Multicenter Phase I Study of Triapine in Combination with [177Lu]Lutetium-Dotatate in Patients with Well-Differentiated Neuroendocrine Tumors (ETCTN 10388)

Abstract

Purpose:

The next generation of theranostics studies is looking at novel combinations with radiation sensitizers to enhance peptide receptor radionuclide therapy’s efficacy. In this study, we present safety and efficacy data from a multicenter phase I study of a [¹⁷⁷Lu]lutetium-dotatate–triapine combination in patients with gastroenteropancreatic (GEP) neuroendocrine tumors (NET).

Patients and Methods:

The study consisted of 2 phases (part A: dose escalation; part B: dose expansion) administering the recommended phase II doses (RP2D) for the 2 agents. All patients received 200 mCi of [¹⁷⁷Lu]lutetium-dotatate on day 1 of four 8-week cycles, in combination with oral triapine administered at assigned dose levels (100, 150, or 200 mg) from day 1 to day 14 of each cycle.

Results:

A total of 31 (A: 15; B: 16) patients received treatment. Across the study, 9 patients experienced dose-limiting toxicities: grade 3 anemia (n = 1), grade 5 cardiac arrest (unrelated to investigational product; n = 1), and grade 4 neutropenia (n = 7). Based on the overall safety and pharmacokinetics data, a [¹⁷⁷Lu]lutetium-dotatate (200 mCi) plus triapine (150 mg) dose was selected as the RP2D for the expansion phase. Grade ≥3 treatment-related adverse events were observed in 77% of patients: anemia (23%), nausea and vomiting (3%), lymphopenia (58%), neutropenia (35%), and decreased white blood cell count (39%). The majority of cytopenias were transient and resolved within 2 weeks. Among the 28 evaluable patients, the objective response rate was 21.4%, and the median progression-free survival has not been reached (median follow-up period, 23.7 months).

Conclusions:

The combination of [¹⁷⁷Lu]lutetium-dotatate (200 mCi) and oral triapine (150 mg D1-14) was well tolerated and demonstrated preliminary activity in patients with well-differentiated GEP-NET (NCT04234568).

Translational Relevance.

Theranostics has reshaped modern oncology. Landmark approvals such as [¹⁷⁷Lu]lutetium-dotatate and [¹⁷⁷Lu]lutetium vipivotide tetraxetan have transformed the therapeutic landscape and validated targeted radiopharmaceutical therapy as a powerful modality. Despite these advances, emerging clinical experience suggests that current theranostic agents may be approaching a therapeutic ceiling, particularly in patients with resistant or heavily pretreated disease. To meaningfully expand the impact of radiopharmaceutical therapy, next-generation trials must move beyond monotherapy paradigms and rigorously test biologically grounded combination strategies—especially the integration of effective radiation sensitizers.

Our study represents one of the early prospective clinical efforts to combine a systemic DNA damage–enhancing agent with an FDA-approved radioligand therapy. We establish feasibility, define the safety profile and recommended phase II dose, and provide preliminary clinical activity. Importantly, this work offers a mechanistic framework for rational theranostic combinations and has broader implications for radiopharmaceutical drug development across multiple tumor types beyond neuroendocrine malignancies.

Introduction

Neuroendocrine tumors (NET) are rare neoplasms that can arise anywhere in the body (1). A recent review of the Surveillance, Epidemiology, and End Results database confirms a fivefold overall increase in the incidence of NET over the past 50 years (2). Due to the indolent nature of NET, their prevalence is disproportionate to their incidence. There are currently more than 170,000 patients diagnosed with NET in the United States (3, 4). Despite the development of several effective targeted therapies, including everolimus and sunitinib, resistant disease is common, and treatment options remain limited. Peptide receptor radionuclide therapy (PRRT) using [¹⁷⁷Lu]lutetium-dotatate has been approved as a second-line treatment by the FDA (1) for metastatic progressive gastroenteropancreatic (GEP) NET following progression on first-line octreotide. Long-term follow-up of the NETTER-1 trial reported a significant prolongation of median progression-free survival (mPFS) of 28.4 months (5, 6). Furthermore, the majority of patients who initially show objective radiologic responses will eventually progress after PRRT (7–9). Despite progress in drug development, metastatic NET are fatal, with a median overall survival (OS) of 79 months (10). There is an unmet need to further improve on the efficacy of PRRT agents using novel therapeutic strategies.

Somatostatin receptor (SSTR) type 2 is a G-protein–coupled receptor expressed on the surface of epithelial and diffuse neuroendocrine system cells arising predominantly (90%) from a gastroenteropancreatic origin (3). Activation of SSTR by somatostatin has been shown to control hormonal uptake, cell metabolism, proliferation, and apoptosis (3, 11). Neoplastic GEP-NET are classified by 3 grades based on the mitotic rate and Ki-67 index (G1, <2 mitoses/10 HPF and Ki-67 < 3%; G2, 2–20 mitoses/10 HPF or Ki-67 3%–20%; G3, >20 mitoses/10 HPF or Ki-67 > 20%). The G1 to G2 GEP-NET represent 60% of all NET and are an emerging cause of cancer-related deaths in the United States (12, 13). Patients with G1 to G2 well-differentiated GEP-NET have an indolent course; however, resistance to therapy and progressive metastatic disease are inevitably fatal, primarily from liver failure, gastrointestinal obstruction, and carcinoid heart disease (14, 15). SSTR type 2 receptor expression in NET introduces a potential therapeutic sensitivity for a radiopharmaceutical, such as [¹⁷⁷Lu]lutetium-dotatate, which uses SSTR type 2 as a means to preferentially deliver a radioactive payload to neoplastic cells. [¹⁷⁷Lu]lutetium-dotatate directs β-particle–emitting ¹⁷⁷Lu to SSTR type 2–expressing neoplastic cells with high affinity, allowing it to enter cells and irradiate nuclear DNA, resulting in neoplastic cell death (16). The optimism surrounding current targeted radiopharmaceuticals, along with the potential to further enhance their efficacy, motivated the design of this phase I study to test the hypothesis that radiosensitization with an oral ribonucleotide reductase inhibitor would be safe and tolerable and provide an initial assessment of clinical efficacy.

