Integrating antibody drug conjugates in the management of gynecologic cancers

Introduction

The American Cancer Society estimates that 115 130 new cases of gynecological cancer arose in 2022, causing 32 830 deaths in the United States and 26 500 deaths in Europe.1 Ovarian and cervical carcinomas are the most frequent malignancies and are responsible for the highest number of cancer-related deaths among gynecological cancer.1 A platinum-based doublet chemotherapy remains the gold standard for advanced, metastatic or recurrent gynecological malignancies, and it is usually indicated in combination with targeted agents or immunotherapy.2–8 However, many patients who initially respond to platinum and targeted agents develop resistance, and an increasing number of prior treatments lead to decreased objective response rates for subsequent therapy.9 10 Therefore, developing a new class of drugs to treat these malignancies is an urgent unmet need.

Antibody drug conjugates (ADCs) have been used in clinical practice for the treatment of hematological malignancies for more than 20 years.11 12 In particular, the unique structure of this class of drugs, which improves chemotherapy efficacy while limiting the systemic toxicity, made ADCs attractive therapeutic candidates that rapidly gained traction in clinical trials.13 However, the era of ADCs in solid tumors evolved once the Food and Administration (FDA) approved trastuzumab emtansine (T-DM1) for the treatment of metastatic and early-stage breast cancer.14 15

This article briefly details the ADC mechanism of action and highlights potential targets in gynecological malignancies. We then summarize findings from recent clinical trials, discuss the first FDA regulatory approvals in cervical cancer and ovarian cancer. Finally, we present novel ADCs for gynecological malignancies at different stages of research and development. Further, we outline potential future directions of clinical trials by combining ADCs with chemotherapy and/or other targeted therapies approved in gynecological cancers, such as bevacizumab and immune checkpoint inhibitors.

ADCs Design and Mechanism of Action

The concept of ADCs as a “magic bullet” was introduced by Paul Ehrlich, describing the selective delivery of a cytotoxic drug to the cancer cell.16 This selective delivery is mediated by humanized monoclonal antibody (mAb) binding to a target, either solely expressed or overexpressed on the surface of cancer cells and with a low expression in healthy tissues.13 Following target binding, the ADC molecule should be subsequently internalized, ideally through a cellular receptor, facilitating the direct delivery of the cytotoxic agent into the tumor cells.17

Because the IgG1 subtype is the most common immunoglobulin found in circulation and elicits minimal immunogenicity, it is usually the IgG of choice for ADCs.13 Interestingly, some antibodies elicit an antitumor effect alone which has been shown to improve the overall antitumor efficacy, especially in breast cancer. The humanized mAb trastuzumab, which exerts its anticancer effect by inhibiting the proliferation induced by human epidermal growth factor receptor-2 (HER-2), was the first in its class to be widely adopted in the clinical practice.18 A second humanized mAb with the same target, pertuzumab, further improved overall survival in patients with metastatic disease, lessening the risk of recurrence and improving pathologic complete response rate when used as neoadjuvant therapy.19 20

Despite these impressive responses of humanized mAb in breast cancer, early efforts to use mAbs (“naked antibodies”) to therapeutically target other receptors such as folate receptor-α (FRα) in ovarian cancer failed to achieve clinical efficacy. This is the case for the humanized anti-FRα mAb farletuzumab as well as vintafolide, a small molecule antigen drug conjugate, both targeting FRα. In the case of farletuzumab, phase II trials were promising in terms of both efficacy and safety.21 Unfortunately, the subsequent phase three trials in both platinum-sensitive (NCT00849667) and platinum-resistant (NCT00738699) disease22 both failed to meet primary outcome endpoints. Similarly, the phase III clinical trial PROCEED (NCT01170650) was discontinued due to the failure of vintafolide to demonstrate an increase in progression free survival (PFS).23 Nevertheless, FRα has emerged as one of the most attractive candidates for molecularly targeted approaches in ADC development for ovarian cancer due to its almost ubiquitous expression on the surface of high grade serous ovarian cancer and its ability to internalize large molecules containing a cytotoxic payload. Thus far, and it is the best characterized target in clinical trials for ovarian cancer (Figure 1).24–26 Aberrant FRα overexpression is found almost exclusively in epithelial tumors, and previous studies have reported that 80–90% of high grade serous ovarian cancers constitutively express FRα.27

Figure 1Figure 1Figure 1

ADC timeline in ovarian cancer.

