A Phase 1b Dose Escalation Trial of NC-6300 (Nanoparticle Epirubicin) in Patients with Advanced Solid Tumors or Advanced, Metastatic, or Unresectable Soft-tissue Sarcoma

Anthracyclines are widely used in the treatment of many hematologic and solid tumor malignancies, including sarcomas. Although doxorubicin remains the most commonly used anthracycline, the potential for significant hematologic and cardiac toxicity exists. Epirubicin, a 4′-epimer of doxorubicin, was specifically synthesized to address cardiac toxicity concerns. Despite its use and other anthracycline formulations, many individuals still suffer from severe rate-limiting side effects. Thus, there remains an unmet medical need for a safer anthracycline agent.

NC-6300 is a novel formulation of epirubicin that utilizes cutting-edge nanotechnology. Epirubicin hydrochloride is covalently bound to a block-copolymer with a poly (ethylene glycol; PEG)-poly(l-aspartic acid) framework via an acid-labile hydrazine linkage that enables the selective release of epirubicin into the acidic environment of tumor tissue and late endosomes in tumor cells. The epirubicin-conjugated block-copolymer spontaneously forms micelles (40–80 nm in diameter) in aqueous solution. Selective tumor accumulation of nanoparticles has been attributed to the enhanced permeability and retention (EPR) effect. Structures or molecules ranging from 90 to 130 nm in diameter tend to penetrate tumor tissues through tumor vessel pores not frequently observed in nontumor vascular smooth muscle layers (1, 2). Consequently, macromolecules can accumulate in tumor tissues and remain for longer periods of time. Immature growth of lymphatics associated with tumors also contributes to less effective drainage of macromolecules from tumors (3). The EPR effect enables NC-6300 to deliver epirubicin more efficiently to tumor tissue than to nontumor tissue.

In NC-6300 nonclinical studies, extensive tumor tissue distribution of NC-6300 was observed when compared with cardiac and other nontumor tissue (4, 5). In vivo, cardiac function assessed by echocardiography was well maintained. When compared with equal doses of conventional epirubicin, NC-6300 showed significant tumor growth inhibition as well as prolonged survival (4, 5).

A first-in-human trial of NC-6300 was conducted in solid tumor subjects in Japan by Mukai and colleagues (6). The MTD was evaluated (at conventional epirubicin-equivalent doses) from 15 to 225 mg/m² by a continuous reassessment method (CRM). An MTD of 170 mg/m² was identified. Importantly, NC-6300 was administered at doses exceeding those of conventional epirubicin while still maintaining tolerability. Finally, the spectrum of adverse events (AEs) did not differ between recipients of NC-6300 and those who received conventional epirubicin.

We now report an open-label, nonrandomized, phase 1b dose-escalation study assessing the safety, preliminary activity, pharmacokinetics (PK), and quality-of-life (QOL) following administration of NC-6300 monotherapy in patients with advanced, metastatic, or unresectable solid tumors, including soft-tissue sarcomas.

Adult subjects (age ≥ 18 years) that met the following criteria were recruited: a histologically or cytologically confirmed diagnosis of advanced solid tumor, including soft-tissue sarcoma, that had relapsed or was refractory to standard therapy; Eastern Cooperative Oncology Group (ECOG) performance status, defined as 0 or 1; adequate bone marrow reserve (absolute neutrophil count ≥1.5 × 10⁹/L; platelet count ≥100 × 10⁹/L; and hemoglobin level ≥10 g/dL), cardiac function [left ventricular ejection fraction (LVEF) ≥50%; QTcF ≤470 ms; no history of uncontrolled cardiac dysrhythmia, hepatic dysfunction, or renal dysfunction].

Subjects who had been treated previously with an anthracycline were eligible if cumulative exposure had not exceeded acceptable limits, defined as ≤375 mg/m² doxorubicin or liposomal doxorubicin, or ≤675 mg/m² epirubicin. Subjects who had palliative surgery and/or radiation treatment within 30 days prior to screening, or had active central nervous system or brain metastasis at the time of screening were excluded.