Triapine (3-aminopyridine-2-carboxaldehyde thiosemicarbazone) acts as a radiosensitizer through its potent inhibition of ribonucleotide reductase (17–20)—the enzyme responsible for catalyzing the rate-limiting conversion of ribonucleotide diphosphates to deoxyribonucleotide diphosphates used in DNA synthesis and repair (21). After DNA damage, such as from irradiation, inhibitors of ribonucleotide reductase protract DNA damage repair and promote cell kill (22, 23). The antitumor activity of the ribonucleotide reductase inhibitor triapine in combination with irradiation in preclinical NET models was synergistic (24). Herein, we report the clinical findings of a first-in-human phase I study evaluating the combination of [¹⁷⁷Lu]lutetium-dotatate with increasing doses of oral triapine administered over four 8-week cycles in patients with well-differentiated GEP-NET (NCT04234568). The primary endpoints were to assess the safety of the [¹⁷⁷Lu]lutetium-dotatate–triapine combination and to determine the recommended phase II dose (RP2D) of the agents. Secondary endpoints included the evaluation of antitumor activity and pharmacokinetics (PK). Exploratory endpoints assessed [¹⁷⁷Lu]lutetium-dotatate biodistribution in the setting of ribonucleotide reductase inhibition, pharmacodynamic effects, and molecular genetic profiling of metastatic NET.

Patients and Methods

Patient eligibility

Patients 18 years or older with a diagnosis of measurable (RECIST 1.1) metastatic well-differentiated NET with a positive gallium-68 or copper-64 dotatate scan (within 6 months) and who had failed at least 1 prior systemic treatment—progressive disease based on RECIST criteria, version 1.1, and evidenced by CT/magnetic resonance imaging (MRI) obtained within 24 months from enrollment, including somatostatin analogs—were eligible. Long-acting somatostatin analogs were allowed to be continued if a patient reported a history of carcinoid syndrome and required them for control of their functional syndrome. Patients had no prior exposure to PRRT; an Eastern Cooperative Oncology Group performance status score of 0, 1, or 2; and adequate kidney, liver, and bone marrow function as determined by laboratory assessments. Key exclusion criteria included untreated central nervous system metastases, major surgery within 6 weeks, or chemotherapy and/or radiotherapy (i.e., irradiating >50% of bone marrow) within 4 weeks before entering the study. All persistent toxicities from prior treatments must have resolved to grade 1, except for alopecia.

The study was approved by the National Cancer Institute (NCI) Central Institutional Review Board and conducted in accordance with the International Conference on Harmonisation Good Clinical Practice Guidelines and the Declaration of Helsinki. This study was registered with clinicaltrials.gov (NCT04234568). Written informed consent was obtained from all patients before enrollment. Triapine was provided under an NCI Cooperative Research and Development Agreement with Nanopharmaceutics.

Study design and objectives

This was a multicenter, open-label, phase I study, enrolling patients from across 6 NCI Experimental Therapeutics Clinical Trials Network (ETCTN)–supported cancer centers in the United States. Patients received between 1 and 4 cycles (cycle length = 8 weeks) of study treatment—triapine at the assigned dose level (100, 150, or 200 mg) and [¹⁷⁷Lu]lutetium-dotatate (200 mCi) on day 1 of each cycle until disease progression or toxicity (Supplementary Fig. S1). On day 1 of each cycle, patients received triapine within 90 minutes after [¹⁷⁷Lu]lutetium-dotatate treatment and once daily for a total of 14 days. The study included a dose escalation part and a dose expansion part.

The primary objective of this trial was to determine the maximum tolerated dose (MTD) and the RP2D of triapine in combination with [¹⁷⁷Lu]lutetium-dotatate in patients with well-differentiated SSTR-positive GEP-NET.

The secondary objectives were to observe and record antitumor activity, determine the overall response rate, PFS, and OS. Exploratory objectives included evaluating oral triapine plasma PK and molecular characterization of our study cohort with the help of tumor tissue and circulating DNA-based next-generation sequencing.

Determination of MTD

In the dose-escalation part, patients were enrolled following a Bayesian Optimal Interval (BOIN) design to define the primary objectives of the study—MTD, RP2D, and PK profile of triapine added to the [¹⁷⁷Lu]lutetium-dotatate. Patients were enrolled in cohorts of 3, beginning at dose level 1 (100 mg). A target dose-limiting toxicity (DLT) rate of 25% was assumed, with dose escalation (λe) and de-escalation (λd) boundaries set at 0.197 and 0.298, respectively. Based on the BOIN design, dose escalation was conducted using increasing doses of triapine (Supplementary Table S1) until DLT and safety were determined.

