Mitochondrially targeted tamoxifen in patients with metastatic solid tumours: an open-label, phase I/Ib single-centre trial

Summary

Background

Mitochondria present an emerging target for cancer treatment. We have investigated the effect of mitochondrially targeted tamoxifen (MitoTam), a first-in-class anti-cancer agent, in patients with solid metastatic tumours.

Methods

MitoTam was tested in an open-label, single-centre (Department of Oncology, General Faculty Hospital, Charles University, Czech Republic), phase I/Ib trial in metastatic patients with various malignancies and terminated oncological therapies. In total, 75 patients were enrolled between May 23, 2018 and July 22, 2020. Phase I evaluated escalating doses of MitoTam in two therapeutic regimens using the 3 + 3 design to establish drug safety and maximum tolerated dose (MTD). In phase Ib, three dosing regimens were applied over 8 and 6 weeks to evaluate long-term toxicity of MitoTam as the primary objective and its anti-cancer effect as a secondary objective. This trial was registered with the European Medicines Agency under EudraCT 2017-004441-25.

Findings

In total, 37 patients were enrolled into phase I and 38 into phase Ib. In phase I, the initial application of MitoTam via peripheral vein indicated high risk of thrombophlebitis, which was avoided by central vein administration. The highest dose with acceptable side effects was 5.0 mg/kg. The prevailing adverse effects (AEs) in phase I were neutropenia (30%), anaemia (30%) and fever/hyperthermia (30%), and in phase Ib fever/hyperthermia (58%) together with anaemia (26%) and neutropenia (16%). Serious AEs were mostly related to thromboembolic (TE) complications that affected 5% and 13% of patients in phase I and Ib, respectively. The only statistically significant AE related to MitoTam treatment was anaemia in phase Ib (p = 0.004). Of the tested regimens weekly dosing with 3.0 mg/kg for 6 weeks afforded the best safety profile with almost all being grade 1 (G1) AEs. Altogether, five fatalities occurred during the study, two of them meeting criteria for Suspected Unexpected Serious Adverse Events Reporting (SUSAR) (G4 thrombocytopenia and G5 stroke). MitoTam showed benefit evaluated as clinical benefit rate (CBR) in 37% patients with the largest effect in renal cell carcinoma (RCC) where four out of six patients reached disease stabilisation (SD), one reached partial response (PR) so that in total, five out of six (83%) patients showed CBR.

Interpretation

In this study, the MTD was established as 5.0 mg/kg and the recommended dose of MitoTam as 3.0 mg/kg given once per week via central vein with recommended preventive anti-coagulation therapy. The prevailing toxicity included haematological AEs, hyperthermia/fever and TE complications. One fatal stroke and non-fatal G4 thrombocytopenia were recorded. MitoTam showed high efficacy against RCC.

Funding

Smart Brain Ltd.

Translation

For the Czech translation of the abstract see Supplementary Materials section.

Research in context.

Evidence before this study

We searched for articles in PubMed up to December 1, 2022, using the terms Mitochondrially Targeted Anti-cancer Agents, Tamoxifen, Metastatic Tumors, and Phase I/Ib trial, in title or abstract. After manual literature search we did not find any relevant reports. We decided to test clinically the premise that mitochondrially targeted anti-cancer agents provide benefit to patients with metastatic solid tumors, and have launched Phase I/Ib clinical trial.

Added value of this study

This study provides evidence for safety of the novel, first-in-class mitochondria-targeting anti-cancer agent, MitoTam, and points to potential therapeutic benefit in metastatic tumours. MitoTam showed benefit evaluated as clinical benefit rate (CBR) in 37% patients with the largest effect in renal cell carcinoma (RCC) where four out of six patients reached disease stabilisation (SD), one reached partial response (PR) so that in total, five out of six (83%) patients showed CBR.

Implications of all the available evidence

The result of this phase I/Ib clinical trial provides evidence of safety of MitoTam. Of the various diagnoses, patients with RCC experienced the highest benefit rate. These data will be used when planning a phase II trial.

Introduction

Mitocans are a group of anti-cancer agents that act by targeting mitochondria.¹ Of particular interest are small compounds tagged with triphenylphosphonium (TPP⁺), a group previously used for mitochondrial delivery of coenzyme Q.² Due to the presence of the lipophilic delocalised cationic group, mitocans tagged with TPP⁺ readily cross phospholipid bilayers with membrane potential, including both the plasma membrane and inner mitochondrial membrane (IMM). Based on the relatively high potential across the IMM, that can be higher in cancer cells than in their non-malignant counterparts,³ TPP⁺ acts as a molecular anchor, accumulating and positioning these biologically active agents at the interface of the IMM and mitochondrial matrix.^1,4

We have synthetised and tested TPP⁺-tagged mitocans with molecular targets in the mitochondrial respiratory chain, of which MitoTam that interferes with complex I (CI)-dependent respiration^5,6 was evaluated in pre-clinical settings. Molecular modelling and biochemical data revealed that MitoTam blocks movement of electrons generated by oxidation of NADH at the catalytic centre of CI to the Q-domain-bound ubiquinone, their physiological electron acceptor. This generates superoxide that triggers a cascade of reactions culminating in preferential demise of cancer cells.⁵ Moreover, localization of MitoTam at IMM induces depolarization of mitochondrial membrane, affecting mitochondrial function and integrity.^6,7 This dual mechanism of action gives MitoTam its unique properties that make it effective against a broad spectrum of cancer cells.