The study protocol was approved by an institutional review board or independent ethics committee at each study center, and was performed in accordance with ethical principles that have their origin in the Declaration of Helsinki, International Council for Harmonization Good Clinical Practice, and all applicable regulations. All subjects provided written informed consent prior to study participation. This trial is registered with ClinicalTrials gov, identifier NCT03168061.

This open-label, nonrandomized, phase 1b dose-escalation trial of NC-6300 monotherapy was conducted at seven sites in the United States. The primary endpoint was determination of the MTD and RP2D of NC-6300. The secondary endpoints included evaluation of antitumor activity, overall safety, and tolerability of NC-6300 and assessment of changes in health-related QOL using the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30. Given the recognized cardiac toxicity of anthracyclines, LVEF was measured by echocardiography or by multigated acquisition scanning at Screening, day 1 of every cycle and end of treatment visit. PK of NC-6300 was assessed as an exploratory endpoint. Images of tumors by CT or MRI were assessed according to RECIST Version 1.1. AEs were graded using the NCI Common Terminology Criteria for Adverse Events Version 4.03 (CTCAE v4.03). Patient visits took place at Screening, days 1, 8, and 15 of every cycle and end of treatment. Hematology and biochemistry blood test were done every patient visit.

Dose-limiting toxicity (DLT) was defined as any occurrence of the following conditions during the first treatment cycle, graded in accordance with NCI CTCAE version v4.03: grade 2 increase in serum aspartate aminotransferase or alanine aminotransferase in combination with a grade 2 increase in serum bilirubin; febrile neutropenia; grade 4 neutropenia lasting more than 7 days and unresponsive to growth factor administration; grade 4 thrombocytopenia; grade 3 thrombocytopenia, with clinically significant bleeding or requiring a platelet transfusion; grade 3 or grade 4 nonhematological toxicity (excluding alopecia and grade 3 nausea, vomiting, diarrhea, and electrolyte imbalances lasting less than 48 hours or with suboptimal prophylactic); grade 3 nausea, vomiting, diarrhea, or electrolyte imbalances lasting >48 hours despite optimal prophylactic; grade 4 vomiting, diarrhea, or electrolyte imbalances of any duration despite optimal prophylactic; ≥grade 3 hypersensitivity reaction and dosing delay >21 days due to toxicity.

NC-6300 was administered intravenously over a period of at least 10 minutes on day 1 of each 21-day treatment cycle until disease progression, intolerance, or withdrawal of consent. Prophylaxis for chemotherapy-induced nausea and vomiting was allowed but not mandated. Prophylactic antiallergic agents and glucocorticoids were recommended according to standard-of-care and institutional treatment protocols. Use of prophylactic growth factor support was allowed but not required.

Subject enrollment began as a single-subject run-in that continued until either 1 subject experienced a grade 2 or higher treatment-related AE or until the 170 mg/m² dose level was reached. After the single-subject run-in period, four subjects were enrolled at each dose level predicted by the CRM until the MTD was determined. Subjects were enrolled sequentially from 125 to 215 mg/m² until a DLT was observed or until subjects were treated at 230 mg/m² for one cycle without a DLT. Dosing cohorts of 125, 150, 170, 185, 200, 215, and 230 mg/m² were planned.

A withdrawal criterion associated with cardiac toxicity was changed during the study from “Decrease in LVEF by ≥10% as an absolute value compared with Baseline, or LVEF falls below 50%” to “Decrease in LVEF by ≥10% as an absolute value compared with Baseline, and LVEF falls below 50%.”

A Bayesian N-CRM model was applied, wherein model-based estimates of the probability of a DLT were classified into four mutually exclusive categories: underdosing: p∈ (0, 0.20], targeted toxicity: p∈ (0.20, 0.33], excessive toxicity: p∈ (0.33, 0.60], unacceptable toxicity: p∈ (0.60, 1.00] (7).

