Potential break through technology poised
Potential break-through technology poised to overcome these above-mentioned limitations is that of the “substrate mediated enzyme prodrug therapy”, SMEPT (Fig. 1). Learning from the previously established enzyme prodrug therapies (EPT) and specifically the antibody-directed EPT (ADEPT) , development of SMEPT was carried out over the past decade such as to engineer robust and flexible instruments into the design of implantable biomaterials and achieve localized drug synthesis rather than drug delivery , . Envisioned advantages of SMEPT over standard drug-eluting implants are similar to those of ADEPT over ADC (Fig. 2). Most importantly, EPT makes it easy to achieve the synthesis, and therefore the delivery, of combination of drugs, concurrently or in sequence, at the individually nominated doses and times of administration. This is a highly desired opportunity which remains a challenge specifically in the design of conventional drug eluting matrices. Further, EPT offers a drastically increased deliverable payload, achieved through numerous cycles of catalysis performed by each enzyme. Finally, EPT is well suited for the synthesis of short-lived drug molecules for which drug delivery is challenging, such as nitric oxide. In this review, we aim to present the historical developments of techniques that led to the establishment of SMEPT and the state of the art of this methodology. We also briefly discuss the envisioned avenues for subsequent development of EPT as engineered into implantable biomaterials, including the discussion over translational potential of SMEPT.
Early developments Historically, the advent of SMEPT can be traced to the 1980′s, where the first example of localized prodrug conversion was achieved using a judiciously placed enzyme – in order to achieve a localized feed of 5-fluorouracil (5-FU). Immobilization of E. coli-derived cytosine deaminase (CDase) was employed for enzymatic conversion of the non-toxic 5-fluorocytosine (5-FC) into a highly cytotoxic chemotherapeutic agent 5-FU (Fig. 3, A), in an attempt to achieve a local antineoplastic activity , . At the time, 5-FC was employed for anti-fungal treatments due to its selective toxicity towards CDase-expressing fungi. Understanding of the CDase-mediated prodrug activation inspired the development of an implantable biomedical device to contain immobilized CDase for local conversion of 5-FC into 5-FU for applications in anti-cancer therapy. Initially, CDase was injected intratumoraly (s.c. Amikacin synthesis tumor model in rat) followed by intraperitoneal (i.p.) administration of 150mg/kg 5-FC which resulted in a local conversion of 5-FC into 5-FU and increased the survival outcome , . These findings resulted in a patent on the use of 5-FC and CDase for antitumor therapies in 1982 . In 1985, Sakai et al.  and Nishiyama et al.  immobilized CDase in capsules comprised of cellulose dialysis tubing (Fig. 3, B) and implanted these capsules in the s.c. xenograft tumor bed, in an attempt to obtain local release of 5-FU after 5-FC administration (Fig. 3, C). Sakai et al.  investigated stability of the encapsulated CDase after long-term incubation at 37°C (in vitro) and after 30days s.c. implantation of the capsules (in vivo) revealing half-lives of 10.3days and 10±2days, respectively. For the latter, the capsules were covered by thick fibrous connective tissue after 30days implantation. Sakai et al. conclude this work by proposing the use of the capsules as a post-operative treatment of remnant brain tumor tissue that cannot be surgically resected . Next, Nishiyama et al.  investigated the therapeutic potential of these CDase-containing capsules using a murine brain tumor model (s.c. EA-285 glioma xenograft in rats). Bioavailability study for 5-FC following i.p. administration of 150mg/kg 5-FC (i.e. ~15mg 5-FC per rat weighing ~100g) showed that 98μg/mL 5-FC was detected in the serum 1h post injection and a comparable concentration (88μg/mL) of 5-FC was detected in the tumor tissue. Subsequently, the 5-FC concentration gradually decreased, giving a residence half-life of 4.5h for 5-FC in the tumor (Fig. 3, D). This work clearly indicates that the 5-FC prodrug effectively migrates from the peritoneum to the tumor tissue. 5-FU was detected locally in the tumor as early as 30min after prodrug administration, peaking at 2h with a concentration of 8μg/g. The in-tumor drug concentration was maintained at 1μg/g concentration or higher between 1 and 6h post-administration of 5-FC, whilst the half-life of 5-FU was 3.2h. The local concentration of 5-FU within the tumor was markedly higher than in serum (Fig. 3, E) which clearly illustrates the advantage of EPT as a strategy for localized drug delivery. After 30days, the tumor volume in CDase-capsule treated rats was only 19% (Fig. 3, F). Interestingly, at least 25% of the original enzymatic activity was still observed even after 3month of implantation of the enzyme indicating that therapy can be extended well over the 30days' time frame used in the reported study. Subsequent work  focused on improving stability of the encapsulated enzyme and afforded enzyme preparations with in vitro half-life of 28days for the initial 4months and 99days for the subsequent 5months.