Motivated by our results with MitoTam,⁵ we launched phase I/Ib MitoTam trial (MitoTam-01, EudraCT 2017-004441-25) in patients with various types of metastatic solid tumours after terminated standard therapy. Here we present results of phase I/Ib clinical trial of MitoTam, showing its safety and potential clinical benefit.

Methods

Study design and participants

The MitoTam-01 study was designed as an open-label, single-arm, non-randomised, single centre phase I/Ib trial carried out in the Department of Oncology, General Faculty Hospital, Charles University (Prague, Czech Republic), and included patients of 18–75 years with different metastatic tumours (performance status 0–2 according to the Eastern Cooperative Oncology Group, ECOG; life expectancy >3 months). All patients had been treated for oncological diseases, and their therapy was terminated or they refused standard therapy. Inclusion/exclusion criteria are listed in the Appendix (Clinical Study Protocol and Table S1, p 3). Patient ID (e.g., P1-1.0) consists of the number of patient (P1) and dose of MitoTam (1.0 mg/kg).

Phase I followed the 3 + 3 design, where patients received gradually increasing doses of MitoTam. In case dose-limiting toxicity (DLT) was observed, the initial three-patient cohort was expanded to six patients. The doses studied were 0.25–6.0 mg/kg until DLT was reached; the dose before DLT was declared MTD. For 0.25–3.0 mg/kg, one escalation cycle consisted of three doses of MitoTam administered within one week on days 1 (D1), D3 and D5, followed by observation on D6 and D7, and by follow-up visits on D14 and D28. After reaching 3.0 mg/kg, the treatment was altered to one administration per week on D1; this was used for 4.0–6.0 mg/kg and included observations on D2 and D3 plus follow-up visits on D8, D14 and D28 (Table 1; Appendix Fig. S1, p 2).

Table 1.

Treatment regimens in phase I and phase Ib trial.

Phase I n = 37
Schemes of M administration	Dosages of M (mg/kg)/No of pts	Screening period
1 cycle/3 doses, D1, D3, D5 in 1 week	0.25/4, 0.5/6, 1.0/3, 1.5/3, 2.25/6, 3.0/5	05/2018–07/2019
1 cycle/1 dose, applied on D1	4.0/3, 5.0/6, 6.0/1	10/2019–03/2020

Phase Ib n = 38
Regimen of M administration	Dosages of M (mg/kg)/No of pts	Screening period
4 biweekly cyclesa, applied on D1+D3+D5	1.0/20	01/2019–07/2019
6 weekly cyclesa, applied on D1	3.0/9	11/2019–03/2020
6 weekly cycles, applied on D1	4.0/9	02–07/2020

D = day. M = MitoTam.

^aThere was a possibility to repeat M over the 4 or 6 cycles in pts with stable disease or partial response according to CT examination.

In phase Ib, long-term administration Regimen 1 consisted of four biweekly cycles. In the first week, MitoTam (1.0 mg/kg) was administered on D1, D3 and D5, the second week was recovery period. Treatment Regimens 2 and 3 consisted of six weekly cycles with MitoTam (3.0 and 4.0 mg/kg, respectively) administered once per week on D1 (Table 1; Appendix Fig. S1, p 3). All regimens included follow-up visits on D14 and D28 after treatment. When clinical benefit and low toxicity were observed on D28 of the follow-up period, patients were allowed to repeat the treatment for up to 16 cycles for Regimen 1 and up to 12 cycles for Regimen 2. Patients who finished 1 cycle of MitoTam therapy in phase I were allowed to continue in phase Ib irrespective of the result of the control CT scan in case they met inclusion criteria and in case phase Ib was open at the time of the patient screening.

The trial was performed according to the current version of the World Medical Association (WMA) Declaration of Helsinki, and was conducted in agreement with the International Conference on Harmonisation (ICH) guidelines on Good Clinical Practice (GCP). It was reviewed and approved by the Clinical Centre Ethical Committee of the State Institute for Drug Control (Prague, Czech Republic) and registered with the European Medicines Agency under EudraCT 2017-004441-25. All patients signed written informed consent before undergoing any study-related procedure.

Procedures

In phase I at dose of 0.25 and 0.5 mg/kg, MitoTam was administered into the peripheral vein (10/37 patients). The other 27 patients at doses of 1.0–6.0 mg/kg were given MitoTam into the central vein via Peripheral Inserted Central (PIC) catheter, central vein catheter or port catheter because of high frequency of peripheral veins inflammation. In phase Ib, MitoTam was delivered via central vein in all 38 patients. Detailed list of examinations is in the Appendix (Clinical Study Protocol and Tables S2–S5, pp 4–7).

AEs were classified according to the National Cancer Institute Common Terminology Criteria for Adverse Effects (NCI-CTCAE) v.4.0 and according to the protocol, which pre-defined five-level evaluation of peripheral phlebitis, four-level evaluation of central vein phlebitis, two grading systems for evaluation of liver toxicity in patients with and without liver metastases, and assessment of hyperthermia (Appendix Tables S6 and S7, pp 8–9). Hyperthermia refers to temperature of 37.1–37.9 °C, fever to temperature above 37.9 °C.