The dose-response relationship was estimated through a two-parameter logistic regression model, wherein the log odds of the probability of DLT at dose i was modeled as a simple linear equation using the adjusted dose strength of dose d_i: ln (d_i/d_r)

where P_i is the probability of toxicity at dose level i; d_i is the dose in mg at level i; d_r is the reference dose, which is defined as median dose in this study; α is the logit of toxicity at the reference dose; β is the parameter to determine the slope of the curve. An informative bivariate normal prior for the model parameters [α, log(β)] was obtained. On the basis of preclinical data and complete clinical studies, the starting dose for NC-6300 was determined to be 125 mg/m², and subsequent doses included 150, 170, 185, 200, 215, and 230 mg/m². The median dose of 185 mg/m² was selected as reference dose for the prior. A noninformative prior was selected per the methods of Neuenschwander and colleagues (2008; ref. 7).

The probability of an unacceptable toxicity at the lowest dose was taken to be 5% and the probability under dosing at the highest dose was taken to be 5%. Assuming minimally informative unimodal beta distributions for each dose, prior median toxicities were estimated for the lowest and highest doses. Median toxicities for the intervening doses were assumed to be linear in log dose on the logit scale. The uninformative prior was generated by FACS. All modeling and calculations were done with SAS (version 9.4) IML software.

Following a single-subject run-in at the 125 mg/m² dose, subjects were enrolled sequentially in cohorts of n = 4. The study was designed to stop when: seven cohorts had completed, or two cohorts had been treated at the MTD, or no dose level assured posterior probability of excessive or unacceptable toxicity at ≤ 25%.

The dose level with the greatest posterior probability of targeted toxicity that assured posterior probability of excessive or unacceptable toxicity at ≤ 25% was determined to be the MTD. Analyses of all endpoints were conducted using the safety analysis set (defined as all subjects enrolled and who received ≥ one dose of NC-6300). Descriptive statistics were used to summarize select demographic and safety data.

Plasma samples were collected from all subjects for measurement of total epirubicin, released epirubicin, and epirubicinol (metabolite) by validated LC/MS-MS. In cycle 1, plasma samples were collected at predose (0 hour), at the end of NC-6300 infusion (10 minutes) and at 0.5, 1, 2, 4, 8, 24, 48, 72, 168, and 336 hours after start of infusion. From cycle 2 onward, samples were taken only at predose and at the end of infusion. Plasma concentration data were imported to the Phoenix WinNonlin software (v8.2, Certara) for calculation of PK parameters in the noncompartmental model.

Between August 16, 2017 (the first subject treated) and November 13, 2018 (the last subject treated), 29 enrolled subjects received at least one dose of NC-6300. Demographics are shown in Table 1. Overall, 55% of subjects were men and 66% were white. Median age was 61 years (range, 24–79 years). Fifty-nine percent of the enrolled subjects had a soft-tissue sarcoma. Among the soft-tissue sarcoma population, the most common subtype was leiomyosarcoma (n = 6, 35%) followed by undifferentiated, unclassified sarcoma (n = 3, 18%). The median number of prior anticancer treatments was two (range, 0–9), with details as follows: zero in three subjects (10%), one in six subjects (21%), two in six subjects (21%), three in four subjects (14%), and four or more in 10 subjects (35%). Ten subjects (35%) had received prior anthracyclines before enrolling in this study.

Eight DLT events were observed in six subjects at 170 mg/m² or above. Among six subjects, grade 4 thrombocytopenia and grade 4 febrile neutropenia were the most common events (Table 2). All DLTs were resolved by dose reduction or no action taken during the treatment period, except for an event of grade 4 thrombocytopenia at 215 mg/m². The MTD and the RP2D were determined to be 185 mg/m² and 150 mg/m², respectively. No DLTs were observed at the MTD or the RP2D.

