Status: withdrawal from world markets announced in late 2011

Organizations involved:

BioCritica – USA mark.

Lilly, Eli & Co. – R&D; Tech.; Intl. mark.

Lonza Biologics plc – Manuf.; Tech.

ARIAD Pharmaceuticals, Inc. – Patent dispute

University of Glasgow – Tech.

National Institutes of Health (NIH) – Patent dispute

Oklahoma Medical Research Foundation – Patent dispute

University of Washington – Patent dispute

Description: Xigris is a lyophilized (freeze-dried) formulation of drotrecogin alfa (activated), recombinant human activated Protein C, a serine protease enzyme with thrombolytic activity. Drotrecogin alfa is expressed by Human Embryonic Kidney 293 (HeK293) cell line transformed with the cDNA for the human Protein C zymogen (inactive precursor form of the enzyme), which is then enzymatically activated by cleavage/removal of a portion of molecule using bovine thrombin (bovine source). The resulting drotrecogin alfa (activated) is substantially identical to human plasma-derived activated Protein C, including having the same amino acid sequence. Drotrecogin alfa (activated) and human activated Protein C are dimeric glycoproteins consisting of a heavy chain and a light chain linked by a disulfide bond, with 10 other intrachain disulfide bonds (7 light chain, 3 heavy chain), four glycosylation sites (1 light chain, 3 heavy chain), a molecular weight of ~55 kDa, and molecular formula of C2071H3165N581O640S31.

In vivo, both drotrecogin alfa (activated) and endogenous activated Protein C (in conjunction with Protein S, a cofactor) hydrolyze coagulation Factors Va and VIIIa. This inhibits the formation of thrombin and prothrombin, resulting in antithrombotic and anticoagulatory effects (inhibition of blood clotting and coagulation). The specific mechanisms by which drotrecogin alfa (activated) exerts an effect on survival in patients with severe sepsis have not been determined.

Xigris is supplied in 5 and 20 mg preservative-free vials containing 5.3 mg or 20.8 mg of drotrecogin alfa (activated), respectively; plus 40.3 or 158.1 mg of sodium chloride, 10.9 or 42.9 mg of sodium citrate, and 31.8 or 124.9 mg of sucrose, respectively. Xigris should be stored at 2-8°C (refrigerated). The dating period is 18 months from the date of manufacture, the date of final sterile filtration of the formulated product.

Nomenclature: Protein C, activated, rDNA [BIO]; Xigris [TR]; drotrecogin alfa (activated) [FDA]; drotrecogin alfa [INN]; Zovant [TR former]; Factor XIV [CAS]; 98530-76-8 [CAS RN]; APC [SY]; activated Protein C [SY]; aPC [SY]; DAA[SY]; NDC 0002-7559-01 [NDC]

Note, activated protein C and other terms are commonly be used to refer to recombinant and/or endogenous forms.

Biological.: Endogenous protein C circulates in the blood as a zymogen (inactive form of an enzyme). Thrombin rapidly activates Protein C in vivo after it binds to thrombomodulin, a membrane receptor protein on endothelial cells. When thrombin is bound to thrombomodulin, it undergoes conformational changes, losing its procoagulant activities and accelerating its ability to activate protein C zymogen as much as 20.000 fold. This provides a natural mechanism for control of the coagulation system and prevention of excessive clotting (as occurs in persons with congenital Protein C deficiency). See the Factor VIII Products entry in the Blood Products, Human section for further discussion of blood coagulation mechanisms.

Activated Protein C (aPC; both endogenous and recombinant) exerts an antithrombotic effect by inhibiting endogenous Factors Va and VIIIa. APC also degrades the non-activated forms of Factor V and Factor VIII, although at a slower rate. In vitro data indicate that aPC has indirect profibrinolytic activity through inhibition of plasminogen activator inhibitor-1 (PAI-1) and limiting generation of activated thrombin-activatable-fibrinolysis-inhibitor. In vitro data also indicate that aPC exerts an anti-inflammatory effect by inhibiting human tumor necrosis factor (TNF) production by monocytes, by blocking leukocyte adhesion to selectins, and limiting thrombin-induced inflammatory responses within the microvascular endothelium. Drotrecogin alfa (activated) and endogenous activated Protein C are inactivated by endogenous plasma protease inhibitors. Plasma concentrations of endogenous activated Protein C in both healthy subjects and patients with severe sepsis are usually below detection limits. The concentration of APC in normal plasma is is reported to be about 2 ng/ mL, and its in vivo half-life is about 15 minutes.

