Commercial development considerations for biotechnology-derived therapeutics.

Although it seems unlikely, it has only been 20 years since the US Food and Drug Administration (FDA) approved the first recombinant protein as a therapeutic modality. Unbelievably, an average of slightly more than two approvals per year of monoclonal antibodies (MAbs) and other human protein therapeutics has been achieved by this burgeoning industry 43 recombinant protein therapeutics in the two decades since 1982 (see Table 1), and the pace is increasing.

The effect of interleukin-2 administration on wound healing in adriamycin-treated rats.

Adriamycin (doxorubicin hydrochloride), an effective chemotherapeutic drug, is also a potent inhibitor of wound healing. Conversely, certain polypeptide growth factors are capable of stimulating fibroblasts to secrete collagen, thus enhancing wound healing. The purpose of this study was to determine if interleukin-2 (IL-2), a T-cell growth factor, could reverse the wound healing deficit caused by Adriamycin. Adriamycin treatment caused a significant decrease in wound-breaking strength (P less than 0.005). IL-2 administration increased wound-breaking strength in Adriamycin-treated animals (2126 g vs 1549 g, P less than 0.005). In control animals, IL-2 did not increase wound-breaking strength significantly (2708 g vs 2608 g, P greater than 0.1). Histologically, wounds from Adriamycin-treated animals were less cellular, demonstrated less collagen in the dermis, and a lesser degree of capillary ingrowth. The number of fibroblasts in the dermal layer was increased in animals receiving IL-2. Control rats gained an average of 1.4% of their original body weight, while Adriamycin-treated rats lost an average of 19% of their original body weight (P less than 0.0005). IL-2 administration did not influence weight loss or gain. Hematologically, animals receiving Adriamycin had lower hemoglobin and hematocrit values and higher platelet counts. There were no differences in total white blood cell counts; however, animals receiving Adriamycin showed a predominance of polymorphonuclear leukocytes and a relative decrease in lymphocytes. Animals receiving IL-2 demonstrated a significant eosinophilia. (1) Adriamycin impairs normal wound healing. (2) Interleukin-2 administration improves the wound healing impairment caused by Adriamycin. (3) Interleukin-2 appears to increase infiltration of inflammatory cells, fibroblasts, and capillaries into the wound, which may account for the observed increase in wound breaking strength.

The role of oxidant injury in tumor cell sensitivity to recombinant human tumor necrosis factor in vivo. Implications for mechanisms of action.

The intracellular glutathione levels of two human tumor lines and seven murine tumor lines were determined in order to investigate the role of oxidant injury in tumor cell sensitivity to human rTNF (rhTNF). Correlations were found between high intracellular glutathione levels and in vivo tumor resistance to rhTNF, and on the other hand, low glutathione levels and rhTNF sensitivity. The transplantable murine fibrosarcoma, Meth A, a TNF-sensitive line in vivo, was less sensitive to rhTNF and host toxicity was reduced when the hosts were pretreated with uric acid, a major reactive oxygen scavenger in humans and certain other primates. Conversely, pretreatment of the tumor-bearing hosts with DL-buthionine-(S,R)-sulfoximine, an inhibitor of GSH biosynthesis, resulted in an increased sensitivity of Meth A to rhTNF. This effect was not limited to tumor-bearing mice, as rats pretreated with diethyl maleate, a compound which irreversibly binds glutathione, were more sensitive to rhTNF toxicity than control rats. On the other hand, pretreatment with N-acetyl cysteine, an oxidant scavenger, reduced the toxicity of rhTNF treatment in rats. The data are consistent with the hypothesis that tumor cell sensitivity to rhTNF in vivo is dependent on its capacity to buffer oxidative attack. In addition, host toxicity is also related to the production of reactive oxygen species. Activated effector cells such as granulocytes and macrophages are hypothesized to produce most of this damage by their respiratory burst and oxidant release, although the direct action of rhTNF may also contribute to oxidative injury in vivo.

Effect of human gamma interferon on yellow fever virus infection.

We studied yellow fever virus infection in two species of monkey: Saimiri sciureus (squirrel monkeys) and Macaca mulatta (rhesus monkeys). Human gamma interferon was administered intravenously in five equal doses, one was given 24 hr before infection followed by four doses 24 hr apart. Interferon reduced the levels and duration of viremia and the severity of hepatitis in squirrel monkeys. Interferon prolonged survival time and delayed the appearance of viremia and hepatitis in infected rhesus monkeys, but it did not change overall mortality.