All 29 subjects had at least one AE. No treatment emergent deaths were observed. AEs, across all cycles, for all grades and grades 3–4 in the entire population, in the MTD population, and in the RP2D are reported in Table 2. The most common AEs at all grades (n = 29) in the entire population included nausea (62%), neutropenia (62%), fatigue (55%), anemia (52%), decreased appetite (52%), constipation (41%), thrombocytopenia (38%), stomatitis (35%), vomiting (35%), alopecia (31%), and diarrhea (31%). The most common AEs in the MTD population (n = 4) included decreased appetite (75%), fatigue (75%), neutropenia (75%), nausea (50%), and thrombocytopenia (50%). The most common AEs in the RP2D population (n = 4) included fatigue (75%), nausea (75%), chromaturia (50%), lymphocyte count decreased (50%), neutropenia (50%), stomatitis (50%), and white blood cell count decreased (50%). The most common grade 3–4 AEs at all grades (n = 29) in the entire population included neutropenia (59%), anemia (24%), thrombocytopenia (24%), febrile neutropenia (21%), and white blood cell count decreased (17%). The most common grade 3–4 AEs in the MTD population (n = 4) included neutropenia (75%). The most common grade 3–4 AEs in the RP2D population (n = 4) included neutropenia (50%). Grade 3 or greater nonhematologic toxicities were rarely observed in each population.

Cardiac toxicity was observed in four subjects [grade 3 tachycardia (n = 1), grade 2 palpitations (n = 1), grade 1 tachycardia (n = 2) and grade 1 ejection fraction (EF) decrease (n = 1)]. One subject at 170 mg/m² discontinued study medication due to decreased ejection fraction (grade 1) before LVEF associated withdrawal criterion changed. One subject had an EF fall less than 50% during the study treatment. That subject did not have a postadministration LVEF assessment at the 215 mg/m² dose level. Among the subjects who received a cumulative dose of 900 mg/m² or more of NC-6300 (n = 7; three subjects at 170 mg/m²; three subjects at 200 mg/m²; one subject at 215 mg/m²), one had grade 1 tachycardia while six experienced no cardiac toxicity.

Twenty-eight subjects were evaluable for radiographic assessment of tumor response. One subject was excluded due to lack of posttreatment tumor assessment. In the overall patient population, the objective response rate (ORR) was 11% (n = 3), the disease control rate (DCR) was 64% (n = 18), stable disease (SD) was 54% (n = 15) and progressive disease (PD) was 36% (n = 10). In soft-tissue sarcoma subset (n = 17), ORR was 18% (n = 3), DCR was 76% (n = 13), SD was 59% (n = 10), and PD was 24% (n = 4). Partial response (PR) was observed in angiosarcoma (two subjects) and endometrial stromal sarcoma (one subject). The waterfall plot in the soft-tissue sarcoma population (n = 17) is shown in Fig. 1. A leiomyosarcoma subject at the 170 mg/m² dose experienced tumor shrinkage over 30%, but also developed a new lesion. A melanoma subject who had received prior therapy with nivolumab alone, pembrolizumab alone, and nivolumab in combination with ipilimumab showed long-term stable disease period (9 months). Median treatment duration was three cycles (range, 1–12) in the overall patient population and four cycles (range, 2–12) in the soft-tissue sarcoma subset. Progression-free survival (PFS) data in individual subjects as well as treatment duration is shown in Supplementary Table S1.

PK parameters of NC-6300 were evaluated for total, released epirubicin and epirubicinol. Figure 2 shows total and released epirubicin concentration–time profile in cycle 1 for the 170 mg/m² cohort. In this cohort, at least three subjects completed eight cycles of treatment and total epirubicin concentrations at the end of infusion in each cycle are plotted in Supplementary Fig. S1. For all cohorts receiving 125 to 215 mg/m², major PK parameters are presented in Table 3, and mean C_max and AUC values at different dose levels are shown in Fig. 3. As a major metabolite, epirubicinol was detected at very low concentrations, typically between 24- and 72-hour postinfusion. C_max and AUC values for epirubicinol are shown in Supplementary Table S2.

Twenty-eight subjects were evaluable for assessment of QOL. Due to lack of posttreatment assessment of QOL, one subject was excluded. Change from baseline data is shown in Supplementary Fig. S2. Among the 28 evaluable subjects, more than half showed numeric improvement or stability in final QOL scores when compared with baseline (Supplementary Fig. S2).

