2026-06-01

digest generated 2026-06-02

DeLLphi-304: tarlatamab CNS PFS HR 0.40 in brain mets, CR 14.9% vs 5.4%; intracranial control joins OS/PFS benefit in 2L SCLC.

TL;DR

Tarlatamab extends intracranial control in 2L SCLC (brain mets HR 0.40), though CNS endpoint is post-hoc. Lower-GI: ctDNA risk-stratifies NOM rectal pts for distant mets (74% sens) but misses 59% of local regrowths; endoscopy + MRI remain essential.

GI Lower

NOM for rectal cancer is expanding; ctDNA integration into surveillance protocols remains an active question.

ctDNA surveillance in non-operative rectal cancer management

ForStage I-III MSS rectal, cCR/nCR after NAT, undergoing NOM

TL;DRctDNA sensitivity only 41% for local regrowth vs 74% for distant mets in 110 NOM rectal pts; risk stratifies but cannot replace imaging.

Curative Real-world evidence Early signal

ctDNA surveillance in non-operative rectal cancer management — Outcome Sensitivity Specificity Accuracy

Local regrowth 12/29 (41%) 480/509 (94%) 492/538 (91%)
Distant metastasis 31/42 (74%) 611/627 (97%) 642/669 (96%)

Outcome	Sensitivity	Specificity	Accuracy
Local regrowth	12/29 (41%)	480/509 (94%)	492/538 (91%)
Distant metastasis	31/42 (74%)	611/627 (97%)	642/669 (96%)

7 details

Methods

🔍 N=110, INTERCEPT program, MD Anderson 2020-2024; stage I-III MSS rectal adenocarcinoma, cCR/nCR after NAT → NOM; Signatera™ tumor-informed ctDNA
🔍 22/23 pts with local regrowth underwent salvage surgery; ctDNA-positive at regrowth: ypT3-4 75% vs 21% (ctDNA-negative), p=0.01

Results

📊 Ever-positive longitudinal ctDNA: worse 2-yr local regrowth-free survival (log-rank p=0.0002) and metastasis-free survival (p<0.0001)
📊 First post-NAT ctDNA positive (within 180 days): worse regrowth-free (p=0.0006) and metastasis-free survival (p<0.0001)
📊 Positive ctDNA associated with regrowth (~60%) and distant mets (~60%) in ctDNA-positive pts

Critique

⚠️ Sensitivity only 41% for local regrowth — negative ctDNA does not exclude local recurrence; endoscopy and MRI remain essential
⚠️ Single-institution non-randomised cohort, N=110, median f/u 25 months; no ctDNA-guided vs standard surveillance arm

Open questions

Does ctDNA-guided intensified surveillance improve salvage surgery success rates?
Optimal ctDNA testing frequency and timing within NOM protocols
Performance in dMMR/MSI-H pts (all MSS here)

Sourced from @NiuSanford

📚 Sources · 🐦 1 tweet

Important study re: ctDNA for non-op surveillance in rectal ca.

Pos ctDNA associated w regrowth (60%) & distant mets (60%), but neg ctDNA doesn't exclude local regrowth (sensitivity only 41%)

ctDNA good for risk stratification but not surveillance replacement #ASCO26 @OncoAlert pic.twitter.com/UQQub5QfNB
— Dr. Nina Niu Sanford (@NiuSanford) June 1, 2026

Thoracic / Lung

2L SCLC CNS control is a growing priority as systemic responses improve; DeLLphi-304 is the first randomised data showing intracranial benefit with a BiTE.

DeLLphi-304 (tarlatamab CNS outcomes)

For2L ES-SCLC; brain mets subgroup ≥1 BM at baseline, >70% prior CNS treatment

TL;DRCNS PFS HR 0.54 (ITT), HR 0.40 in brain mets; CNS CR 14.9% vs 5.4% with tarlatamab vs chemo 2L SCLC.

vs leading data

DeLLphi-304 primary results: tarlatamab OS + PFS benefit in 2L SCLC; CNS data extends intracranial signal

Systemic Palliative Phase 3 RCT Caveats dominate

DeLLphi-304 (tarlatamab CNS outcomes) — Arm n Median CNS PFS (mo) HR (95% CI)

Tarlatamab 67 6.5 (4.3-13.7) 0.40 (0.24-0.66)
Chemotherapy 56 4.2 (2.9-5.5) ref

Arm	n	Median CNS PFS (mo)	HR (95% CI)
Tarlatamab	67	6.5 (4.3-13.7)	0.40 (0.24-0.66)
Chemotherapy	56	4.2 (2.9-5.5)	ref

+2 more figures

Arm	n	Median CNS PFS (mo)	HR (95% CI)
Tarlatamab	254	NE (13.7, NE)	0.54 (0.39-0.75)
Chemotherapy	255	7.2 (5.6, NE)	ref

Endpoint	Tarlatamab (n=67)	Chemo (n=56)
CNS CR	10 (14.9%)	3 (5.4%)
Non-CR/Non-PD	42 (62.7%)	37 (66.1%)
CNS DCR	52 (77.6%)	40 (71.4%)
Median duration CNS CR (mo)	NE	3.6
Median duration CNS DC (mo)	8.2	5.2

5 details