Characterization of the antitumor activities of human tumor necrosis factor-alpha and the comparison with other cytokines: induction of tumor-specific immunity.

We have investigated the in vitro and in vivo antitumor activities of recombinant human tumor necrosis factor-alpha (rHuTNF-alpha) against Meth A sarcoma. Meth A sarcoma cells were found to a) be relatively insensitive in vitro to rHuTNF-alpha, and b) express low numbers of TNF-alpha receptors. Intraperitoneally implanted Meth A sarcoma was insensitive to the antitumor effects of rHuTNF-alpha. In contrast, rHuTNF-alpha was highly efficacious against subcutaneously implanted Meth A sarcoma. Biodistribution studies with 125I- or 3H-labeled rHuTNF-alpha demonstrated that, after intravenous administration, the majority of the labeled rHuTNF-alpha localized in the kidney, lungs, and liver. Only low levels of radiolabel were found in subcutaneous Meth A implants. These results support the in vitro data on the low number of TNF-alpha receptors on Meth A sarcoma cells. The ability of rHuTNF-alpha to induce regression of established (7 days) subcutaneous Meth A implants, positively correlated with the degree of both macroscopic and microscopic tumor necrosis. In addition, recombinant human tumor necrosis factor-beta (lymphotoxin) and recombinant murine tumor necrosis factor-alpha induced similar levels of necrosis. Other lymphokines with known antitumor activities, recombinant human interferon-gamma, murine interferon-gamma, and human interleukin 1 alpha, failed to induce detectable necrosis of Meth A sarcoma. Mice which had rejected subcutaneous Meth A sarcoma implants after rHuTNF-alpha treatment and which were later challenged subcutaneously with Meth A sarcoma or other noncross-reacting chemically induced sarcomas were found to be specifically immune to Meth A sarcoma. In addition, low levels of cytotoxic antibodies reactive to Meth A sarcoma were detected in the sera of 21 of 30 Meth A immune mice. Histological evaluation of the hemorrhagic tumor necrosis induced by rHuTNF-alpha suggests that the primary lesion is vascular, possibly directly on the endothelial cells. The mechanisms involved in the generation of specific cell-mediated antitumor immunity in this model are at present unknown.

Interferon reduces hepatic drug metabolism in vivo in mice.

The effects of two highly purified human leukocyte interferons (IFN-A and IFN-AD) on drug-metabolizing capacity in mice have been investigated. IFN-AD was found to produce significant changes in antipyrine half-life, assessed by analysis of 14CO2 exhalation rates following 14C-antipyrine administration. By contrast, IFN-A, which has considerably less antiviral potency than IFN-AD, was found to have no effect on antipyrine half-life. The administration regimen was found to markedly alter the effects seen with IFN-AD. When IFN-AD was given as single daily doses (5 X 10(7) units/kg/day X 3 days), the half-life of antipyrine increased by a mean of 40% (from 21.0 to 28.9 min). However, when a smaller daily dose (3 X 10(7) units/kg/day) was given as a continuous infusion, the antipyrine half-life increased by more than 3-fold (from 20.8 to 68.5 min) after 3 days of administration. Continued infusion for a further 3 days produced no additional change in antipyrine half-life. These results demonstrate that human leukocyte interferons can significantly inhibit hepatic metabolic activity in vivo.

In vivo antiviral activity of recombinant murine gamma interferon.

Recombinant murine gamma interferon (gamma-IFN) was tested for its antiviral activity in vivo. IFN preparations purified to greater than 95% purity were administered to CD-1 mice infected with lethal doses of encephalomyocarditis (EMC) virus. An initial treatment with rMuIFN-gamma administered 4 h prior to infection with virus, followed by daily treatment for 3 consecutive days significantly protected mice against EMC virus as evidenced by animal survival after 3-4 weeks post-viral infection. Variations in the antiviral effect relative to dose levels and routes of administration were also studied.

In vitro fibrinolytic activity of recombinant tissue-type plasminogen activator in the plasma of various primate species.

The ratio of fibrinolytic to fibrinogenolytic effect of recombinant human tissue-type plasminogen activator was investigated at several concentrations in an in vitro system consisting of 125I-fibrin labeled autologous plasma clots immersed in the plasma of several different primate species. A concentration-dependent fibrinolysis was obtained in each case; however, the degree of fibrinolysis differed markedly from one primate species to the other. At a concentration of 30 IU/ml (300 ng/ml) of recombinant tissue-type plasminogen activator nearly complete thrombolysis was achieved within 4 hr in human plasma, in the chimpanzee, in the cynomolgus macaque, in the bonnet macaque and in the rhesus macaque. The common baboon, the olive baboon and the squirrel monkey were rather resistant to thrombolysis. In all species the thrombolysis achieved was not associated with fibrinogenolysis nor with significant decreases in plasminogen or alpha 2-antiplasmin.

