FF-10850, a Novel Liposomal Topotecan Achieves Superior Antitumor Activity via Macrophage- and Ammonia-Mediated Payload Release in the Tumor Microenvironment

Ovarian cancer continues to be a leading cause of death in women with gynecologic cancers, with numbers of new ovarian cancer cases and deaths in the United States estimated at 19,710 and 13,270 in 2023, respectively (1). More than 75% of ovarian cancer patients are diagnosed in the advanced stage due to difficulties in early-stage detection. Although molecular targeted therapies such as bevacizumab and PARP inhibitors have changed the treatment landscape in more recent years, chemotherapies still play a central role (2). Treatment of advanced ovarian cancer remains a challenge due to the high frequency of recurrence: ovarian cancer recurs in approximately 70% of patients within 3 years of first-line platinum-based chemotherapy treatment (3). Patients with platinum-refractory/resistant ovarian cancer is an especially hard to treat population, and non-platinum chemotherapies such as topotecan, liposomal doxorubicin, or paclitaxel show overall response rates of only about 10% in this setting (4). While combination of bevacizumab with these chemotherapies significantly improves progression-free survival (PFS), the median PFS with combination therapy remains at 6.7 months (5, 6). Treatment is further complicated by a general decline in efficacy with recurrent disease (7, 8). Therefore, there remains a significant need for new treatment options for platinum-refractory/resistant ovarian cancer.

Topotecan is a topoisomerase I inhibitor approved for ovarian cancer, cervical cancer, and small cell lung cancer treatment. A factor that may limit its clinical efficacy is its rapid clearance from blood. Although its tumor growth inhibitory activity is dependent on exposure time, topotecan distribution (T_1/2 α) and elimination (T_1/2 β) half-lives in human plasma are 8 minutes and 132 minutes, respectively (9, 10). Topotecan also has a challenging safety profile. Grade 4 neutropenia, leukopenia, and thrombocytopenia were observed in 79.3%, 33.6%, and 25.2% of patients, respectively, in the phase III clinical trial for recurrent ovarian cancer (11). These safety issues limit the clinical use of topotecan in combination with other anticancer agents such as platinum and PARP inhibitors, despite studies indicating enhanced antitumor activity in combination settings (12–16).

Liposomal formulations represent a promising approach to improve the pharmacokinetics, efficacy, and safety of chemotherapies, and a number of liposomal chemotherapies have been approved for various cancer types (17). The leaky vasculature and impaired lymphatic drainage in the tumor microenvironment (TME) is thought to allow for increased extravasation and retention of macromolecules like liposomes, a mechanism known as the enhanced permeability and retention (EPR) effect (18). Preferential accumulation of liposomal drugs in tumor tissue could therefore result in improved efficacy and decreased toxicity in normal tissues compared with their non-liposomal counterparts. Payload release in the TME is another critical factor impacting liposomal drug efficacy, and modified liposomal formulations designed to trigger payload release in the TME show superior antitumor activity compared with non-modified liposomal formulations (19).

To maximize the clinical benefit of topotecan, we designed a novel liposomal formulation, FF-10850, with an optimized compositions of dihydrosphingomyelin (DHSM), cholesterol, and polyethylene glycol (20). Here, we report on the favorable pharmacokinetic, antitumor activity, and safety profiles of FF-10850 in a variety of murine models. Data demonstrated the mechanisms of topotecan released from FF-10850 that improve on its antitumor activity both alone and in combination with platinum, PARP inhibitor, and an anti–PD-1 antibody therapies.

Study protocols were approved by the Institutional Animal Care and Use Committee of FUJIFILM Corporation. Animal experiments were conducted in compliance with the Act on Welfare and Management of Animals and Code of Welfare of Laboratory Animals of FUJIFILM Corporation. For the DF181 patient-derived xenograft (PDX) model experiment, the protocol was approved by the Institutional Animal Care and Use Committee of the Dana-Farber Cancer Institute following the guidance of the Association for Assessment and Accreditation of Laboratory Animal Care.

