What’s new in the use of antibody-drug conjugates (ADCs) delivering pyrrolobenzodiazepine (PBD) DNA cross-linkers for cancer therapy?

KEYWORDS:

• ADC
• Antibody-drug conjugate
• pyrrolobenzodiazepine dimer
• PBD dimer
• loncastuximab tesirine
• camidanlumab tesirine
• Zynlonta^TM

1. Introduction

In this journal in 2020, I wrote a review on the previous decade of discovery and development of PBD dimer-containing antibody-drug conjugates (ADCs) [1]. Impressive pre-clinical data had resulted in 20 different ADCs entering the clinic. Despite some high-profile failures, significant clinical activity with manageable toxicity was demonstrated in several early-phase clinical trials with the potential for the first approval of an ADC containing the linker-payload tesirine. So what has been the progress over the subsequent three years?

2. Loncastuximab tesirine

The final results of the phase 1 dose escalation and dose expansion study of CD19-directed loncastuximab tesirine in relapsed or refractory (r/r) B-cell lymphoma were published in 2021 [2]. In 183 patients, receiving between 15 and 200 μg/kg, dose-limiting toxicities (all hematologic) were observed in four patients and the MTD was not reached. Efficacy is summarized in Table 1. A pivotal phase 2 study in r/r diffuse large B-cell lymphoma (DLBCL) (LOTIS-2) was also published in 2021 [3]. The most common treatment-emergent ≥ grade 3 adverse events were neutropenia, thrombocytopenia, and increased gamma-glutamyltransferase (GGT). Substantial single-agent activity was confirmed (Table 1), and long-term follow-up data have shown durable responses, including in patients with high-risk characteristics for poor prognosis, with a median duration of response not reached in patients who achieved a complete response (13.7 months in all responding patients) [6].

Table 1. Summary of efficacy and toxicity data for loncastuximab tesirine and camidanlumab tesirine in single-agent phase 1 and phase 2 clinical trials.

Clinical Trial	Indication	Number of patients	Dose ($\mu$g/kg)	ORR (%)	CR (%)	mDoR (Months)	mPFS (Months)	mOS (Months)	Most common grade ≥3 TEAEs
Loncastuximab tesirine
Phase 1 [2]	R/R B-cell non-Hodgkin lymphoma	180	15–200	45.6	26.7	5.4	3.1	8.3	Neutropenia Thrombocytopenia Increased GGT
	Diffuse large B-cell lymphoma	139	15–200	42.3	23.4	4.5	2.8	7.5
	Mantle cell lymphoma	15	15–200	46.7	33.3	n.r	4.8	n.r.
	Follicular lymphoma	14	15–200	78.6	64.3	n.r.	n.d	n.r.
Phase 2 [3]	R/R Diffuse large B-cell lymphoma	145	150/75^a	48.3	24	10.3^c (13.4 with CR)	4.9	9.9	Neutropenia Thrombocytopenia Increased GGT
Camidanlumab tesirine
Phase 1 [4]	R/R Lymphoma	130	5–300	58	29				Increased GGT Maculopapular rash Anemia
	Non-Hodgkin lymphoma	53	5–300	38	9
	Classical Hodgkin lymphoma	77	5–300	71	42	6.6	6.8	n.r.
	Classical Hodgkin lymphoma	37	45	86	49	7.2	7	n.r.
Phase 2 [5]	R/R Classical Hodgkin lymphoma	117	45/30^b	70.1	33.3	13.7 (14.5 with CR)	9.1 (15.9 with CR)		Thrombocytopenia Anemia, Neutropenia Hypophosphatemia Maculopapular rash

n.r. – not reached. n.d. – not determined due to low numbers, ORR – overall response rate (CR + PR), CR – complete responses, mDoR – median duration of response, mPFS – median progression free survival, mOS – median overall survival. TEAE – treatment-emergent adverse events, R/R – relapsed or refractory. GGT – gamma-glutamyltransferase.

^a150 $\mu$g/kg for two cycles followed by 75 $\mu$g/kg for subsequent cycles.

^b45 $\mu$g/kg for two cycles followed by 30 $\mu$g/kg for subsequent cycles.