Biological properties of human tissue-type plasminogen activator obtained by expression of recombinant DNA in mammalian cells.

Human tissue-type plasminogen activator (t-PA), obtained by expression in mammalian cells of recombinant DNA coding for the entire sequence of t-PA (rt-PA), was compared with natural activator from melanoma cell culture (mt-PA). In an in vitro system, composed of [125I]fibrinogen-labeled plasma clot suspended in circulating human plasma, rt-PA and mt-PA caused a very similar dose-related degree of fibrinolysis without causing extensive fibrinolytic activation and fibrinogen breakdown in the surrounding plasma. Urokinase only induced fibrinolysis at a 5- to 10-fold higher concentration and in association with extensive fibrinogenolysis. Intravenous injection of mixtures of labeled (0.4 microCi/kg) and unlabeled (2000 I.U./kg) mt-PA or rt-PA resulted in a rapid but similar disappearance of activity from plasma (T1/2 of 3 min) and specific accumulation of tracer in the liver. In rabbits with experimental jugular vein thrombosis, rt-PA and mt-PA caused a very similar dose-dependent thrombolysis without causing substantial systemic activation of the fibrinolytic system and fibrinogenolysis. Urokinase induced significant thrombolysis only at a 10-fold higher dose and this was associated with systemic fibrinolytic activation. Infusion of 96,000 I.U./kg (approximately equal to 1 mg/kg) of mt-PA or rt-PA over 4 hr induced approximately 70% lysis, whereas a 10-fold higher dose of urokinase yielded 35 to 40% lysis. Two subfractions of rt-PA differing in the extent of glycosylation had very similar thrombolytic properties. It is concluded that the potentially more readily available rt-PA could constitute a specific, fibrin-selective thrombolytic agent.

Effects of verapamil on doxorubicin-induced mortality and electrolyte changes in the mouse heart.

The effects of verapamil on doxorubicin-induced mortality and electrolyte changes were investigated in mice using either 0, 8, 16, or 32 mg/kg verapamil. The verapamil doses were injected by subcutaneous route 24 and 3 hr prior to administration of a single dose of doxorubicin (20 mg/kg, IP), followed by daily treatment with verapamil for 20 subsequent days. Using total deaths or changes in ventricular electrolyte levels as indices of toxicity, there was no evidence that verapamil decreased doxorubicin toxicity. At this dose, doxorubicin increased myocardial calcium and sodium, and decreased magnesium and potassium. Verapamil alone was shown to significantly reduce total ventricular calcium and increase magnesium in a dose-dependent manner. The ventricular sodium and potassium levels were not affected by verapamil treatment. Neither the increase in calcium nor the decrease in heart magnesium content of the ventricles in response to doxorubicin, was lessened by verapamil treatment. Doxorubicin-associated sodium increase in ventricular tissue was greater in groups receiving higher doses of verapamil but decrease of potassium in the ventricles of these animals remained unaffected. We conclude from the findings of the present study that verapamil is inappropriate as an antagonist of the cardiac toxicity resulting from treatment with doxorubicin.

Effects of feed-pairing and different doses of doxorubicin on mortality and electrolyte changes in the mouse heart.

Various electrolyte changes have been documented in the hearts of different species of laboratory animals intoxicated with doxorubicin. However, it has not been determined what role the drug-induced relative nutritional lack plays in the electrolyte imbalance of this antineoplastic drug, nor to what extent the electrolyte imbalance is dose-related. A feed-pairing study performed after a single 20 mg/kg i.p. dose of doxorubicin demonstrated that relative heart weight was significantly increased in responding drug treated mice over ad libitum controls. The heart weights of feed-paired animals were also elevated to the same extent. Nutrition did not appear to have any effect on the drug-induced elevation of total ventricular calcium levels in responding animals. Although heart sodium concentrations were increased over ad libitum controls by feed restriction, they were still significantly lower than the doxorubicin-treated animals. Cardiac levels of magnesium and potassium were significantly decreased over both ad libitum and feed-paired controls. Calcium and sodium levels were increased significantly at 32 mg/kg but not at 50 or 80 mg/kg doxorubicin, while potassium and magnesium were depressed at these doses. Both the median time of death and total weight loss were inversely related to dose. It is concluded that the electrolyte disturbances are not due to the nutrition-related weight loss, and may be a proximal cause of doxorubicin-induced cardiotoxicity.