Outcomes

Phase I had four primary objectives: 1) MitoTam safety, 2) identification of MTD, 3) establishing appropriate doses for phase Ib, and 4) description of pharmacokinetic (PK) profile of MitoTam. The primary objective of phase Ib included long-term safety together with PK analysis after repeated MitoTam administration. The secondary objective was classification of the clinical benefit rate (CBR) according to RECIST 1.1⁸ and identifying a possible target group of malignancies for MitoTam phase II trial. CBR was defined as percentage of patients who experienced complete response (CR), partial response (PR) or disease stabilisation (stable disease, SD), but not progressive disease (PD).

Statistical analysis

In phase I/Ib, the focus was on studying toxicity by means of AE analysis. Logistic regression was used to analyse presence of possible AEs, where size of the dose was considered a covariate (a dose–response model). The analysis of PK included the following parameters: half-time of elimination from plasma (t_1/2), area under the curve (AUC) to the last quantifiable concentration, and maximal observed concentration in serum (C_max). Longitudinal data analysis for all three parameters was performed using the random effect mixed model (generalised linear mixed model) for repeated observations in patients. In phase Ib, the secondary objective was CBR presented by an ordinal variable (ordered levels CR<PR<SD<PD). Cumulative link model for ordinal regression was applied. All patients were subjected to CBR evaluation. In phase Ib, six patients missed the CT scan after treatment because of disease progression and/or death, and they were considered PD patients.

Sample size calculation

For the logistic regression with a continuous dose covariate, significance level of 0.05 and power equal to 0.85 were assumed for one-sided test of the dose effect significance.⁹ If 0.9 is the considered change in log odds for the difference between the mean of dose and at one standard deviation above the mean, then 36 patients are needed in phase I. For ordinal regression in phase Ib, the design strategy for single-arm clinical trials was based on literature.¹⁰ Cut-off for the unacceptably low efficacy was set to 0.90 and cut-off for the unacceptably high adverse outcome rate to 0.90, which leads to 38 patients. The R statistical software (v.4.2.0) with libraries nlme (v.3.1-157) and ordinal (v.2019.12-10) was used for analyses.

Role of the funding source

The sponsor provided MitoTam and participated in the study design and in data collection and analysis. The course of the study and results were supervised by and consulted with independent Data and Safety Monitoring board (DSMB). Z.B., L.P., and L.W. had full access to all the data. Z.B., L.W., and J.N. had final responsibility for the decision to submit the manuscript for publication.

Results

From May 23, 2018 to July 22, 2020, 37 patients were enrolled into phase I and 38 in phase Ib. All patients received at least one dose of MitoTam (Table 2). 18 of 37 patients treated in phase I transitioned to phase Ib (Fig. 1). Baseline characteristics of patients were as follows: median age was 62 years in phase I and 63 years in phase Ib, all patients in phase Ib had visceral metastases and 70% of them were pre-treated with 3 or more lines of palliative systemic therapy (Table 2). Patients enrolled in the study were diagnosed with more than 10 types of solid tumours, the largest groups being colorectal cancer (CRC) and renal cell carcinoma (RCC) (Appendix Table S8, p 9).

Table 2.

Baseline characteristics of enrolled patients.

	PHASE I, n = 37	PHASE Ib, n = 38
Age, years	62.0 (40–72)	63.0 (40–73)
Gender
Female	19 (51%)	15 (40%)
Male	18 (49%)	23 (60%)
Eastern Cooperative Oncology Group performance status score
0	18/37 (48%)	23/38 (61%)
1	15/37 (41%)	13/38 (34%)
2	4/37 (11%)	2/38 (5%)
No of previous therapy lines in metastatic setting
0	0	1a/38 (3%)
1	1/37 (3%)	4/38 (10%)
2	8/37 (22%)	6/38 (16%)
3	12/37 (32%)	13/38 (34%)
4	11b/37 (30%)	10b/38 (26%)
5	5/37 (13%)	2/38 (5%)
6 and more	0	2/38 (5%)
Type of metastases
Visceral	36/37 (97%)	38/38 (100%)
Skeletal	7/37 (19%)	11/38 (29%)
Central nervous system	2/37 (5%)	1/38 (3%)

^aWoman (with breast cancer without treatment effect of study drug) signed a reverse with standard treatment.

^bOne man treated with four different types of alternative treatment but he underwent no standard palliative therapy.

Fig. 1.