FF-10850 (ref. 20; Supplementary materials and methods) or topotecan (Biocompounds Pharmaceutical, Shanghai, China) was intravenously administered at the specified doses to BALB/c nude mice (CLEA Japan, Tokyo, Japan) bearing ES-2 ovarian cancer (ATCC, Manassas, VA) xenografts. All FF-10850 doses are provided in terms of mg/kg topotecan. Plasma samples were prepared by centrifugation at 800 × g for 10 minutes at 4°C. For isolation of unencapsulated topotecan, supernatants were harvested after ultracentrifugation of plasma at 200,000 × g for 2 hours at 4°C. For bone marrow cell preparation, femurs were placed in microtubes after cutting their ends and centrifuged at 1,200 × g for 1 minute at 4°C. For tumor tissue samples, frozen tumors were pulverized using a Multi-beads Shocker (YASUI KIKAI, Osaka, Japan).

To extract topotecan for quantification, plasma, bone marrow, and tumor samples prepared above were resuspended in methanol containing 0.1% formic acid, centrifuged at 1,820 × g for 10 minutes at room temperature, and supernatants were mixed with an equal volume of 0.1% formic acid. Topotecan was quantified by LC/MS-MS [ultra-performance liquid chromatography (ACQUITY UPLC, Waters, Milford, MA) and tandem mass spectroscopy (Triple Quad 5500, AB Sciex, Framingham, MA)]. Pharmacokinetic parameters were calculated using Phoenix WinNonlin (Certara, Princeton, NJ).

For the subcutaneous ES-2 xenograft model, BALB/c nude mice were subcutaneously inoculated with 3 × 10⁶ ES-2 ovarian cancer cells. Tumor volume was calculated as (width)² × (length) × 0.5, and mice were randomized into groups (N = 8/group) based on tumor volumes 7 days after tumor inoculation using the randomized block method by StatLight (Yukms, Kanagawa, Japan). Statistical analysis was performed by Steel-Dwass test using JMP (SAS Institute, Cary, NC).

For the intraperitoneally disseminated ES-2 xenograft model, BALB/c nude mice were intraperitoneally inoculated with 2 × 10⁶ ES-2 cells and randomized into groups (N = 10/group) 7 days after tumor inoculation. Mice were euthanized under anesthesia when they presented severe abdominal distention, emaciation or decrease of locomotor activities. Statistical analysis was performed by log-rank test with Holm method adjustment for multiple comparisons using JMP.

An intraperitoneally disseminated DF181 PDX model was established as described previously by Liu and colleagues (21). NSG mice (Jackson Laboratory, Bar Harbor, ME) were intraperitoneally inoculated with 5 × 10⁶ DF181 cells. Luciferin was injected at 150 mg/kg into anesthetized mice to monitor tumor growth by bioluminescence imaging (BLI) using an IVIS Spectrum In Vivo Imaging System (PerkinElmer, Waltham, MA). Mice were randomized into groups (N = 5/group) based on BLI signals at day 17 after tumor inoculation. Mice were euthanized under anesthesia when they presented severe abdominal distention, emaciation or decrease of locomotor activities. Dunnett test comparing day 15 tumor volume in treated versus vehicle groups was performed using JMP. Tukey test comparing tumor volume data among treated groups on day 22 was performed using JMP.

FF-10850 was intravenously administered at the indicated doses once weekly for 2 weeks. Topotecan was intravenously administered at 2 mg/kg for 5 consecutive days. Liposomal doxorubicin 16.7 mg/kg (Janssen Pharmaceuticals, Beerse, Belgium) was intravenously administered once weekly for 2 weeks.

