Carried-forward safety-signal baselines" sub-bullet that feeds directly into the trend-analysis escalation path (§ Trend analysis).

Point 3 — Surveillance-cadence rationale

Legacy input. R-TF-007-003 umbrella legacy PMS Report §2 (passive surveillance), §3 (proactive surveillance), §5 (trend analysis) — ≈ 250,000 diagnostic reports over four-plus years of continuous commercial deployment across 21 active client contracts; 0 Article 87 serious incidents (rule-of-three one-sided 95 % upper bound ≤ 3 / 250,000 ≈ 0.0012 %); 0 Article 88 trend reports triggered; 0 FSCAs; 7 closed non-serious complaints. Planned under R-TF-007-005 § 5 (surveillance cadence and responsibilities).

Plus output. R-TF-007-001 § Surveillance period updated to declare the annual / quarterly-interim cadence as calibrated against the legacy-device operational experience rather than defaulted from a template. New cross-reference paragraph in § Legacy Device Data (existing subsection under § Data collection activities) names R-TF-007-003 as the evidence basis for why Plus vigilance is reviewed annually (PSUR) with quarterly quality-indicator reviews rather than at a higher or lower frequency. No cadence values change; the rationale becomes visible.

Point 4 — Residual uncertainties → PMCF objectives

Legacy input. R-TF-015-012 §11 Methodological limitations and §Discussion residual-uncertainty bullets:

• physician-reported nature of outcomes (not independently measured patient outcomes),
• cross-sectional design with retrospective recall (no within-subject longitudinal tracking),
• attribution uncertainty (observational design; no concurrent control).

Consolidated in R-TF-007-003 §4.7.6 (Appendix D conclusions) and §9 (Conclusions).

Plus output. R-TF-007-002 new subsection § Residual uncertainties from legacy PMS: confirmation in PMCF (lives after § Continuous Evaluation Lifecycle, before § Evaluation of Equivalent and Similar Devices). For each residual uncertainty, names which existing Plus PMCF activity delivers the confirmation or strengthening (MDCG 2020-6 §6.3 permits confirm / strengthen; §6.4 forbids fills / closes):

• Physician-reported → Activities B.1–B.5 (prospective ICC against expert consensus on device-captured images; independently measurable outcomes per benefit 5RB) and Activity A.2 (histopathology / clinical-outcome reference standard for benefit 7GH malignancy detection).
• Cross-sectional / non-randomised → Activities B.1, B.2, B.3, B.4, B.5 (prospective longitudinal with repeated measurements where applicable) and Activity A.1 (large-scale retrospective with temporal stratification).
• Attribution uncertainty → Activities A.1, A.2, A.3 (site-stratified pre / post or concurrent-site comparisons; before-and-after device-deployment workflow analysis embedded in the activity designs).

No new PMCF activities are added; the mapping is structural.

Point 5 — Continuity of the evidence-quality control

Legacy input. R-TF-015-012 §8.4 (evidence-quality substantiation principle for quantitative endpoints), §10.7 (extension of that principle to Section F conditional safety items), §10.4 (records-consulted-vs-estimate-based sensitivity analysis; aggregate records-consulted proportion 35.9 %, sensitivity-analysis conclusions unchanged). Consolidated in R-TF-007-003 §4.7.

Plus output. R-TF-007-002 new subsection § Evidence-quality substantiation: continuity from legacy PMS (lives immediately after § Residual uncertainties from legacy PMS: confirmation in PMCF). Declares that all Plus PMCF quantitative data collection (Activities A–C) applies the same substantiation principle: a binary response without a corresponding free-text substantiation is not evidentially usable. Specifies the records-consulted threshold (≥ 35.9 % of any quantitative PMCF data batch must be records-based rather than estimate-based) and the consequence when the threshold is missed — repeat collection, not adjust conclusions. This is a quality-control principle; it does not override any individual PMCF activity's own acceptance criteria.

CER §Post-market surveillance cross-reference

Plus output in R-TF-015-003. §Stage 4 (PMS / PMCF feedback loop) paragraph updated to reference the five integration points explicitly; the existing § R-TF-015-012 per-study subsection gains a closing sentence naming the Plus PMS / PMCF subsections that consume the legacy evidence. No new benefits or performance claims are introduced in the CER — the continuity strengthens the benefit-risk argument without changing the evidence portfolio.

Validation

• No dangling inputs: every legacy conclusion above is routed to one named Plus-side subsection.
• No dangling outputs: every new Plus-side subsection is grounded in a named legacy input.
• Regulatory framing discipline (task-3b5 / task-3b6 parity): "confirm" / "strengthen" everywhere; "fills" / "closes" nowhere. The legacy evidence is a named post-market input to Plus, not a premarket evidence substitute.
• R-number anchoring: every cross-reference names documents by R-number; no file-path leaks.
• No double-counting: legacy evidence rows in the CER evidence-classification table remain authoritative; the Plus PMS / PMCF Plan references those rows rather than re-listing the evidence as a second stream.