MTD and RP2D were estimated using isotonic regression based on toxicity data from all patients; an isotonic regression curve plot (Supplementary Fig. S2) has been included as a supplementary document. Dose level 2 (150 mg) was selected as the RP2D for the dose expansion cohort of 16 patients to further characterize the safety profile. All patients were followed for 24 months from the time of enrollment or until death, whichever occurred first. Patients removed from the study for unacceptable adverse event(s) (AE) were followed until resolution or stabilization of the AE.

Safety

Safety was evaluated based on reported treatment-emergent AE, physical examination, vital signs, and laboratory test examinations. AE were monitored from the time of the first dose of [¹⁷⁷Lu]lutetium-dotatate–triapine until 24 months after discontinuation or the start of a new anticancer treatment. AE were classified using the Common Terminology Criteria for Adverse Events (version 5).

DLT was defined as grade 3 or greater hematologic or nonhematologic toxicity that was possibly, probably, or likely related to study treatment. The following were not considered DLT unless determined to be clinically significant and related to study treatments by the treating physician: grade 1 to 4 lymphopenia and grade 3 nausea, vomiting, or diarrhea lasting <72 hours and controlled by optimal antiemetic/antidiarrheal therapy. DLT was assessed at the end of cycle 1 (8-week period). All patients who received at least 1 dose of [¹⁷⁷Lu]lutetium-dotatate and at least 80% of the cohort-specific triapine doses during cycle 1 were evaluable for DLT criteria for dose-escalation/de-escalation decisions.

Antitumor activity

Tumor response was evaluated using contrast-enhanced CT (or MRI scan or noncontrast CT scan if contrast was contraindicated). Radiographic tumor assessments of the chest, abdomen, and pelvis were performed at baseline ≤4 weeks before the start of treatment and every 8 weeks after the start of [¹⁷⁷Lu]lutetium-dotatate–triapine treatment for patients in the dose-escalation part or every 12 weeks for patients in the dose expansion part until disease progression, study termination, or the final visit for patients without documented radiographic progression if clinically warranted. Changes in the size of target lesions were evaluated using RECIST version 1.1 (25), and the objective response rate (ORR), duration of response (DOR), PFS, and OS were calculated.

PK and pharmacodynamics

Blood samples from 14 patients were collected for triapine PK analysis in EDTA tubes before and at 0.5, 1, 1.5, 2, 3, 4, 6, and 8 hours after oral administration. Each blood sample was centrifuged at approximately 1,000 × g, and plasma was stored at −70°C or colder until analysis with an LC/MS-MS assay validated to FDA guidance (26). Plasma PK parameters were derived from the data by noncompartmental methods with WinNonlin (Certara), as before (27).

Correlative endpoints

For whole-exome sequencing (WES) data analysis, Burrows-Wheeler Aligner (RRID:SCR_010910) was used for sequence alignment, and somatic mutations were identified using the GATK4 (RRID:SCR_001876) workflow and MuTect2 (RRID:SCR_026692; bioRxiv 2019.12.02.861054) software. The copy-number calls were inferred by Sequenza (RRID:SCR_016662; ref. 28). The copy-number variation (CNV) based signature was estimated by performing nonnegative matrix factorization and profile comparison with existing Catalogue of Somatic Mutations in Cancer (RRID:SCR_002260) version 3 signatures using sigminer (29). For the gene expression analysis, alignment using STAR (RRID:SCR_004463) was performed; then the normalized counts were obtained by running RNA sequencing (RNA-seq) by expectation maximization (RRID:SCR_000262; ref. 30), tximport (RRID:SCR_016752; ref. 31), and DESeq2 (RRID:SCR_015687; ref. 32) in sequence. WES and circulating tumor DNA (ctDNA) markers (especially MEN1, DAXX, ATRX, TSC2, and PTEN) were determined centrally by the NCI National Clinical Laboratory. NET work (Frederick, MD) using the Illumina NovaSeq 6000 sequencing platform and established techniques, as elsewhere.

WES

Whole-exome libraries were prepared from 50 ng of genomic DNA sheared to 150 to 180 bp using a Covaris LE220 sonicator (Covaris, Inc.). Library construction was automated on a Revvity Sciclone G3 platform (Revvity, Inc.) using the KAPA HyperPrep Library Construction and Dual Index Kit (Roche Sequencing). Following end-repair, A-tailing, and adapter ligation, libraries were purified with AMPure XP beads (Beckman Coulter) and PCR-amplified (14 cycles) using KAPA HiFi polymerase. Amplified libraries were hybridized with SureSelectXT Human All Exon V6 RNA baits (Agilent Technologies) at 65°C for 16 hours. Captured DNA was enriched with Dynal MyOne Streptavidin T1 beads (Thermo Fisher Scientific) and reamplified (12 cycles) using Hercules DNA polymerase (Agilent). Final libraries were purified, analyzed for size (300–400 bp) using the Agilent TapeStation 4200, and quantified with the Quant-IT PicoGreen double-stranded DNA fluorescent quantitation kit (Thermo Fisher Scientific) on a SpectraMax M Fluorometer (Molecular Devices). Equimolar pooled libraries were created and diluted to 1.2 nmol/L for XP mode clustering on a NovaSeq S4 flow cell (Illumina, Inc.). Paired-end sequencing (2 × 150 bp) was performed on an Illumina NovaSeq 6000 system (Illumina, Inc.).