BALB/c mice (Japan SLC, Inc., Shizuoka, Japan) were randomized into groups (N = 5/group). FF-10850 was intravenously administered once at a dose of 0.5 or 2 mg/kg. Topotecan 2 mg/kg was intravenously administered for 5 consecutive days. Blood was collected from the post-caval vein under anesthesia at 1, 3, 5, 7, or 10 days after initial dosing and analyzed with an XT-2000i hematology analyzer (Sysmex, Hyogo, Japan).

DiI-fluorescent labeled FF-10850 (Supplementary materials and methods) was intravenously administered to ES-2 subcutaneous tumor-bearing mice (N = 3/group) 7 days after tumor inoculation. Tumors were collected 24 hours after administration and dissociated into single-cell suspensions using a gentleMACS dissociator (Miltenyi Biotec, North Rhine-Westphalia, Germany). Cells were stained with the LIVE/DEAD Fixable Yellow Dead Cell Stain Kit (Thermo Fisher Scientific, Waltham, MA). After blocking with anti-CD16/CD32 (Thermo Fisher Scientific), cells were stained with PerCP-Cy5.5-conjugated anti-CD11b (BD Biosciences, Franklin Lakes, NJ) and APC-conjugated anti-F4/80 (Miltenyi Biotec). Data were acquired using an Attune NxT Flow Cytometer (Thermo Fisher Scientific) and analyzed by FlowJo (BD Biosciences).

Macrophage colony-stimulating factor (M-CSF)-induced murine bone marrow–derived macrophages were incubated with 10 μg/mL FF-10850 for 2 hours. Cells were washed 6 times and incubated in fresh culture medium at 37°C. Unencapsulated topotecan in culture medium was isolated by ultracentrifugation at 200,000 × g for 90 minutes at 4°C. Intracellular topotecan was quantified in macrophage lysates prepared by addition of methanol/0.1% formic acid using LC/MS-MS. The percentage of topotecan released was calculated as (unencapsulated topotecan in culture medium)/[(unencapsulated topotecan in culture medium) + (intracellular topotecan)] × 100.

Four μg/mL of FF-10850 or liposomal doxorubicin was incubated in murine plasma in the presence or absence of 10 mmol/L NH₄Cl based on the ammonia concentration in tumor interstitial fluid (TIF). Unencapsulated topotecan or doxorubicin was isolated from plasma samples by ultracentrifugation at 200,000 × g for 90 minutes at 4°C and quantified via LC/MS-MS. The percentage of payload release was calculated as (unencapsulated payload) / [(unencapsulated payload) + (encapsulated payload)] × 100.

BALB/c nude mice were subcutaneously inoculated with 5 × 10⁶ A2780 platinum-sensitive ovarian cancer cells (ECACC, Salisbury, UK). Tumor-bearing mice were randomized into groups (N = 8/group) based on tumor volume 7 days after tumor inoculation. FF-10850 was intravenously administered at 0.5 mg/kg once weekly for 2 weeks. Carboplatin (TCI, Tokyo, Japan) was intraperitoneally administered at 40 mg/kg once weekly for 2 weeks. Mice were euthanized under anesthesia when tumor volume exceeded 2,000 mm³. Statistical analysis for survival was performed with log-rank test with Holm method adjustment for multiple comparisons using JMP.

BALB/c nude mice were subcutaneously inoculated with 1 × 10⁷ BRCA2-deficient Capan-1 pancreatic cancer cells (ATCC). Mice were randomized into groups (N = 8/group) based on tumor volume 11 days after tumor inoculation. FF-10850 was intravenously administered at 0.5 mg/kg once weekly for 4 weeks. Olaparib (MedChemExpress, Monmouth Junction, NJ) was intraperitoneally administered at 50 mg/kg for 6 consecutive days for 4 weeks. Tumor volumes were statistically analyzed using Steel-Dwass tests with JMP.