Several other antigens have been exploited for the use of ADCs in gynecological tumors. These include antigens such as MUC-16 (CA125), mesothelin (MSLN), tumor-associated calcium signal transducer 2 (TROP-2), sodium-dependent phosphate transport protein 2B (NaPi2b), tissue factor (TF), protein tyrosine kinase 7 (PTK7), T cell immunoglobulin and mucin domain (TIM), neurogenic locus notch homolog protein 3 (NOTCH3), activated leukocyte cell adhesion molecule (CD166) and cadherin 6 (CDH6) have been reported as ADC targets.22 Table 1 presents an overview of these antigens and their expression across gynecological malignancies.

Table 1

Selected tumor types with reported target overexpression

Microtubules, inhibition of which induces cell cycle arrest in the G2/M phase, have been an attractive drug target since 1970 as clinicians struggled to treat patients who either did not respond or developed resistance to available treatment options.28 Tubulin inhibitors can be found in nature, and their synthetic derivatives such as taxanes (eg, paclitaxel) and vinca alkaloids (eg, vinblastine) are currently among the most successful cytotoxic drugs used for standard therapy in both hematological and solid cancers.29 However, the limitations of these earlier tubulin inhibitors led to the development of a new generation of tubulin inhibitors with greater potency.30 Although these highly potent compounds were too toxic to be clinically useful on their own, they stimulated considerable research effort into developing suitable payloads for ADCs.

One of the first compounds to be successful as the cytotoxin conjugate for ADCs was the tubulin inhibitor maytansine, with picomolar IC50 values and higher potency (100- to 1000-fold) compared with doxorubicin or paclitaxel.31 The two existing maytansine conjugates, DM1 and DM4, are currently being investigated in different types of cancers.17 Other tubulin inhibitors being used as cytotoxins for ADCs are auristatins derivates, including monomethyl auristatin-E (MMAE) and monomethyl auristatin-F (MMAF).31 Finally, DNA damaging agents, such as derivates of calicheamicin, duocarmycin, pyrrolobenzodiazepine and topoisomerase I inhibitors also have promising antitumor activity.17

The third essential structural component of an ADC is the linker, which functions to connect the antibody to the cytotoxic payload.32 Several important characteristics should be considered when choosing linkers for ADCs. The linker should be stable enough in the bloodstream to transport the cytotoxic molecules to the target antigen without premature cleavage.13 A second characteristic that needs to be considered is the ability of the linker to be cleaved on internalization in the tumor cell, followed by the efficient release of the cytotoxic molecules.13 Although the intracellular cleavage is responsible for the delivery and antitumor effect of an ADC, the particular chemistry of the linker can also dictate whether a portion of the drug can be cleaved extracellularly and diffuse into surrounding cells.31 This results in a ‘bystander killing’ effect, or the death of the surrounding tumor cells that may not express the ADC target antigen, resulting in an antigen-independent cytotoxic effect.32 Taken together, the conjugation chemistry of the linker has a crucial role in the therapeutic window of an ADC. The field of linkers has been challenging over the years and represents an important research area on its own.17

Although linker characteristics are crucial for ADC design, the drug-to-antibody-ratio (DAR) has also been reported to influence the therapeutic effect of ADCs.33 A higher number of cytotoxic molecules on an antibody confers greater cytotoxicity, as opposed to an antibody that carries fewer payloads.34 However, this may alter the structure of the ADC and compromise its stability, reduce the antigen affinity, or affect its clearance.35