Trial profile. The flow chart shows the number of patients enrolled in phase I and phase Ib trial and the number of patients transferred from phase I to phase Ib. Patients who finished 1 cycle of MitoTam therapy in phase I were allowed to continue in phase Ib irrespective of the result of control CT scan if they met inclusion criteria and if phase Ib was open at the time of patient screening. Overall, 18 patients made the phase I to phase Ib transition. Regimen 1: 4 biweekly cycles/8 weeks, maximally 16 cycles/32 weeks (timing of the trial see Table 1). Regimen 2 and 3: 6 weekly cycles/6 weeks, maximally 12 cycles/12 weeks in Regimen 2 and 6 cycles in Regimen 3 (timing of the trial see Table 1). ∗COVID-19 restrictions, worsening ECOG, disease progression, progressing anemia, atrial fibrillation. ^&Re-screening performed if some cohort in phase Ib was open and informed consent was obtained from patient. ^£Diagnoses of transiting patients: 2 x lungs, 2 x pancreas, 1 x endocrine gland, 1 x unknown origin, 5 x kidney, 5 x colorectal, 1 x bile ducts, 1 x ovary; ⁺additional treatment offered to patients with stable disease (SD) or partial response (PR) if other conditions have been met (see Fig. 2). ^$In six patients CT examination was not performed because of death or worsening of clinical condition (see Table S15 in Appendix) and these patients were included in the group with progressive disease (PD).

The number of patients enrolled in phase I for doses of 0.5, 2.25 and 5.0 mg/kg was increased due to AEs linked to DLT observed in at least one patient per dose (Appendix Table S9, p 10). Due to the frequency of haematological (leucocytes, neutrophils, thrombocytes) AEs at 5.0 mg/kg and since the AEs were expected to be more pronounced in patients administered 6.0 mg/kg, 5.0 mg/kg was considered MTD.

In phase I, safety of three doses per week (0.25–3.0 mg/kg) was evaluated in 27 patients and that of one dose per week (4.0 and 5.0 mg/kg) in 10 patients (summarised in Table 3 and Appendix Table S10, p 10). The most common was reversible haematological toxicity in 49% (18/37), anaemia in 30% (11/37), neutropenia in 30% (11/37) and thrombocytopenia in 19% (7/37) patients. Both incidence and grading of haematological AEs increased with increasing dosage of MitoTam. The only exception was anaemia in four patients treated in first two cohorts (0.25 and 0.5 mg/kg). Because there were no other cases of anaemia, leucopenia, neutropenia or thrombocytopenia in next two cohorts (1.0 and 1.5 mg/kg), we consider the four cases of anaemia a co-incidence with the advanced disease, the condition of patients and/or with complications of the disease. Isolated hyperthermia was detected in 5% (2/37) and fever in 24% (9/37) patients. Phlebitis and/or phlebothromboses of peripheral veins occurred in 19% (7/37) patients, six out of seven patients in first two MitoTam doses of 0.25 and 0.5 mg/kg until October 2018. Inflammation of peripheral veins was prevented by administration of MitoTam via central vein since November 2018 and by flushing the vein with 250 ml of saline after each MitoTam administration since January 2019 based on the DSMB recommendation. Following these preventive measures, no TE complications linked to MitoTam in phase I trial were observed prior to its termination in February 2020. The probability of haematological toxicity and hyperthermia increased with increasing dose of MitoTam. Majority of the total of 103 statistically evaluated AEs were observed in at least one patient and at least once within two weeks after the last dose. No AE was significantly related to MitoTam treatment (Appendix Table S11, p 11; p-values > 0.05).

Table 3.

Summary of adverse events of phase I trial by CTCAE (v.4) grade in all treated patients (n = 37).

	Any grade	Grade 1–2	Grade 3	Grade 4
Anaemia	12 (32%)	10 (27%)	2 (5%)	–
Leucopenia	23 (62%)	16 (43%)	7 (19%)	–
Neutropenia	20 (54%)	17 (46%)	3 (8%)	–
Thrombocytopenia	9 (24%)	7 (19%)	1 (3%)	1 (3%)§
Fevera	9 (24%)	9 (24%)	–	–
Nausea	2 (5%)	2 (5%)	–	–
ALT + AST elevation	5 (14%)	5 (14%)	–	–
Fatigue	2 (5%)	2 (5%)	–	–
Elevated creatinine	2 (5%)	2 (5%)	–	–
Urinary infection	1 (3%)	1 (3%)	–	–
Phlebitis	6 (16%)	6 (16%)	–	–
Phlebothrombosis∗	2 (5%)	2 (5%)	–	–
Enterorrhagia∗	1 (3%)	–	1 (3%)	–
QTC Interval extension	1 (3%)	1 (3%)	–	–
Loss of appetite	1 (3%)	1 (3%)	–	–

Data are n (%) where n represents number of patients. The same patient could have experienced different grades of adverse events and therefore might be presented in more than one column. Two phlebothromboses of peripheral veins (G1 according to CTCAE) were reported as SAE∗ due to protocol definition (see Appendix, p 10). One G3 enterorrhagia (SAE∗) and G4 thrombocytopenia (SUSAR^§), occurred in patient with small bowel cancer.

ALT = alanine aminotransferase. AST = aspartate aminotransferase. CTCAE, v4.0 = Common Terminology Criteria for Adverse Events, version 4.0. SAE = Serious Adverse Event. SUSAR = Suspected Unexpected Serious Adverse Events Reporting.

^aExcept of nine cases of fever four cases of hyperthermia (defined as temperature 37.1–37.9 C according to the protocol) occurred (not seen in table). Hyperthermia and fever were detected in 11 (30%) of patients together.