BALB/c mice were subcutaneously inoculated with 1 × 10⁶ CT26 murine colon cancer cells (ATCC). Mice were randomized into groups (N = 8/group) based on tumor volume 14 days after tumor inoculation. FF-10850 was intravenously administered at 4 mg/kg once weekly for 3 weeks. Anti–PD-1 antibody clone RMP1–14 (Bio X Cell, Lebanon, NH) was intraperitoneally administered at 10 mg/kg twice weekly for 3 weeks. Mice were euthanized under anesthesia when tumor volume exceeded 2,000 mm³. Statistical analysis of survival was performed with log-rank test with Holm method adjustment for multiple comparisons using JMP. Mice that achieved complete tumor regression with combination treatment of FF-10850 and anti–PD-1 antibody were reinoculated with 1 × 10⁶ CT26 cells on day 66 after initial treatment, then subsequently inoculated with 5 × 10⁴ 4T1 cells (ATCC) 122 days after initial treatment (N = 5). Naïve mice were inoculated with CT26 or 4T1 cells as described above as a control group (N = 8/group).

The data generated in this study are available upon request from the corresponding author.

Pharmacokinetic analysis was performed in mice bearing the ES-2 subcutaneous ovarian cancer xenografts treated with FF-10850 or topotecan (Fig. 1A). FF-10850 showed >1,700-fold higher plasma exposure (AUC_0-inf) and >16-fold longer half-life than topotecan (Supplementary Table S1), which was rapidly cleared from plasma as previously reported (9). Exposure of FF-10850 was linear and dose proportional. Analysis of unencapsulated topotecan in plasma of FF-10850–treated mice revealed that topotecan was stably encapsulated throughout the tested time points (Fig. 1B).

To examine whether improved pharmacokinetics of FF-10850 could result in enhanced antitumor activity, tumor growth inhibition was evaluated in the ES-2 subcutaneous ovarian cancer xenograft model. FF-10850 was administered at the indicated doses once weekly for 2 weeks and results compared with those of topotecan administered at 2 mg/kg for 5 consecutive days. The schedule selected for topotecan administration represents the approved schedule for use in human, and was also shown to have a favorable therapeutic index in our studies (Supplementary Fig. S1). Liposomal doxorubicin, an approved liposomal drug for recurrent ovarian cancer treatment, was also tested at a dose of 16.7 mg/kg once weekly for 2 weeks. Topotecan and liposomal doxorubicin showed statistically significant tumor growth inhibition compared with the vehicle control and induced a 10% to 20% loss of body weight (Fig. 2A and B). FF-10850 showed comparable tumor growth inhibition to topotecan and liposomal doxorubicin at 0.5 mg/kg and achieved almost complete tumor regression at 1.3 mg/kg without significant body weight loss. Mice treated with 4 mg/kg FF-10850 exhibited a ∼10% decrease in body weight that was reversible; this dose was considered to be the maximum tolerated dose for FF-10850. We then evaluated survival in an ES-2 intraperitoneal disseminated xenograft model that resembles metastasis of advanced ovarian cancer. FF-10850 induced survival prolongation in a dose-dependent manner. Furthermore, FF-10850 showed statistically significant survival prolongation at 2 mg/kg compared with topotecan, and at 4 mg/kg compared with both topotecan and liposomal doxorubicin (Fig. 2C; Supplementary Table S2).

Antitumor activity of FF-10850 was further evaluated in a luciferized DF181 PDX model of platinum-refractory ovarian cancer that allows for quantitative monitoring of intraperitoneal tumor growth via bioluminescence. The DF181 PDX model is a well-characterized model that retains patient-relevant properties in terms of tumor histopathology, molecular profiles, and platinum resistance (21). FF-10850 showed statistically significant tumor growth inhibition at 2 mg/kg compared with topotecan without significant body weight loss, and achieved almost complete growth inhibition at 4 mg/kg (Fig. 2D and E). Collectively, FF-10850 consistently demonstrated superior antitumor activity in conventional cell line-based xenograft models as well as the clinically relevant PDX model of ovarian cancer.