Effects of nephrectomy on the pharmacokinetics of various cloned human interferons in the rat.

The pharmacokinetics of two human leukocyte interferon (IFN) subtypes (IFN-alpha A and IFN-alpha D) and two molecular hybrids [IFN-alpha AD(Bgl) and IFN-alpha AD(Pvu)] was studied in rats receiving single i.v. injections, 5.0 X 10(6) I.U./kg, of these materials. All four preparations were highly purified and resulted in plasma decay curves with similar elimination half-lives of 95.3 +/- 13.5 min. The four IFN-alpha subtypes investigated showed differences when their clearances are compared, suggesting that the 13 residue differences between amino acids 92 and 165 affect the clearance. The fact that only IFN-alpha D has an apparent volume of distribution (Vd beta) of 12.9 +/- 3.4 liters/kg whereas the other three IFN-alpha subtypes have Vd beta of 5.7 +/- 1.0 liters/kg suggests that the amino acid differences between residues 1 and 61 affect the Vd beta. Only in the case of IFN-alpha D did functional nephrectomy significantly (P = .02) increase the apparent volume of distribution at steady state. Functional nephrectomy before i.v. injection resulted in a 4-fold decrease in IFN-alpha AD(Bgl) and IFN-alpha AD(Pvu) clearance, an 8-fold decrease in IFN-alpha A clearance and a 10-fold decrease in IFN-alpha D clearance. The elimination half-life of each interferon was unchanged after nephrectomy. Intramuscular administration of the IFN-alpha subtypes in rats produced low plasma concentrations. IFN-alpha D peaks at 39 +/- 10 I.U./ml of plasma, whereas IFN-alpha A reaches 201 +/- 53 I.U./ml, IFN-alpha AD(Bgl) reaches 396 +/- 47 I.U./ml and IFN-alpha AD(Pvu) results in a maximum plasma concentration of 747 +/- 66 I.U./ml.(ABSTRACT TRUNCATED AT 250 WORDS)

Influence of various purified interferons on effects of drugs in mice.

Direct determinations indicate that some types of interferon increase the duration of hexobarbital-induced sleep time and decrease acetaminophen toxicity. Pretreatment of mice with cloned human leukocyte interferon subtype A, IFN-alpha A, did not increase hexobarbital sleep time but pretreatment with mouse interferon or a hybrid human leukocyte interferon, IFN-alpha AD (Bgl), increased the duration of the effect of the barbiturate. Hybrid IFN-alpha AD (Bgl) also decreased the lethality of acetaminophen. These findings are predictable based on reports of hepatic cytochrome P450 depression resulting from treatment with some interferons.

Pharmacokinetics, covalent binding and subcellular distribution of [3H]doxorubicin after intravenous administration in the mouse.

Some aspects of the in vivo disposition of [3H]doxorubicin were investigated in mice in the first 6 hr after a 5 mg/kg i.v. dose. The disappearance of radioactivity from plasma and erythrocytes was consistent with biphasic kinetics. Highest peak radioactivity was found in lung, liver and kidney. Retention half-lives of radioactivity in heart, lung, kidney and small intestine were between 4.5 to 6.5 hr, while in liver, adrenal gland and spleen they were 7 to 13 hr and in skeletal muscle and bone marrow were 17 and 35 hr, respectively. Radioactivity in pancreas and epididymal fat had not equilibrated by 6 hr, and brain and testicle contained little radioactivity. Exhaustive methanol-ether extraction of acid-precipitated homogenates showed covalent binding of radioactivity to lung, liver, kidney and spleen at 40 to 100 pmol/mg of protein, while binding in plasma, small intestine and pancreas ranged from 4 to 30 pmol/mg of protein; heart and skeletal muscle showed 0 to 1.4 pmol/mg of protein at all times examined. Differential centrifugation revealed no site of subcellular accumulation within liver, kidney or most notably heart, other than the nuclear fraction. Specific levels of drug were highest in nuclei of liver, requiring 45 min to reach peak amounts; the lowest quantity per milligram of nuclear protein was found in the heart, and had already peaked by 10 min. These results suggest that the cardiotoxicity of doxorubicin may not be due to concentration or retention of drug within heart tissue as a whole or any non-nuclear subcellular fraction of it nor to the covalent binding of drug to heart protein, but may follow directly from the interaction of doxorubicin with the cardiac nuclei.