Derivatives of melphalan designed to enhance drug accumulation in cancer cells

A rapid intracellular enrichment of alkylating payload is essential for melphalan flufenamide potency and mechanism of action

Despite decades of development of treatments and the successful application of targeted therapies for multiple myeloma, clinical challenges remain for patients with relapsed/refractory disease. A drug designed for efficient delivery of an alkylating payload into tumor cells that yields a favorable therapeutic window can be an attractive choice. Herein we describe melphalan flufenamide (melflufen), a drug with a peptide carrier component conjugated to an alkylating payload, and its cellular metabolism. We further underline the fundamental role of enzymatic hydrolysis in the rapid and robust accumulation of alkylating metabolites in cancer cells and their importance for downstream effects. The formed alkylating metabolites were shown to cause DNA damage, both on purified DNA and on chromatin in cells, with both nuclear and mitochondrial DNA affected in the latter. Furthermore, the rapid intracellular enrichment of alkylating metabolites is shown to be essential for the rapid kinetics of the downstream intracellular effects such as DNA damage signaling and induction of apoptosis. To evaluate the importance of enzymatic hydrolysis for melflufen's efficacy, all four stereoisomers of the compound were studied in a systematic approach and shown to have a different pattern of metabolism. In comparison with melflufen, stereoisomers lacking intracellular accumulation of alkylating payloads showed cytotoxic activity only at significantly higher concentration, slower DNA damage kinetics, and different mechanisms of action to reach cellular apoptosis.

A novel transport and delivery mechanism underpins the effectiveness of prolyl-m-sarcolysyl-p-fluorophenylalanine (PSF) in a human melanoma xenograft nude-mouse model

The alkylating peptide PSF shows very promising results in vitro on different cancer cells but its efficacy in animals has not been assessed. Here we evaluate the efficacy of PSF in human melanoma-bearing nude mice and examine the underlying mechanism. In melanoma-bearing nude mice, escalating doses of PSF showed dose-dependent responses and reached tumor regression with an optimal dose of 20 mg/kg for 1 month. A comparison of PSF with its free moiety m-sarcolysin and melphalan showed a highly significant advantage of PSF. Furthermore, dose fractionation yielded an even better control of tumor regrowth. In vitro studies unraveled an original delivery mechanism based on the rapid binding of PSF mainly due to red blood cells to form a pro-drug complex and the subsequent release of active metabolites by tumor-associated proteolytic enzymes. Blood kinetics showed one major metabolite partially released over time, while in the presence of melanoma cells three additional metabolites are generated. Interestingly, tumor-shed proteases also induce the production of these metabolites and varying combinations of enzyme inhibitors indicate the involvement of metallo- and other families of proteases in the delivery process. This particular transport and delivery of such an alkylating agent may have several benefits, mainly lowering the drug-free moiety in plasma and at the same time increasing its concentration in protease rich areas such as tumors.

D- and L-[123I]-2-I-phenylalanine show a long tumour retention compared with D- and L-[123I]-2-I-tyrosine in R1M rhabdomyosarcoma tumour-bearing Wag/Rij rats

The aim of this study was the comparison of the tumour uptake and the long-term retention of [(123)I]-2-I-L-phenylalanine and [(123)I]-2-I-D-phenylalanine with those of [(123)I]-2-I-L-tyrosine and [(123)I]-2-I-D-tyrosine in R1M rhabdomyosarcoma tumour-bearing rats. The biodistribution of the radioactivity as a function of time in R1M tumour-bearing rats was measured by planar gamma camera imaging (dynamic and static). If dissection was applied, the activity in the tumours and tissues of interest was measured by gamma counting. [(123)I]-2-iodo-L-phenylalanine, [(123)I]-2-iodo-D-phenylalaine, [(123)I]-2-I-L-tyrosine showed a considerable tumour uptake reaching a maximum between 10 and 30 min. At 30 min p.i. the differential uptake ratio values of this uptake were, respectively, 2.1, 2.3, 2.5 and 1.7. The activity in the tumour was shown to be related to a tumour cell uptake and not to an increased blood pool activity. All the tracers showed a clearance from the blood to the bladder without renal retention. At longer times both L- and D- [(123)I]-2-I-tyrosine were cleared for a large part from the tumours and the body. [(123)I]-2-I-L-Phe and [(123)I]-2-I-D-Phe showed a considerable and equal retention in the tumours: as compared with 0.5 h, 91% at 24 h and 80% at 48 h. This was related to the longer retention of activity in the blood pool noticed for these compounds (81% at 24 h and 65% at 48 h). The tumour-to-background ratio increased with 25% at those longer times. At short times all the tracers were taken up to a considerable extent in the tumours. In the R1M-bearing Wag/Rij rat model only [(123)I]-2-I-L-phenylalanine and [(123)I]-2-I-D-phenylalanine showed an especially high retention at long times without any significant difference between the enantiomers.