Endogenous and recombinant Protein C, in the zymogen and activated forms, are substantially identical serine protease enzymes. Human Protein C is produced in vivo primarily in the liver as a single polypeptide of 461 amino acids. This single chain precursor molecule undergoes post-translational modifications including 1) cleavage of a 42 amino acid signal sequence; 2) proteolytic removal of the lysine residue at position 156 and the arginine residue at position 157 to make a 2-chain zymogen (inactive form) of the molecule, (i.e., a light chain of 155 amino acid residues attached through a disulfide bridge to the serine protease-containing heavy chain of 262 amino acid residues); 3) vitamin K-dependent carboxylation of nine glutamic acid residues clustered in the first 42 amino acids of the light chain, resulting in nine gamma-carboxyglutamic acid residues; and 4) carbohydrate attachment at four sites (one in the light chain and three in the heavy chain). This process is essentially replicated during manufacture of drotrecogin alfa (activated). The heavy chain contains the well established serine protease triad of Asp 257, His 211 and Ser 360. The 2-chain zymogen is activated in vivo to active enzyme by the action of thrombin at a phospholipid surface in the presence of calcium ion. Activation of protein C (both recombinant and endogenous) results from removal of the dodecapeptide at the N-terminus of the heavy chain, resulting in activated protein C possessing proteolytic enzyme activity.

The activation of protein C requires thrombin (see related product entries in the Blood Products, Human section), the final serine protease in the coagulation cascade, and thrombomodulin, an endothelial cell membrane-associated glycoprotein. Thrombin complexed with thrombomodulin activates protein C at a rate constant some 20,000 fold higher than the rate constant for thrombin alone. Thrombomodulin forms a tight, stoichiometric complex with thrombin, altering the normal functional properties of thrombin. Thrombin complexed with thrombomodulin, unlike uncomplexed thrombin, does not clot fibrinogen, does not activate platelets, and does not convert clotting Factors V and VIII to their activated counterparts.

Activated Protein C in dogs has been shown to sharply increase circulating levels of tissue plasminogen activator (tPA), the major fibrinolytic (clot-dissolving) enzyme. Activated protein C has been shown in vitro to enhance the lysis of fibrin in human whole blood.

The protein C pathway also plays a role as a natural defense mechanism in inflammation. In primate models of sepsis, APC blocks disseminated intravascular coagulation (DIC) initiated by Escherichia coli infusion. Inhibition of APC function worsens both the coagulant and inflammatory responses of the animals to sublethal levels of E. coli.

Companies.: Xigris was developed and is marketed internationally (originally, worldwide) by Eli Lilly and Co., CBER/FDA est. no. 1611. Lilly began development of Xigris in 1990. In May 2011, with sales lagging, Lilly spun off Xigris U.S. marketing to BioCritica, a new company jointly owned by Lilly and two investment firms. This transfer appears not have been completed (before withdrawal of the product).

Drotrecogin alfa (activated) is manufactured by Lonza Biologics under contract to Lilly. As is common practice for contract manufacture, CBER/FDA reports Lilly, CBER/FDA est. no. 1611, as the manufacturer. In 1997, Lilly entered into a manufacturing agreement with Lonza for process optimization, scale-up and manufacture of drotrecogin alfa. A pilot plant in Slough, U.K., was installed, manufactured product for clinical trials, and was used to design a dedicated large-scale manufacturing facility at Lonza facilities in Portsmouth, NH.

Manufacture: Lilly developed a manufacturing process (used by Lonza) using a unique (not standard platform; undisclosed) process and medium for drotrecogin alfa manufacture using use hollow fiber capillary perfusion bioreactor(s). Expression levels are very low, <0.1 g/L (vs. 1-5 g/L for most recombinant monoclonal antibodies. The manufacturing process takes 30 or more days. The manufacturing process involves: expression in perfusion bioreactor(s); harvesting; filtration; viral inactivation; capture (unspecified); bags held at 2-8˚C; back-end purification (completed in 3 days); TFF; activation; viral filtration; chromatography purification; and controlled freezing.

A different source [Farid, S., Operational & Economic Evaluation of Integrated Continuous Biomanufacturing Strategies for Clinical & Commercial mAb Production, presented at ECI Integrated Continuous Biomanufacturing, Barcelona, Spain, 20-24 October 2013] cites perfusion bioprocessing using 2 x 1,500 L bioreactors with gravity settlers for perfusion.