In phase Ib, the most common AEs (at least one occurrence during long-term treatment) were fever and/or hyperthermia in 58% (22/38), haematological AEs in 50% (19/38), anaemia in 26% (10/38), neutropenia in 16% (6/38) and thrombocytopenia in 11% (4/38) patients (Table 4 and Appendix Table S12, p 12). The only significantly MitoTam dose-dependent AE from the total of 169 evaluated AEs was anaemia (p = 0.004); its grade increased with MitoTam dose (Appendix Table S13, p 13; p-values > 0.05). The ratio for G1-2 anaemia in Regimen 1 vs. Regimen 2 vs. Regimen 3 was 15% vs. 22% vs. 33% patients. Similarly, the ratio for G1-2 neutropenia was 15% vs. 0% vs. 33%, the ratio for G1-2 thrombocytopenia was 10% vs. 0% vs. 22%. Incidence of G3 anaemia in Regimen 1 vs. Regimen 2 vs. Regimen 3 was 5% vs. 0% vs. 11%. No G3 neutropenia/leucopenia or thrombocytopenia was recorded in either Regimen. We conclude that single haematological AEs were low in Regimen 1 (up to 15%), minimal in Regimen 2 (only anaemia G1-2 in 22% of patients) and higher (dose-related) only in Regimen 3. TE complications were detected in 13% (5/38) patients and are described below.

Table 4.

Summary of adverse events of phase Ib trial by CTCAE (v.4) grade in all treated patients (n = 38).

	Any Grade	Grade 1–2	Grade 3	Grade 5
Anaemia	12 (32%)	10 (26%)	2 (5%)	–
Leucopenia	10 (26%)	10 (26%)	–	–
Neutropenia	9 (24%)	9 (24%)	–	–
Thrombocytopenia	6 (16%)	6 (16%)	–	–
Fevera	27 (71%)	27 (71%)	–	–
Nausea	8 (21%)	8 (21%)	–	–
Vomiting	5 (13%)	5 (13%)	–	–
Diarrhea	4 (11%)	4 (11%)	–	–
ALT + AST elevation	2 (5%)	2 (5%)	–	–
Elevated GGT	1 (3%)	1 (3%)	–	–
Elevated TSH	1 (3%)	1 (3%)	–	–
Elevated creatinine	4 (11%)	4 (11%)	–	–
Hypokalaemia	1 (3%)	1 (3%)	–	–
Hypocalcemia	1 (3%)	1 (3%)	–	–
Headache	3 (8%)	3 (8%)	–	–
Joint pain	2 (5%)	2 (5%)	–	–
Muscle pain	1 (3%)	1 (3%)	–	–
Back pain	1 (3%)	1 (3%)	–	–
Conjunctivitis	2 (5%)	2 (5%)	–	–
Loss of appetite	8 (21%)	8 (21%)	–	–
Weight loss	6 (16%)	6 (16%)	–	–
Fatigue	5 (13%)	5 (13%)	–	–
Insomnia	1 (3%)	1 (3%)	–	–
Dyspnoea	4 (11%)	4 (11%)	–	–
Decrease of ejection fraction	1 (3%)	1 (3%)	–	–
Sweating	1 (3%)	1 (3%)	–	–
Hypertension	1 (3%)	1 (3%)	–	–
Thigh redness	1 (3%)	1 (3%)	–	–
Skin erythema due to paravasation	2 (5%)	2 (5%)	–	–
DVT and PE∗	4 (11%)	–	4 (11%)	–
Phlebothrombosis of peripheral vein∗	1 (3%)	1 (3%)	–	–
Perimaleolar oedema	1 (3%)	1 (3%)	–	–
Stroke§	1 (3%)	–	–	1 (3%)

Data are n (%) where n represents number of patients. The same patient could have experienced different grades of adverse events and therefore might be presented in more than one column. One phlebothrombosis of peripheral vein (G1 according to CTCAE) and four deep vein thrombosis (DVT) complicated by pulmonary embolism (PE) were reported as SAE∗ due to protocol (see Appendix, p 12) and/or CTCAE definition. One treatment related death, as a result of G5 ischaemic stroke (SUSAR^§) occurred in patient with colorectal cancer. No G4 AE were observed in the Phase Ib. In one case of ALT/AST elevation was diagnosed also bilirubin elevation.

GGT = gamma-glutamyl transferase. SAE = Serious Adverse Event. SUSAR = Suspected Unexpected Serious Adverse Events Reporting. TSH = thyroid-stimulating hormone.

^aExcept of twenty-seven cases of fever fourteen cases of hyperthermia (defined as temperature 37.1–37.9 C according to the protocol) occurred. Fever and/or hyperthermia were detected in 22 (58%) patients together.