Hematologic toxicity is the most frequent and severe toxicity associated with topotecan treatment (11, 22). Thus, in addition to body weight change, we further evaluated the safety profile of FF-10850 by examining its hematologic toxicity in mice. Neutrophils, lymphocytes, platelets, and reticulocytes were decreased following topotecan administered at 2 mg/kg for 5 consecutive days. In contrast, 0.5 mg/kg FF-10850, a dose that demonstrated comparable antitumor activity to topotecan (Fig. 2A), showed marginal effects on these cells. At 2 mg/kg, a dose resulting in significantly improved antitumor activity compared with topotecan (Fig. 2), FF-10850 induced transient decreases in neutrophils and reticulocytes that were milder than those elicited by topotecan treatment and that reverted to near-baseline levels by day 7, when the second dose of FF-10850 would be administered in xenograft models (Fig. 3A–D).

We hypothesized that preferential tumor accumulation of FF-10850 via the EPR effect may contribute to the observed improvement in therapeutic index compared with topotecan. Topotecan concentrations were quantified in bone marrow and tumor tissue after administration of 2 mg/kg FF-10850 or 2 mg/kg topotecan. Consistent with that observed in plasma, higher exposures and a more prolonged half-life were achieved with FF-10850 compared with topotecan administration in both tumor tissue and bone marrow (Fig. 3E; Supplementary Table S3). The tumor to bone marrow ratio for AUC_0-inf was however substantially greater for FF-10850 versus topotecan. These results support that preferential tumor accumulation of FF-10850 could in part contribute to the improvement of its therapeutic index.

We hypothesized that the observed effects may also be driven by tumor-preferential payload release because this phenomenon is critical to the antitumor activity of liposomal drugs (19). Prior studies have shown that tumor-associated macrophages (TAM) can internalize nanoparticles and release payloads in the TME (23, 24). To examine if FF-10850 can be internalized by TAMs, DiI-fluorescent labeled FF-10850 was administered to ES-2 subcutaneous tumor-bearing mice. Tumors were harvested 24 hours after administration and analyzed by flow cytometry. DiI-fluorescence of FF-10850 was predominantly observed in CD11b⁺F4/80⁺ TAMs (Fig. 4A,–C). Payload release was analyzed by quantification of unencapsulated topotecan in culture medium of bone marrow–derived macrophages following internalization of FF-10850. Payload was completely released from macrophages within 24 hours (Fig. 4D). These results support the notion that FF-10850 is internalized by TAMs, followed by complete release of payload to the surrounding tissue.

High concentration of ammonia can trigger payload release from liposomal doxorubicin (25, 26). Ammonia is produced during glutaminolysis, a process that can be activated by tumor metabolic reprogramming (27, 28). TIF of ES-2 subcutaneous xenografts was collected by a centrifugation method and assayed for ammonia content (refs. 29, 30; Supplementary Materials and Methods). Ammonia concentration in TIF was ∼ 10 mmol/L compared with 0.5 mmol/L in plasma (Fig. 5A). We examined whether high concentrations of ammonia could induce payload release from FF-10850. FF-10850 was incubated in murine plasma in the presence or absence of 10 mmol/L NH₄Cl. Unexpectedly, FF-10850 showed much more rapid and greater payload release than liposomal doxorubicin, which was used as a positive control, in the presence of 10 mmol/L NH₄Cl. Approximately 80% of payload was released from FF-10850 within 6 hours, compared with the <20% release from liposomal doxorubicin whereas little payload release was observed in either formulation without NH₄Cl (Fig. 5B). These results support the notion that FF-10850 releases payload efficiently in the ammonia-rich TME where glutaminolysis is activated by metabolic reprogramming.

In addition to monotherapy, the preferential accumulation of FF-10850 in tumor tissue could enhance the antitumor activity of combination therapies while minimizing off-target toxicities. Carboplatin-based combination therapies are standard first-line treatments for advanced ovarian cancer, as well as second-line treatments in platinum-sensitive patients (2). We explored the use of FF-10850 in combination with carboplatin in the A2780 subcutaneous platinum-sensitive ovarian cancer xenograft model. The combination therapy significantly prolonged survival compared with each monotherapy without detrimental effects on body weight loss (Fig. 6A and B; Supplementary Table S4).