Intra-individual comparison of the human biodistribution and dosimetry of the D and L isomers of 2-[123I]iodo-phenylalanine

PURPOSE

Several authors have shown in animal studies that D-enantiomeric amino acid analogues can have better tumour imaging properties compared to their L-analogues. In our group, the D and L isomers of 2-[I]iodo-phenylalanine were identified as tumour-specific tracers in rat and mouse tumour models, with an advantage for the D-isomer. Here we compare, intra-individually, the normal biodistribution and dosimetry of both tracers in healthy human subjects.

METHODS

Five male volunteers received both the L- and D-enantiomers, ranging from 84 to 114 MBq, with a 1 week interval between the tracers, allowing intra-individual comparison. Whole-body scans were performed and blood and urine samples were collected and analysed up to 24 h. Dosimetry was calculated using OLINDA 1.0 software.

RESULTS

The biodistributions of the tracers are comparable as both show a moderate uptake in heart and the liver, a marked uptake in muscle tissue and clearance via the renal system. However, due to faster clearance, from 2.5 h, the uptake of the D-enantiomer was significantly lower compared to the L-isomer in all organs. The radiation dose estimations showed an effective dose of, respectively, 0.0120+/-0.0020 mSv x Bq(-1) and 0.0106+/-0.0038 mSv x Bq(-1) for 2-123I-L-Phe and 2-123I-D-Phe (P=0.18). In both cases the organ receiving the highest dose was the bladder wall.

CONCLUSION

Both 2-123I-L-Phe and 2-123I-D-Phe show comparable moderate uptake in all organs. 2-123I-D-Phe is the more promising tracer, as it shows a faster clearance resulting in a lower dose and a lower background, favouring tumour imaging with respect to the tumour/background ratio.

Comparison of the uptake of [123/125I]-2-iodo-d-tyrosine and [123/125I]-2-iodo-l-tyrosine in R1M rhabdomyosarcoma cells in vitro and in R1M tumor-bearing Wag/Rij rats in vivo

INTRODUCTION

Recently, promising results concerning uptake in vivo in tumors of D-amino acids have been published. Therefore, we decided to evaluate the tumor uptake of the D-analogue of [(123)I]-2-iodo-L-tyrosine, a tracer recently introduced by our group into clinical trials. The uptake of 2-amino-3-(4-hydroxy-2-[(123/125)I]iodophenyl)-D-propanoic acid (2-iodo-D-tyrosine) was studied in vitro in LAT1-expressing R1M rat rhabdomyosarcoma cells and in vivo in R1M tumor-bearing Wag/Rij rats.

METHODS

The uptake of [(125)I]-2-iodo-L-tyrosine and [(125)I]-2-iodo-D-tyrosine into R1M cells was determined in appropriate buffers, allowing the study of the involved transport systems. In vivo, the biodistribution in R1M-bearing rats of [(123)I]-2-iodo-L-tyrosine and [(123)I]-2-iodo-D-tyrosine was performed by both dynamic and static planar imaging with a gamma camera.

RESULTS

In in vitro conditions, the uptake of both [(125)I]-2-iodo-L-tyrosine and [(125)I]-2-iodo-D-tyrosine in the HEPES buffer was 25% higher in the presence of Na(+) ions. In the absence of Na(+) ions, [(125)I]-2-iodo-D-tyrosine was taken up reversibly in the R1M cells, with an apparent accumulation, probably for the larger part by the LAT1 system. Dynamic planar imaging showed that the uptake in the tumors of [(123)I]-2-iodo-D-tyrosine was somewhat lower than that of [(123)I]-2-iodo-L-tyrosine. At 30 min postinjection, the mean differential uptake ratio values of the L- and D-enantiomers are 2.5+/-0.7 and 1.7+/-0.6, respectively. Although the uptake of the D-isomer is lower, probably due to a faster clearance from the blood, the tumor-background ratio is the same as that of the l-analogue.

CONCLUSION

A large part (75%) of [(125)I]-2-iodo-D-tyrosine in vitro and [(123)I]-2-iodo-D-tyrosine in vivo is reversibly highly taken up in R1M tumor cells by Na(+)-independent LAT transport systems, more likely by the LAT1. The clearance from the blood of [(123)I]-2-iodo-D-tyrosine in the rats is faster than that of the L-analogue, resulting in a slightly lower tumor uptake but with the same tumor-background ratio.

