diff --git a/scripts/notebook_tools/check_kernel_drift.py b/scripts/notebook_tools/check_kernel_drift.py
index 3dc1a63f66..afebbfd025 100644
--- a/scripts/notebook_tools/check_kernel_drift.py
+++ b/scripts/notebook_tools/check_kernel_drift.py
@@ -198,12 +198,19 @@ def body_has_derive_exemption(body):
     tolerates the unaccented 'derive' (covers authors who type the header
     without the accent, a common shortcut when reviewing on a non-French
     keyboard layout).
+
+    Fix v3 (c.1367): regex tolerates an optional trailing token on the
+    header line (e.g. ``(C.4)``). The convention adopted in EPIC #16638 PRs
+    REACCENT was ``## Diagnostic dérive (C.4)`` — the v2 regex required
+    end-of-line after the word, silently rejecting the suffix as a non-match
+    despite carrying the same C.4 acknowledgement intent.
     """
     if not body:
         return False
-    # Case-insensitive header, optional whitespace, optional accent on 'e'.
+    # Case-insensitive header, optional whitespace, optional accent on 'e',
+    # optional trailing token (e.g. "(C.4)") on the same line.
     pattern = re.compile(
-        r"^##\s*Diagnostic\s*d[ée]rive\s*$",
+        r"^##\s*Diagnostic\s*d[ée]rive\b[^\n]*$",
         re.MULTILINE | re.IGNORECASE,
     )
     return bool(pattern.search(body))
diff --git a/scripts/notebook_tools/tests/test_check_kernel_drift_fixes.py b/scripts/notebook_tools/tests/test_check_kernel_drift_fixes.py
index 89934809de..2988069a8a 100644
--- a/scripts/notebook_tools/tests/test_check_kernel_drift_fixes.py
+++ b/scripts/notebook_tools/tests/test_check_kernel_drift_fixes.py
@@ -297,6 +297,25 @@ def test_body_has_derive_exemption_unaccented():
     assert ckd.body_has_derive_exemption(body) is True
 
 
+def test_body_has_derive_exemption_c4_suffix():
+    """Fix v3 (c.1367): regex tolerates an optional trailing token on the
+    header line, such as ``(C.4)``. EPIC #16638 PRs REACCENT use this form
+    systematically; the v2 regex rejected it silently.
+    """
+    body = "## Diagnostic dérive (C.4)\nblah\n"
+    assert ckd.body_has_derive_exemption(body) is True
+    # Same with unaccented derive and parenthetical
+    body = "## Diagnostic derive (C.4)\nblah\n"
+    assert ckd.body_has_derive_exemption(body) is True
+    # Trailing token after a longer annotation must also pass
+    body = "## Diagnostic dérive — racine upstream connue\nblah\n"
+    assert ckd.body_has_derive_exemption(body) is True
+    # Header MUST begin with '## Diagnostic dérive' — a body that only
+    # contains 'dérive' in prose should not match.
+    body = "Some prose mentioning Diagnostic dérive (C.4) inline\n"
+    assert ckd.body_has_derive_exemption(body) is False
+
+
 # === Defect 1 v2: end-to-end branchement test (PR_BODY_FILE -> _run() -> exit 0) ===
 
 def test_run_reads_pr_body_file_and_exempts(tmp_path, monkeypatch, capsys):