Whole-transcriptome sequencing

RNA-seq libraries were prepared using 70 to 100 ng of total RNA with the Illumina RNA Exome workflow (Illumina, Inc.) and performed on the Revvity Sciclone G3 automated liquid handler (Revvity, Inc.) using custom scripts. RNA was fragmented, followed by double-stranded cDNA synthesis and purification using AMPure XP beads (Beckman Coulter). cDNA was subjected to dA-tailing and adaptor ligation with Illumina TruSeq RNA Exome indexed adaptors, then purified with AMPure XP beads per the manufacturer’s protocol. Libraries were amplified via 15-cycle PCR and further purified using AMPure XP beads. Library quality and concentration were assessed using the Agilent TapeStation with D1000 ScreenTape. Two hundred nanograms of each amplified library underwent 2 rounds of hybridization with biotinylated DNA baits, captured using streptavidin magnetic beads, then washed and eluted (following the Illumina protocol). Target-enriched libraries were purified with AMPure XP beads, amplified by 10-cycle PCR, and purified again using AMPure XP beads. Final libraries were quantified using the Agilent TapeStation and Molecular Devices SpectraMax M fluorometer. Equimolar pooling was performed; pooled libraries were diluted to 1.2 nmol/L for XP clustering on the NovaSeq S4 flow cell. Sequencing was performed on the Illumina NovaSeq 6000 (Illumina, Inc.) using 2× 150 bp paired-end reads.

Plasma ctDNA profiling

Library preparation from 15- to 20-ng patient plasma cell–free DNA was performed using the Illumina TruSight Oncology 500 ctDNA v2 assay (Illumina, Inc.), a hybrid capture–based next-generation sequencing (20) panel designed for comprehensive genomic profiling of 523 cancer-related genes. Unique molecular identifiers (UMI) were incorporated to enhance error correction and enable accurate detection of low-frequency variants. Hybrid capture enrichment was followed by PCR amplification and quality control using fragment analysis and quantification via qPCR or fluorometric methods. Sequencing was conducted on an Illumina NovaSeq 6000 platform using 2 × 150 bp paired-end reads. Data analysis was performed using the TSO500 ctDNA Local App v2.2 on the Illumina Dragen server, which includes adapter trimming, alignment to the hg19 reference genome, UMI-based error correction, and variant calling (SNVs, indels, CNVs, and gene translocations).

Statistical analysis

The safety population included all patients who received 1 or more doses of [¹⁷⁷Lu]lutetium-dotatate–triapine. The efficacy population included patients who received 1 or more doses of [¹⁷⁷Lu]lutetium-dotatate–triapine and at least 1 postdose tumor assessment. All safety and efficacy analyses were descriptive in intent, with no statistical inference drawn from the data. The ORR was measured as the best overall response recorded from the start of the treatment until disease progression/recurrence. PFS was measured from the start date of [¹⁷⁷Lu]lutetium-dotatate–triapine through the date of first progression of disease or death. OS was measured from the [¹⁷⁷Lu]lutetium-dotatate–triapine start date through the date of death and was estimated by the Kaplan–Meier method (33). DOR was measured from the time interval between the onset of a treatment response and the subsequent progression of disease or death.

Results

Patient demographics and baseline characteristics

From August 2020 to March 2022, 39 patients with well-differentiated NET were screened for the study. Of the 39 patients, 31 met the eligibility criteria and were enrolled in the study and evaluable for safety analysis. Out of the 31 enrolled patients, 15 were in the dose-escalation cohort and 16 in the dose-expansion cohort.

The median and average age of the enrolled study population was 54 years (range, 25–74). All patients enrolled in the study presented with metastatic well-differentiated NET. Based on the study design, all eligible participants had progressed on at least 1 prior systemic treatment. Consequently, none of the enrolled patients were treatment-naïve. Out of the 31 enrolled patients, 11 (35.48%) had received 1 prior line of therapy, 13 (41.94%) had received 2 prior lines of therapy, and 5 (16.13%) had received 3 or more prior lines of therapy. This information was unavailable for 2 participants. Before enrolling in the study, the participants received systemic treatment (e.g., chemotherapy, targeted therapy, hormonal therapy). Of those who received chemotherapy, the majority of the participants received oral chemotherapy, capecitabine and temozolomide (CAPTEM). Study patients with prior exposure to targeted therapies received mostly everolimus and rarely pazopanib, whereas those who received liver-directed therapy included modalities such as bland transarterial chemoembolization, transcatheter arterial chemoembolization, and radiofrequency ablation.

Patient demographic and baseline characteristics are summarized in Table 1. Demographically, patients were generally representative of the intended population. The representativeness of the study participants is shown in Supplementary Table S2.

Table 1.

Patient demographics and baseline characteristics of all patients who enrolled.

