Publications Update # 148
Non-Medical Article of the Week
Idiots are everywhere
What someone believes is not as important as how they chose to believe it.
If you don't know Joshua Johnson, you have not heard NPR enough 😆
Neuroendocrine Carcinoma of the Cervix
Michael Frumovitz, M.D.
This week's single-topic issue features neuroendocrine tumors of the cervix. It is a GYOEDU tradition to honor Dr. Michael Frumovitz every year by discussing this topic. We are all indebited to his contributions to the neuroendocrine tumors of the cervix.
Neuroendocrine neoplasms are broadly divided into well-differentiated neuroendocrine tumors (NETs), graded by proliferative activity, and poorly differentiated neuroendocrine carcinomas (NECs), which are high-grade by definition and have either small-cell or large-cell morphology.
Among extrapulmonary NECs, approximately 35–55% arise in the gastrointestinal or pancreatic tract, with colorectal, esophageal, gastric, and pancreatic primaries among the most frequent; genitourinary sites such as the bladder and prostate are also important, while gynecologic primaries account for roughly 10% of extrapulmonary NECs, with the cervix being the most common gynecologic site.
Neuroendocrine carcinoma of the cervix is about 1% of cervical cancers, but they are highly lethal. These tumors spread to nodes and distant sites early, which is why systemic therapy is central to the management even when the disease looks localized on imaging.
Video: Neuroendocrine Cervical Carcinoma
This video is a part of the Fellow Lecture Series. It will be available for Newsletter subscribers to watch until the next newsletter. If you want full access to the video and several more hours of fellow lecture series videos, please visit https://courses.gyoedu.org
Watched the video? Try the quiz
Shitanshu Uppal
DLL3 and Tarlatamab
Delta-like ligand 3 (DLL3) is a Notch pathway ligand. In normal adult tissue, it is barely expressed and mostly sits inside the cell. In high-grade neuroendocrine cancers, it gets pushed out onto the tumor cell surface.
The reason this ligand has suddenly become important is a new drug called Tarlatamab --> a DLL3 × CD3 bispecific T-cell engager:
DLL3 on the tumor cell ← tarlatamab → CD3 on the T cell
One arm binds DLL3 on the tumor cell, while the other binds CD3 on T cells, bringing the two cells into close proximity and activating T-cell–mediated killing of the tumor.
Tarlatamab: What the Lung Data Actually Show
Tarlatamab (Imdelltra) is FDA-approved for extensive-stage small cell lung cancer that has progressed on or after platinum. It is not approved for cervical neuroendocrine carcinoma as of now.
DeLLphi-301 — the accelerated approval study
A phase II trial in previously treated small cell lung cancer. Patients had a median of two prior lines. The dose taken forward was 10 mg IV every 2 weeks after step-up dosing.
| Outcome | Tarlatamab 10 mg |
|---|---|
| Objective response rate | 40% |
| Median PFS | 4.9 months |
| Duration of response ≥6 months | 59% of responders |
| Grade 3 CRS | 1% |
| Discontinuation for treatment-related toxicity | 3% |
Responses came quickly. Accelerated approval happened in May 2024.
DLL3 testing was not required for enrollment or for the label, as the expression is mainly on tumor cells.
DeLLphi-304 — the randomized data
509 patients with SCLC progressing during or after platinum, randomized to tarlatamab versus investigator-choice chemotherapy (topotecan, lurbinectedin, or amrubicin).
| Outcome | Tarlatamab | Chemotherapy |
|---|---|---|
| Median overall survival | 13.6 months | 8.3 months |
| Hazard ratio for death | 0.60 | — |
| Grade ≥3 adverse events | 54% | 80% |
| Discontinuation for adverse events | 5% | 12% |
A 40% reduction in the risk of death, with less toxicity than chemotherapy. The FDA converted the accelerated approval to traditional approval in November 2025.
Administration
| Treatment day | Dose |
|---|---|
| Cycle 1 Day 1 | 1 mg IV |
| Cycle 1 Day 8 | 10 mg IV |
| Cycle 1 Day 15 | 10 mg IV |
| Thereafter | 10 mg IV every 2 weeks |
One-hour infusion. For Cycle 1 Days 1 and 8, the label asks for dexamethasone 8 mg IV within an hour before treatment, a liter of normal saline afterward, monitoring in a healthcare setting for 22–24 hours from the start of the infusion, and the patient staying within an hour of a facility with a caregiver for 48 hours.