There were 8 episodes of treatment interruption (stopping and subsequent resuming) due to AEs related to MitoTam in 6/20 (30%) patients in Regimen 1, and 8 episodes in 5/9 (56%) patients in Regimen 3. Serious AEs (SAEs) potentially related to MitoTam were observed in 8/75 (11%) patients in both parts of the study (Tables 3 and 4). They included three cases of peripheral phlebothromboses, four cases of deep vein thrombosis (DVT) and asymptomatic pulmonary embolism (PE), one case of ischaemic stroke (Appendix Table S14, p14) and one case of enterorrhagia. Thromboses of peripheral veins were not treated as all were G1 according to NCI-CTCAE v.4.0. In phase Ib, first thromboembolic SAE was observed in October 2019 in a patient treated by 12 cycles of MitoTam. Till this time no other TE complications in patients treated by 4 cycles in Regimen 1 were recorded. Three of four patients with PE were responders, and PE was diagnosed after 16 (P1-1.0) and 12 cycles (P6-1.0 and P14-1.0) of MitoTam treatment in phase Ib. PE led to premature treatment termination in P1-1.0 and P6-1.0. Two more patients (P14-1.0 and P6-4.0) progressed at the time of PE diagnosis.

Five fatalities occurred during the trial, two of them meeting the criteria for SUSAR, and three with disease progression. First SUSAR (P2-0.5, 62-year-old female, phase I) was related to G4 thrombocytopenia due to idiopathic thrombocytopenic purpura and G3 enterorrhagia (SAE) in a patient with small bowel cancer (Table 3). This SUSAR regressed to normal state. Second SUSAR (P9-4.0, 69-year-old male, phase Ib) was clinically diagnosed with ischaemic stroke without CT correlation in a patient with metastatic CRC (Table 4). The symptoms were obvious within 24 h after MitoTam administration. This SUSAR was fatal (Table 4).

The median (range) time to maximum plasma concentration (T_max) of MitoTam was 0.5 h (range 0–1 h) in all regimens. In phase I, terminal half-life (t_1/2) of the drug was 7.33 h (0.25 mg/kg), ∼16 h (0.50–3.0 mg/kg), 44.46 h (4.0 mg/kg), 46.28 h (5.0 mg/kg), and 27.41 h (6.0 mg/kg). In phase Ib, t_1/2 was 6.19–12.46 h (Regimen 1), 6.73–9.49 h (Regimen 2) and 6.17–11.62 h (Regimen 3); accumulative index was 1–1.2666. PK evaluation after MitoTam administration did not show proportional C_max and AUC for the doses in phase I and Ib. There was a significant effect of the administration regimen on the elimination t_1/2 in plasma, AUC to the last quantifiable concentration, and C_max observed in serum, with p-values of 4.3 × 10⁻¹⁰, 0.011, and 2.6 × 10⁻⁵, respectively. The PK parameters indicate a two-compartment model, with the peak of MitoTam plasma concentration followed by its rapid decrease, suggesting fast elimination of MitoTam from the bloodstream and its tissue distribution. MitoTam thus likely accumulates in individual tissues and is subsequently eliminated from the body. This hypothesis is in agreement with distribution of MitoTam in pigs.⁷ In summary, analysis of the PK parameters showed that the extent of exposure was positively related to the dose. The overall effect of the administration regimen (all cohorts) was highly significant (p-value 4.3 × 10⁻¹⁰) with respect to the Wald χ²-test, indicating that t_1/2 was dose-dependent. In the regimen suggested for the prospective Phase II study (3.0 mg/kg once per week), C_max was reached within 0.25–1.00 h, t_1/2 was 6.73–9.49 h, mean residence time (MRT) and clearance (CL) were 7.43–10.74 h and 100.4–264.4 ml/h/kg respectively. Methods and detailed results of PK analysis will be published separately.

The secondary objective of the study, benefit from MitoTam treatment, gave the following results. For patients treated in phase Ib (diagnoses shown in Appendix Table S8, p 9), 37% (14/38) patients experienced CBR (evaluated according to RECIST 1.1 criteria) from the treatment (Fig. 2). In Regimen 1, 30% (6/20) patients benefited from MitoTam therapy; five patients reached SD and one PR. Of the responding patients two had RCC and underwent 12 and 16 cycles of MitoTam, with either SD and/or PR. In Regimen 2, 78% (7/9) patients responded (all reached SD) to MitoTam, of whom 3 were patients with RCC, two of whom underwent 11 and 12 cycles of MitoTam. In Regimen 3, 11% (1/9) patients responded to MitoTam and reached SD. In summary, thirteen patients had SD, one PR, and the remaining fourteen patients had PD as shown in Fig. 2.

Fig. 2.

Benefit for phase Ib patients from MitoTam treatment. Phase Ib patients were treated with MitoTam in Regimens 1, 2 and 3, and responded by progressive disease (PD), stable disease (SD) or partial response (PR). Marked patients did not continue MitoTam therapy due to: ∗patient's request, ^£non-compliance with inclusion criteria, ^§adverse event of MitoTam, ⁺protocol criteria (no possibility to continue therapy over 16 cycles in Regimen 1, over 12 cycles in Regimen 2 and over 6 cycles in Regimen 3, Regimen 3 due to timing of clinical trial), ⁺⁺COVID-19 restrictions.

To avoid the possible masking effect of diagnoses unevenly represented in each regimen,¹¹ we divided diagnoses of recruited patients into groups according to histogenetic origin, i.e., endodermal (EN), mesodermal (ME) and ectodermal (EC) tumours (Appendix Table S15, p 15), and significant CBR of MitoTam was observed in tumours of ME origin with RCC being in this subgroup (p = 0.018).