^cMedian time to first response was 41 days.

In November 2020, the US FDA accepted a Biologics License Application (BLA) from ADC Therapeutics and granted priority review, which resulted in accelerated approval as Zynlonta^TM in April 2021. Full EC approval followed in December 2022, a BLA was granted priority review by the China National Medical Products Administration based on positive results of a phase 2 bridging study in 2023, and a phase 1/2 bridging study has been initiated in Japan.

LOTIS-5 is a confirmatory Phase 3 study to evaluate loncastuximab tesirine + rituximab (Lonca-R) vs. standard immunotherapy of rituximab + gemcitabine + oxaliplatin (R-GemOx) in patients with r/rDLBCL. The results of the nonrandomized safety lead-in part of this study were presented in 2022 [7] with encouraging activity observed (75% ORR, 40% CR) with no new safety signals. The randomized part of this study (330 patients) is due to complete recruitment in 2024.

A phase 1b study of loncastuximab tesirine in combination with other anticancer agents in r/r B-NHL is ongoing (LOTIS-7); however, a phase 2 study of Lonca-R in previously untreated unfit/frail patients with DLBCL (LOTIS-9) was recently discontinued due to potentially excessive respiratory-related events. The company release [8] stated that ‘These respiratory-related treatment-emergent adverse events (TEAEs) included seven Grade 5 fatal events and five Grade 3 or Grade 4 respiratory-related TEAEs. As per the investigator assessment, 11 of the 12 events (including six of the seven Grade 5 fatal events) were individually assessed as unlikely or unrelated to the study drug.’

3. Camidanlumab tesirine

The phase 1 dose escalation and dose expansion study of camidanlumab tesirine targeting CD25 in patients with r/r lymphoma was published in 2021 [4]. Patients received between 3 and 150 μg/kg, dose-limiting toxicities were observed in five patients and the MTD was not reached. Grade 3 or higher treatment-emergent events were increased GGT, maculopapular rash, and anemia. Activity was observed in both non-Hodgkin and classical Hodgkin lymphomas (Table 1), with the latter giving an overall response rate of 71% (86% at the expansion dose of 45 μg/kg). Overall response rate was similar in patients who had failed brentuximab vedotin, PD-1 blockade, and/or hematopoietic stem-cell transplantation (HSCT).

The initial results from the pivotal phase 2 study in r/r classical Hodgkin lymphoma (cHL) were reported in 2022 [5]. Safety was consistent with previous findings and the impressive single-agent activity was confirmed (ORR 70.1%, 33.3% CR) (Table 1). Fourteen patients received HSCT following treatment with camidalumab tesirine. The most common grade ≥ 3 treatment-emergent adverse events were hematological (thrombocytopenia, anemia, neutropenia), hypophosphatemia and maculopapular rash. Across both trials, a small number of patients (12/247) developed Guillain–Barré syndrome.

Although camidanlumab tesirine entered clinical trials before loncastuximab tesirine, early-phase development was slower. With response rates generally higher than loncastuximab tesirine (Table 1), a BLA filing for accelerated approval in a third-line setting in cHL was anticipated. The accelerated approval pathway window had, however, closed with the FDA now requiring a confirmatory phase 3 study to be underway (and likely fully enrolled) at the time of submission.

4. Other PBD dimer-containing ADCs

4.1. Hematological targets

An update on the phase 1 study of ADCT-602 targeting CD22 in relapsed or refractory B-cell ALL was presented in 2022 [9]. The phase 1 trial in multiple myeloma of MEDI2228 targeting BCMA has been discontinued and not yet published.

4.2. Solid tumor targets

The phase 1 study of MEDI3726 (ADCT-401) targeting PSMA as monotherapy for the treatment of metastatic castration-resistant prostate cancer was published in 2021 [10]. Doses ranged from 15 to 300 μg/kg every three weeks. An MTD was not identified but treatment-related adverse events, primarily skin toxicities and effusions, were commonly observed. Clinical responses occurred at higher doses (composite response rate 12.1%) but were not durable as patients had to discontinue treatment due to TRAEs. Pharmacokinetics was nonlinear with a very short half-life (0.3–1.8 days) compared to other clinical PBD-ADCs.