ADCs Landscape in Late-phase Clinical Trial DevelopmentCervical

In September 2021, the ADC tisotumab vedotin (TV) was the first FDA-approved ADC for the treatment of patients with recurrent or metastatic cervical cancer with disease progression on or after chemotherapy.36 TV is an ADC targeting tissue factor (TF) and is conjugated to MMAE, a microtubule disruptor (Table 2). The FDA accelerated approval was supported by findings from innovaTV 204, a multicenter, open-label, single-arm, phase II study which enrolled 102 patients across 35 academic centers, hospitals, and community practices in Europe and the USA. The trial achieved an objective response rate (ORR) of 24% (95% confidence interval (CI), 15.9 to 33.3) with complete responses (CR) in 7%, partial response (PR) in 17%, stable disease in 49%, and progressive disease (PD) in 24%.37 Among patients with response to TV, the response occurred approximately 6 weeks after starting the treatment, and the median duration of response (DOR) was a median of 8.3 months.37 Taken together, TV is an active agent for the second-line therapy of metastatic cervical cancer. The TV approval is an accelerated approval based on ORR and DOR that is superior to what one would expect with available medicines. The confirmed approval is pending the results of the ongoing GOG 3057/ENGOT cx12/innovaTV 301 (NCT04697628) which is a randomized phase III trial comparing TV to investigator choice monotherapy chemotherapy in the second- or third-line metastatic setting. Overall Survival (OS) is the primary endpoint. This study is still recruiting patients at the time of this writing.

Table 2

ADCs landscape in late-phase clinical trial development

Ovarian

In November 2022, the FDA granted accelerated approval for another ADC, mirvetuximab soravtansine (MIRV), for patients with FRα-high, platinum-resistant ovarian cancer who have previously received between one and three systemic therapies.38 MIRV is an antibody drug conjugate that targets FRα and is conjugated to DM4, a microtubule toxin (Table 2). The associated clinical trial was the global, single-arm, phase III, SORAYA trial, which enrolled 106 patients with platinum-resistant ovarian cancer whose tumors expressed high levels of FRα and who had been treated with up to three prior regimens, at least one of which included bevacizumab. Full data from the study demonstrated that ORR was 32.4% (95% CI, 23.6 to 42.2), including five CR (5%) and 29 PR (27.6%).26 Median time to response was rapid (6 weeks), and two-thirds of patients demonstrated tumor reduction.26 At a data cut-off of March 3, 2022, the median DOR was 6.9 months (95% CI, 5.6 to 8.1), with seven responders on ongoing therapy.26 Furthermore, the median PFS was 4.3 months (95% CI, 3.7 to 5.1).26 Notably, the ORR of 32.4% demonstrated by the SORAYA trial is significant when comparing this data to historical studies of similar populations, which showed response rates of around 12%.9 This accelerated approval is pending confirmation with the completed phase III trial GOG 3045/ENGOT-ov55/MIRASOL (NCT04209855) which is a randomized phase III trial of MIRV vs investigator choice chemotherapy among patients with platinum resistant ovarian cancer and up to three prior lines of chemotherapy. Maintenance therapy (eg, bevacizumab, poly-ADP ribose polymerase (PARP) inhibitors) will be considered as part of the preceding line of therapy. Results for MIRASOL are expected in ythe first half of 2023.

An interesting learning that contributed to the design of the MIRASOL trial arose from developing the biomarker and companion diagnostics. There was a phase III trial of MIRV that preceded MIRASOL called FORWARD I clinical trial, which, despite consistent signals of efficacy of MIRV as compared with chemotherapy did not meet its primary endpoint of PFS.25 FORWARD I utilized immunohistochemistry that scored the tumor based on the percentage of positive cells seen at 10 times magnification (10 x scoring). The scoring that had been used in prior trials, utilized both the percentage of positive cells as well as the intensity (referred to as PS-2 scoring). A bridging study was done before FORWARD I to justify this “simpler” scoring methodology however, on re-evaluation of the negative FORWARD I study it was discovered that the 10 scoring had underperformed. Approximately 30% of the patients who were enrolled on the study were not actually eligible as they were FRα low (FORWARD I allowed FRα medium and high). In addition, approximately 50% of the tumors called FRα high on the 10 x scoring were actually FRα medium. PFS in the FRα cohort was a prespecified primary endpoint in FORWARD I and the dilution of this cohort with FRα medium tumors, diluted the benefit. An exploratory re-analysis of FORWARD I was performed using PS-2 scoring and the PFS results for the FRα high cohort were as expected in the original statistical design (unpublished data) and justified the redesign of MIRASOL using PS-2 IHC scoring. Therefore, it is anticipated that the MIRASOL trial, which included only patients with high FRα expressing tumors, appropriately selected, will show improvement in PFS. This would subsequently lead to FDA and European Medicines Agency (EMA) regular approval to patients with recurrent ovarian cancer.