This phase 1b trial evaluated the tolerability/safety, activity, and PK of NC-6300 monotherapy in subjects with advanced, metastatic, or unresectable solid tumors, including soft-tissue sarcomas. Eight DLTs were observed, mostly hematologic, and consistent with what has been reported in studies of conventional epirubicin. MTD and RP2D of NC-6300 were determined to be 185 mg/m² and 150 mg/m², respectively. These doses exceed doses of conventional epirubicin commonly used in the treatment of breast cancer or soft-tissue sarcoma (8–10). Preliminary antitumor activity was observed in heavily pretreated subjects, half of whom had received more than three lines of previous anticancer therapy. PR was observed in three soft-tissue sarcoma subjects (two angiosarcoma; one endometrial stromal sarcoma). Overall, NC-6300 was well tolerated and AEs were manageable. The most common AEs were hematologic or gastrointestinal toxicities, but the severity at MTD and RP2D appeared milder when compared with AEs following treatment with conventional epirubicin (9, 10). Mouridsen and colleagues, and Nielsen and colleagues conducted clinical studies comparing doxorubicin and epirubicin for patients with advanced soft-tissue sarcoma (9, 10). Mouridsen and colleagues administered doxorubicin or epirubicin at the dose of 75 mg/m² every 3 weeks (9). Nielsen and colleagues administered doxorubicin 75 mg/m² once every 3 weeks, epirubicin 160 mg/m² as a single injection in every 3 weeks or epirubicin 60 mg/m² on days 1, 2, and 3 every 3 weeks (13). Although we cannot compare directly, the incidence rate of severe toxicities, particularly leukopenia, thrombocytopenia, or any nonhematologic toxicities, in those studies were higher than NC-6300 at the RP2D (150 mg/m²) or MTD (185 mg/m²), while ORR were similar or slightly lower than the activity of NC-6300 observed in the soft-tissue sarcoma subset of this study. On the basis of this comparison, it is expected that NC-6300 treatment may contribute to maintained efficacy, with improved tolerability.

The risk of cardiac toxicity rises exponentially as cumulative anthracycline dose increases. With a cumulative conventional epirubicin dose of 900 mg/m² (8), the probability of developing congestive heart failure increases steeply. In this study, none of the seven subjects who received more than 900 mg/m² of NC-6300 experienced a clinically significant reduction in LVEF. Although long-term follow-up of adverse cardiac events is required to evaluate possible chronic cardiotoxicity of NC-6300, acute phase assessments indicate that NC-6300 may decrease the risk of cardiotoxicity.

Esposito and colleagues reported that epirubicinol, an active metabolite of epirubicin, may contribute to the increased cardiotoxicity associated with epirubicin (11). Circulating concentrations of epirubicinol in subjects receiving NC-6300 barely exceeded the lower limit of quantification at all dose levels. When delivered through infusion of NC-6300, plasma clearance (CL) of epirubicin (based on total concentrations) was 0.063–0.075 L/hour/m², and steady-state volume of distribution (V_ss) was 2.34-2.76 L/m² (Table 3). This CL is significantly lower than the CL reported with conventional epirubicin (approximately 50 L/hour/m²; ref. 12). The V_ss of epirubicin delivered through NC-6300 is also dramatically smaller than the V_ss of 838 L/m² (approximately equivalent to 20 L/kg) observed with conventional epirubicin (12, 13). For both NC-6300 and conventional epirubicin, these PK parameters are independent of dose levels. The PK difference between NC-6300 and conventional epirubicin is also evident in the concentration–time profiles. Conventional epirubicin rapidly distributes and has a large volume of distribution (12). In contrast, NC-6300 slowly releases epirubicin into the blood stream, thereby allowing prolonged duration of circulation to achieve a more favorable C_max/AUC ratio and potentially mitigate cardiac risk, while delivering doses comparable with those delivered by conventional epirubicin (14).