Characteristic	Escalation (n = 15)	Expansion (n = 16)	Total (n = 31)
Sex, n (%)
Female	3 (20)	10 (62.5)	13 (41.9)
Male	12 (80)	6 (37.5)	18 (58.1)
Age group, n (%)
18–65 years	12 (80)	12 (75)	24 (77.4)
>65 years	3 (20)	4 (25)	7 (22.6)
Mean (SD)	51.7 (12.1)	57.6 (12.2)	54.7 (12.3)
Median	51	59	54
Range	25–69	35–74	25–74
Race, n (%)
White	14 (93.3)	14 (87.5)	28 (90.3)
Black	1 (6.7)	1 (6.3)	2 (6.5)
Unknown	​	1 (6.3)	1 (3.2)
ECOG and Karnofsky, n (%)
0	11 (73.3)	10 (62.5)	21 (67.7)
1	4 (26.7)	6 (37.5)	10 (32.3)
Tumor grade, n (%)
Well-differentiated (grade 1)	15 (100)	15 (93.8)	30 (96.8)
Moderately differentiated (grade 2)	​	1 (6.3)	1 (3.2)
Primary tumor site location, n (%), N = 31
Small intestine			9 (29.03)
Large intestine			4 (12.90)
Pancreas			11 (35.48)
Rectal			2 (6.45)
Lung			1 (3.23)
Unknown			4 (12.90)
Prior anticancer treatments, n (%), N = 31			​
Chemotherapy			11 (35)
Somatostatin analog			25 (81)
Targeted therapy			11 (35)
Liver-directed			5 (16)
Prior number of lines of therapy, n (%), N = 31
One	11 (35.48)
Two	13 (41.94)
Three or more	5 (16.13)

Abbreviations: ECOG, Eastern Cooperative Oncology Group; SD, standard deviation.

Treatment exposure

For each of the 3 dose levels of triapine studied [100 mg (n = 5), 150 mg (n = 25), or 200 mg (n = 1)], all (100%) patients received at least 1 dose of [¹⁷⁷Lu]lutetium-dotatate, 77% received at least 2 doses, 68% received at least 3 doses, and 61% received all 4 doses (Table 2).

Table 2.

¹⁷⁷Lu‐dotatate exposure.

Patients who received [177Lu]lutetium-dotatate (N = 31)
Number of [177Lu]lutetium-dotatate administrationsa	Number of patients, n (%)
At least 1 dose	31 (100)
At least 2 doses	24 (77)
At least 3 doses	21 (68)
4 doses	19 (61)

^aExcludes patients who did not take enough to be evaluable.

Determination of MTD and RP2D

All 31 patients were enrolled in either dose-escalation or dose-expansion cohorts using a BOIN design. During the dose-escalation part, 3 patients received 100 mg of triapine OP once a day plus 200 mCi of lutetium-177. As no DLT were observed in the first 3 patients enrolled at dose level 1 (triapine 100 mg), 3 patients were then enrolled at dose level 2 (triapine 150 mg). One out of 3 patients who received 150 mg triapine orally once a day plus [¹⁷⁷Lu]lutetium-dotatate reported a DLT; therefore, 3 additional patients were enrolled at the 150-mg triapine dose level. None of these 3 patients reported a DLT, and consequently, a patient was treated at dose level 3 (triapine 200 mg). The first patient treated at the [¹⁷⁷Lu]lutetium-dotatate–triapine (200 mg) dose had a grade 5 acute exacerbation of carcinoid heart failure. This prompted dose level de-escalation. After de-escalation, a patient was treated at dose level 2 (triapine 150 mg) once a day plus [¹⁷⁷Lu]lutetium-dotatate. This patient reported a DLT (grade 4 neutropenia), so 4 additional patients were enrolled at dose level 2 (triapine 150 mg).

Based on the overall safety and PK data from 11 patients treated with [¹⁷⁷Lu]lutetium-dotatate (200 mCi) plus dose level 2 (triapine 150 mg), dose level 2 was selected as the recommended dose for the expansion phase (Fig. 1). Dose expansion followed the BOIN design; 14 out of 16 patients were treated at dose level 2 with 150 mg of triapine orally once a day plus lutetium-177, and 2 patients were treated at dose level 1 with 100 mg. The triapine (150 mg) dose was reaffirmed as RP2D after the completion of the expansion cohort.

Figure 1.

CONSORT diagram. n = number of patients; dose levels 1, 2, and 3 are of the study investigational product, triapine.

Safety

All patients who received study drugs have been included in the safety analysis. All 31 patients enrolled in the study experienced at least 1 AE, the majority of which were grade 1 or 2 severity. The most common AE observed across the entire study cohort are summarized in Table 3; they were anemia (87%), lymphopenia (87%), neutropenia (77%), and thrombocytopenia (71%). Overall, grade ≥3 treatment-related AE were observed in 77% of patients (24): anemia (23%), nausea and vomiting (3%), lymphopenia (58%), neutropenia (35%), and white blood cell count decreased (39%). Lymphopenia was considered clinically insignificant, as it did not result in infections associated with lymphocytopenia, and most patients experiencing cytopenia recovered within 2 weeks (Supplementary Fig. S3; Supplementary Table S3). Eighteen out of 25 patients treated at RP2D (dose level 2) reported the following grade ≥3 treatment-related AE: anemia (20%), lymphopenia (56%), neutropenia (40%), thrombocytopenia (28%), and white blood cell count decreased (44%).

Table 3.

Most common adverse events.