Several ongoing studies are investigating MIRV for other indications, such as platinum-sensitive ovarian cancer. MIROVA (NCT04274426) is a multicenter, randomized, two-arm, open-label phase II trial in which clinical investigators are comparing platinum-based combination chemotherapy with an experimental arm of platinum-based chemotherapy plus MIRV, followed by MIRV monotherapy maintenance, until disease progression in recurrent, FRα high ovarian cancer. This study began in October 2021 and is forecast to complete in December 2023. PICOLO (NCT05041257) is an open-label, single arm, phase II study of MIRV in recurrent platinum-sensitive, ovarian cancer with high FRα expression with a primary endpoint of ORR. This study began in August 2021 and is forecast to complete in May 2023.

MIRV has been studied in combination with a variety of agents including carboplatin, pegylated liposomal doxorubicin (PLD), pembrolizumab and bevacizumab (bev). The results of MIRV+bev combinations are quite promising. For platinum resistant tumors, the ORR was 59% with a 9.4 median DOR and 9.7 months median PFS. For platinum sensitive tumors, the ORR was 69% with a 12.7 month median DOR and 13.3 month median PFS.39 The follow-up trial for these promising results is GOG 3078/ENGOT ov-29/GLORIOSA (NCT05445778), a randomized phase III trial which investigates MIRV+bev in combination in the maintenance setting of platinum sensitive recurrent OC. This study started in October 2022 and has an estimated completion date in March 2027.

ADCs Landscape in Early-phase Clinical Trial Development

There are numerous additional ADCs in early development for ovarian cancer.40 41 Given the impressive number of ADCs, not all the agents could be included in the text of this review. We have summarized them with their corresponding clinical trial numbers in Table 3 and Table 4.

Table 3

ADCs landscape in early phase clinical trial development

Table 4

Novel ADCs in gynecological cancers as of November 2022

Initial findings from the phase I study of anti-NaPi2b ADC, lifastuzumab vedotin, (LIFA) in platinum-resistant ovarian cancer showed promising antitumor efficacy in 37% of patients included in the study.42 However, the randomized phase II study of LIFA as compared with PLD failed its primary endpoint of PFS and development was discontinued. Before the discontinuation of LIFA development, a phase Ib study in combination with carboplatin with an option for LIFA maintenance was completed and demonstrated an acceptable safety profile of the combination. Median PFS for the combination was 10.71 months (95% CI: 8.54, 13.86), including confirmed complete and partial responses in 59% of patients.43 Although discontinued, this signal of efficacy in targeting the sodium gated phosphate transporter via ADCs set the stage for upifitamab rilsodotin.

Upifitamab rilsodotin (UpRi) represents a new generation of ADC targeting NaPi2b and contains an innovative DolaLock payload that facilitates both intracellular and extracellular drug diffusion throughout the tumor. Dolaflexin is a novel ADC technology with two key features: a higher drug-to-antibody ratio and a novel cytotoxic payload from the auristatin class.44 One of the first trials to investigate UpRi was the phase Ib expansion study investigating the drug in ovarian cancer. In this study, patients with platinum-resistant disease received intravenous UpRi once every 4 weeks until disease progression, unacceptable toxicity, or study discontinuation. The results showed that, regardless of NaPi2b expression, ORR and DCR were 23% (17/75) and 72% (54/75), respectively.45 This finding led to the registration enabling study UPLIFT (NCT04697472) which is a single-arm, phase II study of UpRi in platinum resistant ovarian cancer at the 36 mg/m2 dose level with ORR as the primary endpoint. This study completed accrual and results are expected in 2023. The starting dose of UpRi was 36 mg/m2 and increased to 43 mg/m2, but response rates were similar in both dose levels. Furthermore, duration of response did not correlate with NaPi2 expression.45 Two additional clinical trials with UpRi are UPGRADE (NCT04907968), an ongoing umbrella study enrolling patients with platinum-sensitive ovarian cancer, and UP-NEXT ENGOT-Ov71-NSGO-CTU /GOG-3049 (NCT05329545), a confirmatory phase III trial examining monotherapy maintenance with UpRi in platinum-sensitive recurrence. Results from these trials are expected in 2023 and 2024, respectively.