If you are thinking about this for a patient off-label, that monitoring requirement is the first thing to consider and perhaps work with a medical oncologist in your institution.
Toxicities
Cytokine release syndrome is the signature toxicity and carries a boxed warning. About half of patients on the 10 mg regimen had CRS in DeLLphi-301, most were grade 1–2 and overwhelmingly in cycle 1 — fever, hypotension, tachycardia, chills, hypoxia, nausea, fatigue. Severe CRS is uncommon but can require steroids, tocilizumab, and hemodynamic support.
Neurologic toxicity, including Immune effector cell-associated neurotoxicity syndrome (ICANS), is the second boxed warning. Confusion, altered mental status, somnolence, language disturbance, tremor, seizures.
Beyond that: fatigue, decreased appetite, fever, dysgeusia, anemia, constipation, musculoskeletal pain, nausea, cytopenias, infection.
Think of this as a T-cell engager, not a cytotoxic.
Does Any of This Apply to Cervical NEC?
Probably. But we do not know yet.
Tsang and colleagues looked at DLL3 immunohistochemistry across 76 samples from gynecologic tract neuroendocrine tumors and comparison malignancies. Among the cervical cases:
- 83% of cervical small cell carcinoma/NETs were DLL3-positive, all with strong 3+ staining
- 100% of cervical large cell NETs were positive
- High-grade serous carcinomas showed no DLL3 expression
Unanswered. Tarlatamab is not standard of care for cervical NEC, and should only be considered on trial or in a very carefully thought-through off-label situation with the monitoring infrastructure to back it up.
Cadonilimab in NEC Cervical
MD Anderson ran a phase II single-arm study of cadonilimab — a bispecific antibody against PD-1 and CTLA-4 — in recurrent or metastatic small cell neuroendocrine cervical cancer. Cadonilimab 6 mg/kg IV on days 1 and 15 of a 28-day cycle. Nine patients treated, eight evaluable.
| Outcome | Result |
|---|---|
| Objective response rate | 0% (0/8) |
| Disease control rate | 25% (2/8) |
| Median PFS | 2.19 months |
| Progression-free at 6 months | 1/8 |
Two patients had stable disease, lasting 9.4 and 16.2 weeks. Toxicity was mild — one grade 3 event (fatigue), nothing grade 4 or 5.
Why this matters. Dual PD-1/CTLA-4 blockade has looked promising in high-grade neuroendocrine neoplasms in general. Unfortunately, results were not impressive. I would put this in the same mental bucket as single-agent PD-1 therapy in this disease — a well-conducted negative study.
Hysterectomy or Definitive Chemoradiation?
Kobayashi and colleagues analyzed 197 patients with T1b3, T2a2, or T2b small-cell neuroendocrine carcinoma of the cervix in the NCDB. The four treatment groups were: EBRT plus chemotherapy without hysterectomy (63.5%), primary hysterectomy followed by postoperative EBRT plus chemotherapy (17.8%), primary hysterectomy followed by chemotherapy without EBRT (9.6%), and EBRT plus chemotherapy followed by hysterectomy (9.1%). Compared with the non-hysterectomy EBRT + chemotherapy group, primary hysterectomy followed by postoperative EBRT + chemotherapy was associated with improved 5-year OS (68.9% vs 46.9%; adjusted HR 0.42, 95% CI 0.24–0.74). Because this was a retrospective analysis subject to treatment-selection bias and unmeasured confounding, the findings are hypothesis-generating rather than practice-changing.
| Treatment | 5-year OS | Median OS | Adjusted HR for death |
|---|---|---|---|
| EBRT + chemotherapy | 46.9% | 35.4 mo | Reference |
| Primary hysterectomy → EBRT + chemotherapy | 68.9% | 133.2 mo | 0.42 (95% CI 0.24–0.74) |
| Primary hysterectomy → chemotherapy | 31.6% | 25.9 mo | 1.21 (95% CI 0.67–2.20) |
| EBRT + chemotherapy → hysterectomy | 61.1% | Not reached | 0.89 (95% CI 0.42–1.91) |
GYOEDU Updates
GYOEDU Institutional Pricing Calculator
GYOEDU Study Summary App
If you have not yet downloaded the GYOEDU Study Summary App, I highly recommend that you download and try it free for 7 days from the App Store. This is the best thing GYOEDU has put out in its three years of existence.
Follow @gyoedu