Adverse events (n = 31)	All grades	≥ Grade 3
	Number of patients (%)
Blood disorders
Anemia	27 (87)	10 (32)
Lymphopenia	27 (87)	20 (65)
White blood cell decreased	24 (77)	16 (52)
Platelet count decreased	22 (71)	8 (26)
Neutrophil count decreased	19 (61)	12 (39)
Gastrointestinal disorders
Nausea	20 (65)	1 (3)
Vomiting	10 (32)	1 (3)
Diarrhea	6 (19)	0
General disorders
Fatigue	15 (48)	0
Edema limbs	7 (23)	0
Metabolism and nutritional disorders
Anorexia	7 (23)	0
Hypocalcemia	9 (29)	0
Hypokalemia	8 (26)	1 (3)
Hyperglycemia	7 (23)	0
Investigations
Blood bilirubin increased	14 (45)	1 (3)
Blood lactate dehydrogenase increased	12 (39)	0
Aspartate aminotransferase increased	9 (29)	0
Alkaline phosphatase increased	8 (26)	0
Alanine aminotransferase increased	7 (23)	0
Creatinine increased	7 (23)	0

Across the study, 9 out of 31 patients evaluable for safety experienced DLT, which resulted in IP discontinuation. However, these patients continued standard of care PRRT. Of the 9 patients, 1 experienced grade 3 anemia, 1 experienced grade 5 cardiac arrest (unrelated to the investigational product), and 7 experienced grade 4 neutropenia.

Two (6%) at least possibly related deaths attributed to [¹⁷⁷Lu]lutetium-dotatate–triapine treatment were reported: One (6%) patient in the dose-escalation/DL3 step had an acute exacerbation of carcinoid heart failure occur at the [¹⁷⁷Lu]lutetium-dotatate (200 mCi)–triapine (150 mg) dose, and 1 (6%) patient in the dose-expansion/DL2 step had posttherapy grade 5 bowel perforation occur at the [¹⁷⁷Lu]lutetium-dotatate (200 mCi)–triapine (100 mg) dose. Out of the 25 patients treated at RP2D (dose level 2), 11 (44%) patients reported grade 4 serious AE: lymphopenia (8%), thrombocytopenia (16%), and neutropenia (28%).

Of note, the majority of grade 3 to 4 hematologic AE were noted during the expected nadir following PRRT and self-resolved within 2 weeks, allowing patients to continue treatment per protocol. No significant triapine drug-related symptomatic dyspnea or methemoglobinemia was reported in the 25 NET patients treated at the RP2D.

Activity

Overall, 28 out of the 31 enrolled patients were evaluable for ORR. Of these, 6 patients [21.42%; 95% confidence interval (CI), 8.3–40.9] achieved confirmed objective response (Table 4; Fig. 2), including no patients with complete responses and 6 (21.42%) patients with partial responses. Median DOR was 20 months (95% CI, 5–23 months), and the disease control rate was 89% (95% CI, 71%–97%). Nineteen patients (67.85%) had stable disease, 6 (21.42%) had partial response, and 3 (10.71%) had progressive disease as their best response. At the time of the data cutoff for the primary analysis on December 9, 2024, 7 events of disease progression or death had occurred in the [¹⁷⁷Lu]lutetium-dotatate–triapine cohorts; 4 events were seen in dose escalation and 3 events were seen in dose expansion. As described before, 2 patients experienced deaths: 1 from a cardiac event deemed unrelated to the investigational product and the other from treatment-emergent colon perforation. The patient with colon perforation had significant disease burden in the abdomen with omental caking. Of note, 1 patient developed AML. The patient progressed after 2 years of follow-up and was enrolled in an experimental alpha PRRT study. Acute myeloid leukemia (AML) developed after 2 cycles of subsequent alpha PRRT treatment. A 2% risk of AML is described after [¹⁷⁷Lu]lutetium-dotatate. In this case, attribution to triapine is difficult due to sequential exposure to subsequent alpha PRRT. The estimated rate of PFS at month 20 was 76.83% (95% CI, 60.5%–93.5%) for the evaluable patient cohort. mPFS has not been reached at the time of data cutoff (23.7-month median follow-up). At the time of data cutoff, data were not sufficiently mature to provide an estimate of median OS.

Table 4.

Best overall responses.

Best overall response Number of patients evaluable = 28	Total n (%)
Complete response	0
Partial response	6 (21.42)
Stable disease	19 (67.85)
Progressive disease	3 (10.71)
Time to response, months; median (range)a	3.5 (1–14)

NOTE: Three patients were nonevaluable. All data are expressed as number of patients (%).

^aTime to response is defined as months from treatment start to first documentation of response or partial response. Time to response includes patients with confirmed response of partial response.

Figure 2.

A, The waterfall plot of best overall response in the evaluable patient population. We have reported the PFS for 27 patients as we did not have the follow-up tumor size data for 1 patient out of 28 evaluable patients. B, The results of the Kaplan–Meier analysis of mPFS for all 28 patients, which had not been reached at the time of the study cutoff point. In B, the + refers to the data censored at the last time the patient was known to be alive and without disease progression, and the solid blue refers to the 95% Hall–Wellner band.

PK and exposure response

PK data were available for 14 evaluable patients enrolled in the dose-escalation phase (Supplementary Fig. S4; Supplementary Table S4). Triapine exposure increased with oral dose, and between-patient variability was as expected for an oral agent.