Activity of hydrolytic enzymes in tumour cells is a determinant for anti-tumour efficacy of the melphalan containing prodrug J1

Recently, we presented a series of melphalan containing di- and tripeptides with high cytotoxic activity and J1 (l-melphalanyl-p-l-fluorophenylalanine ethyl ester) was identified as one of the most interesting compounds. It was speculated that the increased activity compared to melphalan itself, demonstrated both in vitro and in vivo, resided in increased transport over the tumour cell membrane and/or hydrolytic cleavage and liberation of melphalan inside the cells. Indeed, overexpression of hydrolytic enzymes like peptidases, esterases and proteases has been described in several types of human malignancies, thus providing a target for selective chemotherapy. In this work, the details of the increased activity was further investigated and potential tumour selectivity is discussed. The intracellular delivery of melphalan is investigated in detail using peptidase resistant dipeptide derivatives, by enzyme inhibitors and probes for enzymatic activity and by studying the time dependency of drug effect as well as intracellular drug concentrations (cellular pharmacokinetics). The results show that the activity of the dipeptide mustards is highly dependent on intracellular hydrolysis, which result in rapid intracellular release of the alkylating unit (i.e. free melphalan) in cells with high enzymatic activity. The maximum intracellular melphalan concentration following J1 exposure was reached already after 15 min, thereafter declining with a half-life of approximately 1 h. This rapid intracellular loading resulted in less reduction of activity for J1 than for melphalan and six other standard drugs when human tumour cell lines were exposed to the drugs for a limited time (simulating short half-life in vivo). Peptidase inhibitors inhibited the activity and intracellular release of melphalan, and dipeptide derivatives designed to resist the action of peptidases was less active than the corresponding normal dipeptide.

Antitumor activity of the alkylating oligopeptides J1 (L-melphalanyl-p-L-fluorophenylalanine ethyl ester) and P2 (L-prolyl-m-L-sarcolysyl-p-L-fluorophenylalanine ethyl ester): comparison with melphalan

Peptichemio, a mixture of six short oligopeptides all comprising the alkylating amino acid m-L-sarcolysin, has shown clinical activity in several malignancies. Previous studies have suggested that activity mainly resides in one of the peptides, P2 (L-prolyl-m-L-sarcolysyl-p-L-fluorophenylalanine ethyl ester). In the present study the in vitro activity of P2 was further investigated and compared to melphalan and the novel alkylating dipeptide J1 (L-melphalanyl-p-L-fluorophenylalanine ethyl ester), which is structurally related to P2 and melphalan. Cytotoxic activity was studied using patient tumor cells in a non-clonogenic cytotoxicity assay, whereas cellular response, and kinetics thereof, were studied in the lymphoma cell line U-937 GTB. Cellular metabolism was studied using microphysiometry, kinetic effects on macromolecular synthesis by radiolabeled substrate incorporation and, finally, the microculture kinetic assay of apoptosis was used to monitor morphologic changes following drug exposure. The assays compared P2 favorably with melphalan. Interestingly J1 was even more cytotoxic, and produced more pronounced effects in the kinetic assays for macromolecular synthesis, metabolic activity and apoptosis. The results indicate that the delivery properties of J1 are improved compared to those of melphalan and P2.

Esters of 2-(1-hydroxyalkyl)-1,4-dihydroxy-9,10-anthraquinones with melphalan as multifunctional anticancer agents

Eight esters of 2-(1-hydroxyalkyl)-1,4-dihydroxy-9,10-anthraquinone with melphalan were prepared and tested for their antitumor activity (S-180) and cytotoxicity. 2-[1-[4-(p-Bis(2-chloroethyl)-aminophenyl)-butanoyloxy]methyl]-1,4-dihydroxy-9,10-anthraquinone and 2-[1-[4-(p-bis(2-chloroethyl)-aminophenyl)-butanoyloxy]ethyl]-1,4-dihydroxy-9,10-anthraquinone showed remarkable antitumor activity (T/C, 265 and 272%).

Targeted drug delivery systems 6: Intracellular bioreductive activation, uptake and transport of an anticancer drug delivery system across intestinal Caco-2 cell monolayers

We demonstrate transport across, intracellular accumulation and bioreductive activation of a conformationally constrained, anticancer drug delivery system (the CH(3)-TDDS) using Caco-2 cell monolayers (CCMs) as an in vitro model of the human intestinal mucosa. Reverse-phase High Performance Liquid Chromatography (HPLC) coupled with UV detection was used to detect CH(3)-TDDS, the bioreduction product (lactone) and the released drug (melphalan methyl ester; MME). Upon incubation of the CH(3)-TDDS with the apical (AP) surface of 21-day-old CCM, we observed rapid decrease in the AP concentration of the CH(3)-TDDS (60%/hr) as a result of cellular uptake. Rapid intracellular accumulation of the CH(3)-TDDS was followed by bioreductive activation to deplete the cellular levels of CH(3)-TDDS. The drug part (MME) and lactone, as well as CH(3)-TDDS, were detected in the basolateral (BL) chamber. Intracellular Caco-2 levels of TDDS and lactone were also detectable. Bioreductive activation of the CH(3)-TDDS was additionally confirmed by formation of lactone after incubation of the CH(3)-TDDS in the presence of freshly prepared Caco-2 cell homogenates. During transport studies of melphalan or MME alone (as control), the intact drug was not detected in the intracellular compartment or in the BL chamber. These observations demonstrate that CH(3)-TDDS has potential for improving intestinal delivery of MME. TDDS could be useful in facilitating oral absorption of MME as well as the oral delivery of other agents.