Farletuzumab eribulin consists of a humanized mAb targeting FRα and the established chemotherapy eribulin, which is used for metastatic breast cancer.46 Farletuzumab had promising antibody-mediated cytotoxicity, and several phase II clinical trials confirmed its efficacy at the beginning of the last decade.21 47 However, despite the effort that over 1000 patients were enrolled, the phase III study failed to confirm the clinical benefit observed in phase II studies.22 Given that the combination with eribulin will eventually potentiate the antibody-mediated cytotoxicity of farletuzumab, the new construct is considered to be very promising and was already investigated in a first-in-human phase I study in patients with FRα positive advanced solid tumors including ovarian cancer. Even though the study did not reach the maximum tolerated dose, the agent demonstrated clinical activity in patients.48 The study is ongoing to further assess antitumor activity and safety in each solid tumor type (NCT04300556).

Anetumab ravtansine is an IgG1 antibody targeting mesothelin and conjugated to the maytansinoid tubulin inhibitor DM4. The results from a first-in-human study using anetumab ravtansine were encouraging.49 This study enrolled patients with advanced mesothelin-expressing solid tumors, including malignant mesothelioma, ovarian, pancreatic, non-small-cell lung, and breast cancer. Antitumor activity was observed across all groups. The ovarian cancer cohort (n=61) showed 1 complete response, 4 partial responses, and 29 patients with stable disease.49 Interestingly, mesothelin expression was determined retrospectively, and no significant correlation between expression and antitumor activity could be established. In the more advanced randomized phase II study of bevacizumab and weekly anetumab ravtansine or weekly paclitaxel in platinum-resistant or refractory ovarian cancer, patients had positive expression of mesothelin in their archival tissue and a median of three prior lines of therapy (ranging between 1 and 9) with 42% patients receiving bevacizumab. Interestingly, the control arm with weekly paclitaxel reported 16 partial responses and an ORR of 55%.50 The findings demonstrated one complete and four partial responses, with an ORR of 18%, which proved not to be superior to paclitaxel.50 The estimated median PFS was 5.3 (95% CI: 3.7 to 7.4) months for anetumab ravtansine and 9.6 (95% CI: 7.4 to 17.4) months for paclitaxel with HR=1.7 (95% CI: 0.9 to 3.4). However, based on findings from the above-mentioned phase I study, several additional clinical trials in a variety of mesothelin-expressing solid tumors are underway.

Finally, another FR targeting ADC with encouraging data in ovarian cancer, STRO-002, was found to elicit encouraging responses in heavily pretreated patients with advanced ovarian cancer, according to interim data from the dose-expansion portion of the phase I trial (NCT03748186).51 The novelty of this agent is a precise drug-to-antibody ratio of 4 which contains the tubulin-targeting 3-aminophenyl hemiasterlin warhead SC209, a potent cytotoxin. Given the impressive heavily treated population enrolled in this study with a median number of six (range 2–11) prior systemic therapies, investigators reported a disease control rate of 74% at ≥12 weeks and 61% at ≥16 weeks. Following these findings, in August 2021, the FDA granted fast track designation to STRO-002 for use in patients with platinum-resistant epithelial ovarian, fallopian tube, or primary peritoneal cancer who have previously received 1 to 3 lines of systemic therapy.

Clinical Management of Ocular ADC Toxicities

The specific tumor-targeting of ADCs should lend itself to a low toxicity profile. However, ADCs are not completely benign and, while the toxicity profiles of traditional cytotoxic agents are well studied and their clinical management is included in cancer guidelines, the toxicity profiles of ADCs are not as well-defined. A particularly odd adverse event is ocular toxicity, which we will briefly describe in the following paragraphs. Other clinical toxicities of ADCs, such as hematologic, hepatic, and neurologic events, are discussed elsewhere and are beyond the scope of this review.52

A recent retrospective pooled analysis characterizing the tolerability profile of mirvetuximab soravtansine included 464 patients enrolled in three different studies investigating mirvetuximab soravtansine: phase I first-in-human,53 phase III FORWARD I,25 and phase III SORAYA.26 54 Importantly, the most frequent adverse events consisted mainly of gastrointestinal and ocular toxicities, which were low grade and reversible. Notably, only 1% of patients discontinued mirvetuximab soravtansine due to an ocular event. Grade 1 and 2 ocular adverse events were frequent, including blurred vision (42%), keratopathy (26%), and dry eye (22%). Grade 3+ ocular events, including blurred vision (3%), keratopathy (3%), dry eye (1%), were rarely reported. Interestingly, the toxicity profiles of different ADCs can vary dramatically; while grade 1 and 2 toxicities such as dry eye (23%), and keratitis (11%) were common among mirvetuximab soravtansine and tisotumab vedotin, conjunctivitis (26%) was limited to the latter.55