PARP inhibitors have been approved for treatment of newly diagnosed advanced or recurrent ovarian cancer (2). Topoisomerase I inhibitors in combination with a PARP inhibitor have shown superior synergistic effects compared with other chemotherapies, especially in BRCA-deficient tumor cells (14, 31, 32). We examined the antitumor activity of FF-10850 in combination with olaparib in the BRCA2-deficient Capan-1 subcutaneous pancreatic cancer xenograft model. The combination therapy significantly enhanced tumor growth inhibition compared with each monotherapy without impacting body weight (Fig. 6C and D). Maintenance treatment with olaparib after completion of FF-10850 monotherapy also attenuated tumor regrowth (Supplementary Fig. S2).

The number of ongoing clinical trials for immune checkpoint inhibitors in combination with chemotherapies continues to increase (33). We evaluated the antitumor activity of FF-10850 in combination with anti–PD-1 antibody in the CT26 subcutaneous syngeneic tumor model. Combination therapy significantly prolonged survival compared with each monotherapy without exacerbation of body weight loss (Fig. 6E and F; Supplementary Table S5). Notably, 6 of 8 mice achieved complete regression as a result of combination therapy that persisted until the end of observation on day 66. CT26 cells were reinoculated into mice that achieved durable complete regression by combination treatment to examine whether CT26-specific immunity was established. All tested mice rejected reinoculated CT26 cells, while rejection was not observed upon subsequent inoculation of 4T1 cells (Fig. 6G and H). These results indicate that the combination of FF-10850 and anti–PD-1 antibody induces tumor-specific immunity that contributes to long-term tumor suppression.

FF-10850, a DHSM-based liposomal formulation of topotecan, achieves efficient topotecan loading, stable encapsulation, and a prolonged plasma half-life (20). Here, we have shown that FF-10850 demonstrates superior antitumor activity in several ovarian cancer models compared with topotecan, with reduced hematologic toxicity. Our findings also indicate that the improvement in activity was achieved by preferential tumor accumulation and complete payload release mediated by TAMs and ammonia concentration in the TME. Furthermore, FF-10850 demonstrated enhanced antitumor activity in combination with carboplatin, olaparib, or anti–PD-1 antibody, without exacerbation of body weight loss.

Because collectively the EPR effect, TAMs, and ammonia in the TME are important contributing factors to tumor distribution and payload release from FF-10850, evaluation in models that retain TME characteristics similar to that in humans is essential for prediction of clinical activity. The DF181 PDX model is particularly useful in this context, as it retains histopathologic, molecular, and platinum-resistance characteristics observed in the human tumor (21). TAMs and activation of glutaminolysis, which mediate FF-10850 payload release, are known to be associated with tumor progression and poor prognosis (28, 34–37). The internalization of nanoparticles by TAMs and the activation of glutaminolysis have been demonstrated in both preclinical models and in patient tumor biopsies (23, 24, 27, 28, 38). The percentage of TAMs as measured by flow cytometry in the ES-2 model of human ovarian cancer (8.16 ± 4.73%) has also been shown to be similar to that measured by IHC analysis in ovarian cancer patient biopsies (5%), although differences in macrophage markers and analysis methods should be noted (34). While payload release by TAMs and the effect of ammonia concentration on release have been evaluated in preclinical models, further investigation of these payload release mechanisms is needed in clinical samples to fully translate their impact on clinical efficacy. Evaluation of TAM- and glutaminolysis-related biomarkers in parallel with antitumor activity in the ongoing clinical trials of FF-10850 would also be of great interest.