We conclude that the diagnosis most responsive to MitoTam is RCC with CBR of 83% (5/6) patients, who had been pre-treated with 3 or 4 lines of therapy, including tyrosine kinase inhibitors and immunotherapy (Table 5).

Table 5.

Clinical benefit rate (CBR) in patients with renal cell carcinoma (RCC) in phase Ib.

Subject	P1-1.0	P6-1.0	P20-1.0	P3-3.0	P6-3.0	P9-3.0
Gender	Male	Male	Male	Male	Male	Male
Age, years	63	71	62	72	73	67
Histological type of RCC	CCC	CCC	CCSF	CCC	CCC	CCC
Lines of palliative treatment	3xTKI	3xTKI 1xICI	3xTKI 1xICI	3x TKI	2xTKI 1xIFN	4xTKI
Doses per week	3	3	3	1	1	1
No of cycles/weeksa of MitoTam	16/50	12/36	4/11	12/19	6/10	11/20
Clinical Benefit Rate according to RECIST 1.1.	SD SD SD SD	SD PR PR	PD	SD SD	SD	SD SD

CCC = Clear cell carcinoma. CCSF = Claro-cellular carcinoma with sarcomatoid features. ICI = Immune Checkpoint Inhibitor. ID = subject Identification. IFN = Interferon alfa. P1 = patient one in dosage cohort 1.0 mg/kg. PD = progressive disease. PR = partial response. SD = stable disease. TKI = tyrosine kinase inhibitor.

^aNumber of weeks of MitoTam therapy counted from day one of cycle one to day 28 (follow up visit) after the last cycle.

Discussion

Mitochondria present an emerging target for cancer therapy, being key contributors to cellular energy balance and relay points for metabolic pathways,12, 13, 14 and generally differing between non-malignant and cancer cells.¹⁵ To date, only one mitochondria-targeting agent, the BH3 mimetic Venetoclax, has been approved for cancer therapy, viz. acute myeloid leukaemia.¹⁶ Clinical trial of the inhibitor of mitochondria-associated de novo pyrimidine synthesis, BAY240223423, was discontinued due to lack of benefit (https://clinicaltrials.gov/ct2/show/NCT03404726). Of IACS-010759¹⁷ and Gboxin¹⁸ that target oxidative phosphorylation, the former compound underwent phase I clinical trial with anti-cancer effect, albeit in a low number of patients.¹⁹ We prepared and tested mitochondrially targeted tamoxifen (MitoTam),⁵ based on a report showing that tamoxifen at suprapharmacological doses affects CI,²⁰ and launched phase I/Ib trial.

The primary objective of phase I trial was to assess toxicity of MitoTam. At 5.0 mg/kg, the frequency of AEs increased, although their severity was low. Therefore, after the first patient on 6.0 mg/kg who experienced grade 2 and 3 AEs, we decided to discontinue the trial. The dose of 5.0 mg/kg was therefore established as MTD. Phase Ib was conducted in three different regimens. We consider the weekly regimen as optimal from a pharmacological as well as practical viewpoint. Analysis of PK parameters showed that the extent of MitoTam exposure was positively related to the dose. As we will publish elsewhere, higher AUC and C_max in responders in Regimen 2 clearly demonstrate the dose-dependent link of MitoTam exposure to clinical efficacy. There is no obvious correlation between intensity or extent of exposure and increasing number of cycles. PK data and clinical efficacy are also indicative for the scheme of weekly administration of MitoTam rather than giving the agent in 3 doses per week. In Regimens 1 and 2, G1 AEs were most frequent, while in Regimen 3 frequent G1 and G2 AEs, and also G3 and G4 AEs occurred, for which reason some patients did not finish this treatment.

The most frequent systemic AE was reversible leukopenia/neutropenia, which occurred up to G3 severity, however, it did not require administration of antibiotics. Thrombocytopenia as well as anaemia were mostly mild and often presented already at baseline. There was one case of grade 4 thrombocytopenia recorded as SUSAR in phase I, probably linked to gastrointestinal bleeding. Anaemia was the only significant dose-dependent AE in phase Ib, while there was no significant dose-dependent AE in phase I. We can conclude that Regimens 1 and 2 are characterised by low and manageable haematological toxicity. Regimen 2, with 3.0 mg/kg of MitoTam, which we propose for phase II trial, showed almost exclusively G1 and occasional G2 toxicity. On the other hand, Regimen 3 (4.0 mg/kg) was accompanied by high level of haematological toxicity and we do not consider it for phase II. Nausea, vomiting, dyspnoea, ALT elevation, loss of appetite and fatigue may also be associated with MitoTam administration; the last two AEs seemed to be dose-dependent but these effects were not statistically significant. Hyperthermia/fever occurred repeatedly, most often one or two days after MitoTam application, responding well to antipyretics.