The elimination half-life (T_1/2) of NC-6300 (23–25 hours; Table 3) is similar to that observed with conventional epirubicin, with little or no drug accumulation during the 21-day treatment regimen (Supplementary Fig. S1). At doses ranging from 125 to 215 mg/m², exposure increased with increasing dose level. Figure 3 shows a linear relationship of C_max and AUC with dose levels. Administration of NC-6300 also delivers a favorable profile for epirubicinol, which may contribute to cardiotoxicity. Formation of epirubicinol from NC-6300 primarily occurs 24 hours postinfusion, in contrast to immediate formation of epirubicinol from conventional epirubicin (12). Furthermore, exposure to epirubicin is significantly reduced following administration of NC-6300. Epirubicinol is only detectable at low ng/mL concentrations sporadically throughout the entire study. As shown in Supplementary Table S2, C_max ranges from 3.7 to 6.9 ng/mL and AUC from 215 to 639 ng*hours/mL, while the AUC of epirubicinol ranges from 1% to 6% of the AUC of released epirubicin (Table 3). In contrast, the metabolite/parent AUC ratio reported for conventional epirubicin could be as high as 44% (11).

This phase 1b study revealed a favorable benefit/risk profile for NC-6300, with potential activity observed, particularly in patients with angiosarcoma. Both angiosarcoma subjects enrolled in this study had received previous antitumor treatment. One subject had three treatment lines including paclitaxel and pazopanib. The other subject had received doxorubicin (80 mg/m² in total) in the neoadjuvant setting. Angiosarcoma is an uncommon subtype of soft-tissue sarcoma, but is a highly aggressive, malignant endothelial-cell tumor developed from vascular or lymphatic tissues (15, 16). Angiosarcoma remains an area of significant unmet medical need, as treatment options remain limited for patients with advanced disease. On the basis of the results of this phase 1b trial utilizing NC-6300, the study protocol was amended to include an expansion cohort of angiosarcoma; thus, enabling us to explore activity of NC-6300 further. Subject enrollment is ongoing.

F. Braiteh reports receiving speakers bureau honoraria from Bristol-Myers Squibb, Pfizer, AstraZeneca, Puma Biotechnology, Amgen, Eli Lilly, Genentech, Celgene, Incyte, Ipsen, Astella, Coherus, Eisai, Exelexis, Regeneron, Taiho, and Merck, and is an unpaid consultant/advisory board member for AstraZeneca, Seattle Genetics, Astella, Lilly, Ipsen, and Dicephera. A. S. Singh is a paid consultant for Blueprint Medicines, Deciphera, Daiichi-Sankyo, Eisai, and Roche; reports receiving speakers bureau honoraria from Blueprint Medicines, Eli Lilly, and Novartis; holds ownership interest (including patents) in Certis Oncology Solutions; and reports receiving other remuneration from OncLive Speaker and Expert Perspectives. A. Osada is an employee of NanoCarrier. R. F. Riedel reports immediate family members with ownership interests (including patents) in Limbguard, LLC., and is an unpaid consultant/advisory board member for NanoCarrier, Bayer, Blueprint, Ignyta, Springworks, Daiichi-Sankyo, and Lilly. No potential conflicts of interest were disclosed by the other authors.

Conception and design: S.P. Chawla, S. Goel, A. Osada

Development of methodology: S.P. Chawla, S. Goel, A. Osada

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): S.P. Chawla, S. Goel, W. Chow, F. Braiteh, A.S. Singh, J.E. Grilley Olson, A. Osada, R.F. Riedel

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): S.P. Chawla, S. Goel, F. Braiteh, A.S. Singh, J.E. Grilley Olson, A. Osada

Writing, review, and/or revision of the manuscript: S.P. Chawla, S. Goel, W. Chow, F. Braiteh, A.S. Singh, J.E. Grilley Olson, A. Osada, I. Bobe, R.F. Riedel

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): F. Braiteh, A. Osada

Study supervision: S.P. Chawla, S. Goel, F. Braiteh, A. Osada, I. Bobe, R.F. Riedel

Other (PI and contributed significant number of patients to the study): S.P. Chawla

The authors acknowledge the subjects and their families for participating in this study.

This study was funded by Nanocarrier.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.