PK and pharmacodynamics

Triapine PK parameter values observed in this trial with oral administration were comparable with previous reports in which triapine was administered intravenously (34). The triapine exposure in 4 patients dosed with 150 mg of the previous formulation was C_max 759 (2.31) μg/L and AUC 1743 (1.95) μg/L.hour (3.89 µmol/L and 8.93 μmol/L.hour), respectively; geometric mean (SD) from (34). Although exposure is similar, variability seems smaller with the current oral formulation. Of note, none of the patients evaluated for PK were current smokers, a habit that is known to result in the induction of CYP1A2 (35), which is a major contributor to triapine metabolism (36). The previously established MTD in patients with advanced-stage solid cancers was 150 mg twice a day on days 1 to 3, 8 to 10, and 15 to 17 of a 28-day cycle (34). Clinically, our MTD was defined as 150 mg once a day on days 1 to 14 every 8 weeks. Given the relatively short half-life of 1.7 hours, there is no accumulation with daily dosing, and one may compare the 2 MTD for dose, in which they are identical in being 150 mg and exhibit similar PK. Considering overall dosing, the previously established MTD, which was 18 doses over 17 days, is very similar to the current MTD of 14 doses administered over 14 days. In conclusion, the current MTD is very similar to the previously established MTD in terms of per-dose PK, as well as the overall dose administration.

Exploratory biomarkers

In the study, out of the 23 analyzable patients, 3 were found to have a MEN1 mutation, which is the key mutation that leads to MEN1 syndrome. It is known that 80% of people with MEN1 syndrome will develop GEP-NET by the age of 80 (33, 34). The corresponding survival plot stratified by MEN1 mutation in this cohort shows a significant difference (P value < 0.05; Supplementary Fig. S5).

For the CNV result derived from Sequenza, patient IL036-0025 was found to have a similar copy-number profile resembling the COSMOC CN17 signature, which represents homologous recombination deficiency (HRD; Supplementary Fig. S6; Supplementary Table S5).

In the dose-escalation cohort, 1 of the patients with carcinoid heart disease was found to have low HRD. We hypothesize that low HRD is associated with the patient’s exceptional response of greater than 50% NET lysis by day 8 after therapy. Further details are provided in the supplement.

Discussion

PRRT has been instrumental in transforming the therapeutic landscape for patients with advanced, well-differentiated NET. The introduction of [¹⁷⁷Lu]lutetium-dotatate has provided durable disease control and meaningful improvements in PFS for patients with SSTR-expressing disease, firmly establishing PRRT as a cornerstone of systemic therapy in this population. Despite these advances, a substantial proportion of patients eventually experience disease progression, underscoring the need for next-generation PRRT strategies aimed at improving the depth and durability of response. Accordingly, contemporary efforts in the field are focused on the development of novel radiopharmaceuticals, alternative radionuclides, and rational combination approaches designed to enhance therapeutic efficacy without incurring unacceptable toxicity.

One such strategy involves combining theranostic agents with biologically informed radiation sensitizers to potentiate DNA damage induced by targeted radionuclide therapy. Prior experience with PRRT combinations has included empiric chemotherapy pairings (e.g., capecitabine-based regimens) and targeted radiosensitizers such as PARP inhibitors; however, randomized data have not shown consistent improvements in radiographic response or PFS, and these approaches may be limited by toxicity or lack of a selective mechanistic rationale (37). In contrast, the combination of triapine, a potent ribonucleotide reductase inhibitor, with [¹⁷⁷Lu]lutetium-dotatate is grounded in preclinical data demonstrating impaired DNA damage repair and enhanced radiation sensitivity. By targeting a critical pathway involved in deoxynucleotide synthesis and DNA repair, triapine has the potential to amplify PRRT-mediated cytotoxicity while preserving an acceptable safety profile.

It is important to highlight that although the development of therapy-related myeloid neoplasms, including myelodysplastic syndrome/AML, is likely multifactorial—especially in the context of prior and subsequent therapies—the possibility that the combination strategy may contribute to an increased risk cannot be excluded. Continued long-term follow-up in this study, as well as careful monitoring in future studies, will be essential to better characterize the risk of delayed hematologic toxicity.

The results of this phase I study provide early clinical support for this biologically driven combination strategy and suggest that radiosensitization with triapine may represent a viable approach to enhancing the effectiveness of PRRT without compromising tolerability. These findings warrant further evaluation in larger disease-specific studies to better define efficacy, optimal dosing, and patient selection. At present, the triapine and [¹⁷⁷Lu]lutetium-dotatate combination is being investigated in a phase II randomized controlled trial being conducted at more than 14 sites across the United States and has completed enrollment of more than 100 patients. Results from this phase II trial are expected in the next 12 to 24 months (NCT 05724108).

Conclusion

Based on safety and PK, a dosing regimen of [¹⁷⁷Lu]lutetium-dotatate (200 mCi) every 8 weeks plus triapine (150 mg) daily on days 1 to 14 for 4 cycles was selected for further clinical development. Asymptomatic hematologic laboratory findings were noted, largely owing to capturing hematologic data at expected nadir time points, with the majority of patients having rapid count recovery. Continued long-term follow-up in this study, as well as careful monitoring in future studies, will be essential to better characterize the risk of delayed hematologic toxicity. mPFS has not been reached at the time of data cutoff (23.7 months median follow-up), and ORR is 21%. ETCTN 10558, a randomized phase II trial, is ongoing at more than 14 sites in the United States and is comparing the efficacy and safety of the [¹⁷⁷Lu]lutetium-dotatate–triapine combination versus standard-of-care [¹⁷⁷Lu]lutetium-dotatate monotherapy. This synergistic radiosensitizer strategy holds promise for enhancing the efficacy of theranostic agents and may have broader applications across oncology.