Achieving complete payload release through TAM uptake and an increased ammonia concentration in the TME may contribute to the superior antitumor activity observed with FF-10850 compared with liposomal doxorubicin. Laginha and colleagues indicate that less than 50% of payload is released from liposomal doxorubicin in the TME (39), and maximizing payload release through modification of its formulation has been shown to enhance its antitumor activity (19). Both FF-10850 and liposomal doxorubicin formulations consist of a positively charged payload that is stably entrapped, forming a precipitated salt with sulfate inside the liposomes. As it is highly diffusible and membrane permeable (25), ammonia can distribute relatively freely into liposomes where it deprotonates positively charged payload, generating uncharged payload susceptible to leakage. Similar to previous observations with liposomal doxorubicin (25, 26), the ammonia-mediated payload release from FF-10850 has been shown to be concentration and pH-dependent (Supplementary Fig. S3), as increasing pH facilitates conversion of ammonium to ammonia. Although the mechanism of ammonia-mediated payload release is thought to be the same for liposomal doxorubicin and FF-10850, a comparatively greater payload release from FF-10850 was observed. FF-10850, a DHSM-based liposomal topotecan, does exhibit a more prolonged plasma half-life compared with a liposomal topotecan with hydrogenated soy phosphatidylcholine (HSPC)-based formulation like liposomal doxorubicin (Supplementary Fig. S4), presumably due to the more stable retention and lower permeability of DHSM-based liposomal membranes. Permeability of the liposomal membrane, however, is unlikely to be changed by ammonia. It is more likely the physicochemical properties of particular payloads themselves that have the greatest impact on ammonia-mediated payload release. Topotecan has 175-fold solubility (35 mmol/L) of doxorubicin (0.2 mmol/L) at the intraliposomal pH condition, pH 5–6, which should facilitate ammonia-mediated payload release, along with a polar surface area for topotecan (103 Ų) that confers greater membrane permeability compared with that for doxorubicin (206 Ų; ref. 40).

Encapsulating toxic drugs in liposomes is expected to mitigate safety concerns (17). FF-10850 clearly reduced the hematologic toxicity and body weight loss elicited by topotecan. Similar to other liposomal drugs, the surface of FF-10850 is modified with polyethylene glycol to prolong its plasma half-life and reduce uptake by macrophages in the liver and spleen (17, 18, 20), however, substantial distribution is still observed in these tissues (Supplementary Fig. S5). The lack of severe toxicities observed in the liver and spleen, however, are thought related to rapid diffusion of topotecan into the systemic circulation once released from liver and splenic macrophages. Liposomal formulations can sometimes contribute to unexpected toxicities. Palmar-plantar erythrodysesthesia (PPE), which is the most common adverse event induced by liposomal doxorubicin, is attributed to its liposomal formulation. Approximately 50% of patients treated with liposomal doxorubicin develop PPE compared with only 2% of patients treated with doxorubicin (41). In our ES-2 subcutaneous xenograft model, all mice treated with liposomal doxorubicin developed skin abnormalities, whereas such findings were not observed in mice treated with FF-10850 (Supplementary Fig. S6). There have been no additional or unexpected toxicities observed preclinically that were specifically attributed to the liposomal formulation of FF-10850.

The favorable toxicity profile of FF-10850 may allow for improving the antitumor activity of combination therapies. Platinum-based combination chemotherapies are the standard for advanced ovarian cancer treatment. While the combination of topotecan and cisplatin is approved for cervical cancer, reductions in both dose and administration frequency are often required to achieve tolerability (12, 13). Similar observations were reported in clinical trials of topotecan and PARP inhibitor combination therapies (15, 16, 42). Our preclinical studies have shown that FF-10850 significantly enhanced antitumor activity without impacting body weight loss in combination with carboplatin or olaparib, thought related to preferential accumulation of FF-10850 in tumor tissue, thereby minimizing normal tissue exposure. Further optimization of the dosing schedule may further improve the therapeutic index (43). Our results also support the use of FF-10850 in combination with immune checkpoint inhibitors. We have demonstrated that the combination of FF-10850 with an anti–PD-1 antibody established tumor antigen-specific immunity and resulted in durable complete tumor regression. FF-10850 may induce tumor antigen release and the production of fragmented DNA, thereby activating immune cells through the cGAS-STING pathway (44, 45). In addition, the lymphocyte-sparing profile of FF-10850 may prove beneficial for cytotoxic T-cell proliferation.