Pre-clinical accumulation of MitoTam in the kidney, myocardium, lungs and liver was explored.⁷ Although we recorded signs of nephrotoxicity (creatinine elevation), hepatotoxicity (ALT/AST elevation) and cardiotoxicity/pneumotoxicity (dyspnoea), none of these AEs was numerically or statistically significant, or serious. Potential relation of both cases of SUSAR linked to the treatment was mainly due to the time coincidence of the drug application and the event occurrence. Yet, one of them (ischaemic stroke) was fatal, and the TE complications require attention. Vein inflammation as an AE related to drug administration had been described in the pre-clinical model and confirmed in first two doses in phase I as a dose dependent AE. Although phlebitis of peripheral veins was entirely avoided by administration of MitoTam via central vein in phase I, in phase Ib asymptomatic PE in four of 37 patients occurred. Three patients were treated with MitoTam for longer time (12–16 cycles) than the rest of patients enrolled in phase Ib. PE was diagnosed at the end of 2019 in three patients, and shortly before the end of the trial in September 2020 in one patient. The time coincidence of PE and no other TE safety signals till the December 2019, when Regimen 1 of phase Ib trial was terminated, explain the absence of prophylactic indication of anti-coagulation during the study.

Risk factors of PE are presented by the type of diagnosis; for example, pancreatic cancer and RCC are malignancies with more frequent TE complications.²¹ Another possible risk factor is the type of central vein catheter. Three patients with RCC underwent 12, 12 and 16 cycles of treatment with MitoTam applied via PIC catheter. As the optimal duration of PIC catheter insertion is ∼3 months,²¹ it is possible that thrombosis and subsequently PE developed due to long-term insertion of PIC catheter (9, 7 and 11 months in these patients). Another risk factor may be the personal history of patients, two of whom experienced DVT or PE prior to MitoTam treatment. Thus, TE as a consequence of MitoTam therapy is unclear due to the presence of inherent disease risk factors, but cannot be ignored. Anti-coagulation therapy should be therefore considered in any other prospective study.

The initial efficacy analysis showed a positive relation of the dose-to-treatment response in Regimen 1 (1.0 mg/kg) and Regimen 2 (3.0 mg/kg), and paradoxical worsening in Regimen 3 (4.0 mg/kg). More detailed inspection of data showed that while the Regimen 2 subgroup comprised patients with neoplasms of the urogenital organ system (originating from ME), Regimen 3 subgroup consisted mainly of patients with gastrointestinal neoplasms (originating from EN).

Among patients with ME malignancies, five of six RCC patients showed clinical response to MitoTam. Five RCC patients had prognostically better clear cell histology and two patients had long history of RCC, which could have positive prognostic impact. On the other hand, all patients were pre-treated with tyrosine kinase inhibitors and three were pre-treated with immunotherapy, and progressed before MitoTam treatment. In one patient with PR in our series, immunotherapy with an immune checkpoint inhibitor (ICI) was the last treatment prior to MitoTam administration. Although the disease progressed during immunotherapy, SD and subsequently PR were observed after MitoTam, so that a synergistic effect of MitoTam and an ICI cannot be excluded. A potential combination of the ICIs PD-1 and PD-L1 with MitoTam is supported by our recent study showing their additive effect in a syngeneic mouse model of RCC, which also revealed 2- to 10-fold higher accumulation of MitoTam in renal tissue and its persistence in the kidney for almost a week longer compared to most other tissues.⁷

The study has certain limitations, especially in its ability to compare two different schemes of therapy (Regimen 1 vs. Regimens 2 and 3). Another limitation is the small number of patients due to different nature of the Regimens concerning the included diagnoses. Additionally, not all patients had the opportunity to repeat therapy in case of SD depending on the time of enrollment. The study enrollment was also negatively affected by the COVID-19 pandemic in case of several patients. Furthermore, patients were heavily pre-treated, which complicates the evaluation of haematological toxicities in particular.

In conclusion, the study met its objectives. MitoTam is generally well tolerated and its toxicity is manageable. Possible TE complications can be limited by MitoTam administration via vein port catheter and by anti-coagulation therapy. Of the included diagnoses, patients with RCC, all with metastatic disease and after 3–4 lines of therapy, showed greatest benefit from MitoTam. This is intriguing since RCC is a difficult-to-treat urological tumour with 20–40% recurrence after nephrectomy, and current therapies for RCC do not suffice.^22,23 The cohort in phase II study should include patients with RCC, and MitoTam could be combined with an ICI, which presents current standard of care.24, 25, 26

Contributors

The study was designed by Z.B., J.S., L.W. and J.N. based on data provided by L.D., K.R., S.S.H., R.Z., J.R., E.B., K.K., V.B., E.P., P.K., and J.N. Data were collected by Z.B., L.K., M.V., J.S., S.H., V.C., T.P., O.B., K.K., V.B., E.P., and P.K. Statistical analyses were performed by M.P. Full access to all data had Z.B., L.W., and L.P. and first three mentioned verified the underlying data. The manuscript was written by Z.B., L.P., L.W. and J.N. Final responsibility for the decision to submit for publication had Z.B., L.W., and J.N. but all authors contributed to the interpretation of the data, read and edited manuscript and accepted responsibility to submit it for publication.

Data sharing statement

Data collected for this study are not currently available. Sharing of such data for research purposes requires a written request placed with the sponsor of the study via Dr. Lukas Werner (Werner@smart-brain.cz), and, if deemed necessary by sponsor, internal data sharing and the transfer agreement vetted and signed by sponsor before any sharing of any data.

Declaration of interests

J.N., J.S and L.W. are owners of MitoTax s.r.o. that co-owns the MitoTam intellectual property.