Mipasetamab uzoptirine (ADCT-601) targeting AXL, which delivers the same PBD dimer as tesirine (SG3199), but site-specifically conjugated using GlycoConnect™ technology to a valine-alanine linker containing HydraSpace™ [11], is currently in phase 1b dose escalation focusing on patients with NSCLC and sarcoma either as a single agent or in combination with gemcitabine. A tesirine-containing ADC targeting the novel antigen KAAG1 (ADCT-901) entered the clinic in 2021 against advanced solid tumors. TR1801-ADC targeting cMet is listed as active not recruiting.

5. Conclusion

It is now 10 years since the first PBD dimer-containing ADC entered the clinic, and the last 3 years have seen positive outcomes from pivotal phase 2 studies of two agents leading to the first approval.

6. Expert opinion

The development of loncastuximab tesirine has been relatively rapid from first patient in 2016 to accelerated approval for r/r DLBCL in 2021. Many patients in this setting do not respond to salvage therapy and cannot proceed to autologous HSCT. Response rates to CD19 CAR T-cell therapy can be high, but this is not suitable for all patients and many do not have a durable response. The different toxicity profile to other approved therapies and fast time to response with durable activity, even in difficult-to-treat subgroups, makes loncastuximab tesirine an important addition to the treatment landscape. Responses are observed in patients relapsing after CD19 CAR T-cell therapy [3], and a subgroup of patients in the phase 2 study received subsequent consolidation treatments including HSCT or CAR T-cell therapy suggesting the potential as a bridge to other treatments [3].

The successful phase 2 of camidanlumab tesirine [5] has provided further clinical proof of concept for this ADC class, and it is hoped that a route to approval and patient benefit can be found for this highly active agent. PBD ADCs can clearly have significant single-agent activity in a relapsed/refractory setting, but the ability to combine with standard treatments, or rationally with other agents (e.g. gemcitabine [12]), with the potential to reduce dosing and as a result decrease toxicity, will be crucial in the long-term therapeutic strategy for these agents.

With increasing clinical experience, common toxicities related to this ADC class are becoming evident. These include hematological, liver enzyme abnormalities, fatigue, skin toxicities, and effusions, the latter two generally becoming evident after several cycles of treatment and can be cumulative [2–5]. Most of these adverse events are reversible and manageable with dose delays and/or reductions. Clinical experience has also further highlighted the importance of optimizing dose scheduling (including dose reduction in later cycles as in both pivotal phase 2 studies [3,5]) to reduce the potential for delayed toxicities. Target-related toxicities have also been observed, most notably the rare occurrence of Guillain-Barré syndrome or polyradiculopathy in cHL patients treated with camidalumab tesirine targeting CD25 [4,5]. Interestingly, this has not been observed in non-HL [4] or leukemia patients [13] treated with this agent suggesting it may be both target and indication dependent.

The choice of patient population is clearly key for agents such as loncastuximab tesirine as highlighted by the recent discontinuation of the phase 2 study of Lonca-R in previously untreated unfit/frail patients with DLBCL (LOTIS-9). This has some analogy to the experience with vadustuximab talirine, where, despite impressive preclinical and early-phase clinical data in AML, the phase 3 CASCADE trial was terminated early due to toxicity in this older population of HMA-eligible patients who are more susceptible to toxicity [14].

The initial promise of ADCs delivering PBD dimers has borne fruit with the major successes to date being in the hematological setting, where accessibility to ADC, high and relatively homogeneous target antigen expression levels, and inherent sensitivity to PBD dimers such as SG3199 [15] may all play a role. Ongoing studies will determine if a satisfactory therapeutic index can be consistently achieved in a solid tumor setting. It will be interesting to see if the use of the more stable linker in mipasetamab uzoptirine (ADCT-601) [11] has a positive effect, and future strategies employing lower potency PBD dimers (which may necessitate an increased drug-to-antibody ratio) may also have potential [16].

Declaration of interest

In addition to his academic positions, JA Hartley was a founder scientist of Spirogen Ltd and has an equity interest in ADC Therapeutics. He is also on the Scientific Advisory Board of Spirea Ltd. The author has no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This paper was not funded.