Counseling patients on the precautionary accessible measures to lessen eye dryness and reduce possible inflammation, such as avoiding the use of contact lenses and cleaning the exterior eye area, were proposed for the mitigation and management of ocular events associated with ADCs.55 Furthermore, gynecologic oncologists should become familiar with ocular symptoms that typically arise, and eye care doctors should be integrated in the oncologic team (Table 5). An ophthalmologic baseline eye exam and preservative-free lubricating eye drops for the duration of treatment have been successfully integrated in the clinical trials with ADCs.56 In addition, pharmacologic intervention such as use of steroid eye drops on the first day of infusion continuing for several days can ultimately help patients stay on therapy.56

Table 5

Common terminology criteria for ocular adverse events

Taken together, ADC ocular toxicities are manageable with ophthalmic care, including prophylactic, symptom management and dose modifications.25 57 Most ocular adverse events predictably occur during the first week of the second cycle of treatment (with some variation) on both mirvetuximab soravtansine and tisotumab vedotin. Moreover, these resolved in more than 90% of patients before the onset of their next cycle to grade 0 or 1.54 55 Given that targeted agents have low grade and less off-target toxicity relative to the standard chemotherapy, ADCs become attractive, especially for heavily pretreated patients who have previously experienced several other chemotherapy related toxicities such as neuropathy, neutropenia, and thrombocytopenia.58

Future Clinical ADCs Trial Considerations

A fresh or archival tumor biopsy is usually required to determine the expression of the target antigen as part of the eligibility criteria to enroll patients in clinical trials. Clinical trials which enrolled patients based on their FRα and NaPi2b tumor expression demonstrated a trend towards increased drug activity and clinical benefit with increased antigen expression.25 59 In contrast, subsequent trials indicated that antigen expression level is not always a key factor of a prior patient selection, and antigen expression evaluation in tumor samples has been limited to retrospective analyses.37 43

Despite these differences in trial design and patient selection, there are still several ongoing clinical trials in which investigators are limiting patient enrollment based on antigen expression . High FRα positivity is assessed by immunohistochemistry using the Ventana FOLR1CDx Assay (a high expression ≥75% of cells with PS2+ staining intensity). However, this indication might be re-considered in future trials due to findings supporting clinical efficacy of mirvetuximab soravtansine also in medium FRα expressing tumors.25 Second, the timing for testing additional markers such as FRα determined by immunohistochemistry should be optimized and performed as early as possible. In this aim, assessment of FRα could be added to the standard germline BRCA1/2 m and somatic homologous recombination deficiency testing, which are routinely performed at the first diagnosis work-up of ovarian cancer. Previous studies demonstrating that FRα expression observed in archival tissue is not significantly altered in response to chemotherapy,60 an observation consistent with subsequent findings in platinum-resistant ovarian cancer tissue samples,61 support early additional tumor testing.

Finally, another significant clinical challenge is the endogenous chemosensitivity of the tumor to the cytotoxic payload. Much attention will be turned toward the use of ADCs in combination with other therapy regimens. Early-phase drug development findings support the use of anti-angiogenic combination strategies, and the ongoing phase III trial GLORIOSA (NCT05445778) is testing this combination. In addition, there is significant interest in leveraging other cell-death pathways and processes, such as inducing DNA damage by combining ADCs with platinum-based therapy.39 Considering that cytotoxic payloads of ADCs are mainly tubulin inhibitors belonging to the same class of drugs as paclitaxel, synergistic drug combinations such as platinum-based therapy, paclitaxel, anti-angiogenic and checkpoint inhibitors may be more potent than a single cytotoxic drug packed in ADCs. Lastly, given the promising clinical efficacy of checkpoint inhibitors in cervical cancer,8 62 such agents are attractive targets for combination regimens with ADCs. Currently, tisotumab vedotin is being tested in recurrent or metastatic cervical cancer in combination with a PD-1 inhibitor in InnovaTV 205/ENGOT-cx8/ GOG-3024 (NCT03786081).

留言 (0)

沒有登入
gif