Supplementary Material

Supplementary Figure S1

Phase 1 Treatment Schema

Supplementary Figure S2

Isotonic regression curve

Supplementary Figure S3

Cytopenia plotted over time.

Supplementary Figure S4

Pharmacokinetic profiles of triapine

Supplementary Figure S5

Survival plot stratified by AGGF1 and MEN1

Supplementary Figure S6

Genomic signatures

Supplementary Table S1

Dose Escalation Schedule

Supplementary Table S2

Representativeness of study participants

Supplementary Table S3

Grade 3 or higher cytopenia raw data

Supplementary Table S4

Triapine geometric mean (SD) plasma pharmacokinetic parameters after PO administration

Supplementary Table S5

Normalized exposure matrix from Sigminer

Data Availability

Sequencing data (WES and RNA-seq) and derived somatic mutation and copy-number results have been deposited in the Database of Genotypes and Phenotypes under accession phs004766 (controlled access). Additional data are available from the corresponding author on request.

Authors’ Disclosures

A. Chauhan reports personal fees from Novartis and other support from Nanopharmaceutic during the conduct of the study; personal fees from ITM, Boehringer Ingelheim, Sanofi, Ipsen, Tersera, Curium, and Crinetics; and other support from Seneca Therapeutics, EMD Serono, Bristol Myers Squibb, and Clovis outside the submitted work. B. Konda reports grants from RayzeBio, Fore Biotherapeutics, Systimmune, Inc., Eisai, and Merck and personal fees from Exelixis outside the submitted work. H.P. Soares reports other support from ITM, Exelixis, Novartis, Curium, and Ipsen outside the submitted work. D. Li reports grants and personal fees from AstraZeneca and personal fees from AbbVie, Bayer, Boehringer Ingelheim, Coherus, Eisai, Elevar, Exelixis, Genentech, Ipsen, Jazz, Merck, Merus, Ono, Kite, Sanofi, Tersera, Camurus, Novartis, and Crinetics outside the submitted work. T.R. Halfdanarson reports personal fees from Exelixis, Curium, Sanofi, and Boehringer Ingelheim and grants from Rezolute, ITM, RayzeBio, Novartis, Crinetics, Camurus, Exelixis, and Perspective Therapeutics outside the submitted work. H.A. Jacene reports grants from Blue Earth Diagnostics and Lantheus, other support from Luminance Biosciences, and personal fees from Cambridge Publishing outside the submitted work, as well as research support from NIH (NIH R01CA235589-03; NIH/NCI 1R01CA251803-01A1). L.B. Anthony reports grants and nonfinancial support from RayzeBio during the conduct of the study. S.M. Arnold reports grants from the NCI during the conduct of the study. No disclosures were reported by the other authors.

Authors’ Contributions

A. Chauhan: Conceptualization, resources, data curation, software, formal analysis, supervision, funding acquisition, validation, investigation, visualization, methodology, writing–original draft, project administration, writing–review and editing. H.L. Weiss: Data curation, formal analysis, supervision, visualization. C. Kunos: Supervision, methodology, writing–review and editing. R. Jayswal: Data curation, software, formal analysis. B. Konda: Resources, supervision, investigation. H.P. Soares: Resources, supervision, investigation. D. Li: Resources, supervision, investigation. T.R. Halfdanarson: Resources, data curation, software, supervision, investigation, visualization, methodology, writing–review and editing. M.F. Mulcahy: Resources, supervision, investigation. H.A. Jacene: Resources, supervision, investigation, writing–review and editing. A. Khurana: Resources, supervision, investigation. R. Modak: Data curation, formal analysis, visualization, writing–original draft, writing–review and editing. D. Yan: Data curation, software, formal analysis, validation. J.H. Beumer: Conceptualization, resources, formal analysis, supervision, validation, investigation, methodology, writing–original draft, writing–review and editing. S.M. Christner: Data curation, formal analysis, methodology. P.I.-F. Wu: Resources, software, formal analysis, visualization. B. Das: Conceptualization, supervision, validation, investigation. L.V. Rubinstein: Investigation. S.D. Gore: Conceptualization, resources, data curation, software, formal analysis, supervision, funding acquisition, validation, investigation, visualization, methodology, writing–original draft, project administration, writing–review and editing. S.P. Ivy: Conceptualization, resources, data curation, software, formal analysis, supervision, funding acquisition, validation, investigation, visualization, methodology, writing–original draft, project administration, writing–review and editing. E.C. Kohn: Conceptualization, resources, data curation, software, formal analysis, supervision, funding acquisition, validation, investigation, visualization, methodology, writing–original draft, project administration, writing–review and editing. W.E. Carson: Conceptualization, resources, data curation, software, formal analysis, supervision, funding acquisition, validation, investigation, visualization, methodology, writing–original draft, project administration, writing–review and editing. L.B. Anthony: Conceptualization, resources, data curation, software, formal analysis, supervision, funding acquisition, validation, investigation, visualization, methodology, writing–original draft, project administration, writing–review and editing. S.M. Arnold: Conceptualization, resources, data curation, software, formal analysis, supervision, funding acquisition, validation, investigation, visualization, methodology, writing–original draft, project administration, writing–review and editing.