Data from these studies contribute to our understanding of how liposomal delivery can improve drug efficacy. These preclinical results will assist interpretation of data from the ongoing phase I clinical trial of FF-10850 in advanced solid tumors, including ovarian cancer (NCT04047251), and provide further support for analysis of FF-10850 in combination with other anticancer agents.

S. Shimoyama reports a patent for WO2019244979 issued, a patent for WO2020071349 issued, a patent for WO2022250013 pending, a patent for WO2022250015 pending, and a patent for Provisional pending; and the author is an employee of FUJIFILM Pharmaceuticals U.S.A., Inc. K. Okada reports a patent for WO2020071349 issued and a patent for Provisional pending; and the author is an employee of FUJIFILM Corporation. T. Kimura reports a patent for WO2022250013 pending; and T. Kimura is an employee of FUJIFILM Corporation. Y. Morohashi reports a patent for WO2022250013 pending; and the author is an employee of FUJIFILM Corporation. S. Nakayama reports a patent for WO2022250013 pending; and the author was an employee of FUJIFILM Corporation at the time the work was performed. S. Kemmochi reports I was an employee at Fujifilm Corporation at the time the work was performed. K. Makita-Suzuki reports a patent for WO2022250013 pending; and the author was an employee of FUJIFILM Corporation at the time the work was performed. U.A. Matulonis reports other support from FujiFilm during the conduct of the study; personal fees from Allarity, NextCure, Profound Bio, Merck, Trillium, Immunogen, CureLab, Eisai, Boeringer Ingelheim, Symphogen, Alkermes, Pfizer, Med Learning Group, Ovarian Cancer Research Alliance; and personal fees from Novartis outside the submitted work. M. Mori reports a patent for WO2018181963 issued, a patent for WO2019244979 issued, a patent for WO2020071349 issued, a patent for WO2022250013 pending, a patent for WO2022250015 pending, and a patent for Provisional pending; and the author is an employee of FUJIFILM Corporation. No other disclosures were reported.

S. Shimoyama: Conceptualization, resources, data curation, formal analysis, supervision, investigation, visualization, methodology, writing–original draft, project administration, writing–review and editing. K. Okada: Conceptualization, data curation, formal analysis, validation, investigation, visualization, methodology, writing–review and editing. T. Kimura: Conceptualization, data curation, formal analysis, validation, investigation, visualization, methodology, writing–review and editing. Y. Morohashi: Data curation, formal analysis, validation, investigation, visualization, methodology, writing–review and editing. S. Nakayama: Data curation, formal analysis, validation, investigation, visualization, methodology, writing–review and editing. S. Kemmochi: Data curation, formal analysis, validation, investigation, visualization, methodology, writing–review and editing. K. Makita-Suzuki: Data curation, formal analysis, validation, investigation, visualization, methodology, writing–review and editing. U.A. Matulonis: Conceptualization, supervision, investigation, writing–review and editing. M. Mori: Conceptualization, resources, data curation, formal analysis, supervision, project administration, writing–review and editing.

We thank members of the Belfer Center for Applied Cancer Science in Dana-Farber Cancer Institute, FUJIFILM Business Expert Corporation, and KAC Co., Ltd., for technical support and animal care. We thank Timothy Madden, Mary Johansen, Ruth Ann Subach, Naoki Yamada, David Wages, and Catherine Wheeler for providing scientific consultation, and Florencia Pascual for scientific writing support.

The publication costs of this article were defrayed in part by the payment of publication fees. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.