As described in U.S. 4,992,373, assigned to Eli Lilly & Co., plasmid pLPC is used as an intermediate vector in the construction of plasmid pLAPC, containing a segment of DNA encoding the BK virus enhancer and the adenovirus 2 late promoter positioned to drive DNA for expression of human Protein C. The construction of plasmid pLAPC essentially results in the replacement of the human Protein C coding sequence on plasmid pLPC with another Protein C coding sequence from which activation peptide-encoding DNA has been removed. The host Human Embryonic Kidney Cell Line 293 used for manufacture of drotrecogin alfa is available from the American Type Culture Collection as ATCC CRL 1573. Fermentation is carried out in a nutrient medium containing the antibiotic gentimicin sulfate, but this is not detectable in the final product.

As described in U.S. 4,775,624, assigned to Eli Lilly & Co., purified recombinant drotrecogin alfa is prepared for activation by removal of residual calcium by passage over a metal affinity column (e.g., Chelex-100, Bio-Rad); and anion exchange chromatography (e.g., with Fast-Flow Q resin from Pharmacia). Drotrecogin alfa is activated by exposure to bovine thrombin, e.g., passage through a column packed with matrix (e.g., Activated CH-Sepharose 4B from Pharmacia)-bound bovine thrombin. Leached thrombin is removed by binding to an anion exchange resin (e.g., Fast Flow Q from Pharmacia).

FDA class: Biologic BLA

CBER class: Blood and Blood Products

CBER to CDER: Among the products transferred within FDA on June 30, 2003

Approvals: Date = 20011121; BLA (BL 125029/0)

Indications: [full text of "INDICATIONS AND USAGE” section of product insert/labeling]:

Xigris is indicated for the reduction of mortality in adult patients with severe sepsis (sepsis associated with acute organ dysfunction) who have a high risk of death (e.g., as determined by APACHE II, see CLINICAL STUDIES). Efficacy has not been established in adult patients with severe sepsis and lower risk of death. Safety and efficacy have not been established in pediatric patients with severe sepsis.

Status: Xigris is exempt from CBER/FDA lot release requirements.

European Union (MAA) approval was granted in Aug. 2002.

In March 2005, the FDA and Eli Lilly issued warnings to physicians that patients taking Xigris for severe sepsis face an increased chance of death. The warning, based on data from two clinical trials, relates only to patients who have recently had surgery and have a single organ dysfunction. FDA and Lilly reiterated that Xigris should only be taken after careful consideration of the risks.

On Feb. 4, 2009, FDA issued and Early Communication and announced that it is working with Lilly to further evaluate the incidence of serious bleeding events and death in patients who receive Xigris. Early Communications mean that the FDA has begun an ongoing safety review about an important safety issue that has not yet been fully analyzed or confirmed. A recent study and an accompanying editorial published in Critical Care Medicine had reported an increased risk of serious bleeding events and death in patients with sepsis and baseline bleeding risk factors who received Xigris. The study, a retrospective review of medical records of 73 patients who were treated with Xigris, found that serious bleeding events occurred in seven of 20 patients (35%) who had a bleeding risk factor versus two of 53 (3.8%) of patients without any bleeding risk factors. More patients with baseline bleeding risk factors died (13 of 20, or 65 percent) compared with patients without any bleeding risk factors (13 of 53, or 24.5 percent). As noted by the authors, limitations of this study include the retrospective design and small patient population. Note, the current prescribing information (labeling) includes a warning that describes bleeding as the most common serious adverse effect and lists a number of risk factors that should be carefully considered when deciding whether to use Xigris. The labeling contraindicates the use of Xigris in several clinical situations where bleeding could lead to significant adverse reactions or death.

On Oct. 26, 2011, Eli Lilly announced it was pulling Xigris from all markets worldwide following results of the PROWESS-SHOCK study, which did not meet the primary endpoint of a statistically-significant reduction in 28-day all-cause mortality in patients with septic shock. The 1,696-patient trial began in March 2008 as a condition for continued market authorization in Europe (it was approved in the EU in 2002 and in November 2001 in the USA). While there were no new safety findings, the study failed to demonstrate that Xigris improved patient survival and called into question its benefit-risk profile and continued use. Lilly reported that it expected to take a loss of $75-$95.0 million related to Xigris.

Tech. transfer: RE 37,806 was extended 497 days to Oct-24, 2016.

Lilly has received a number of patents covering aspects of activated Protein C. The gene encoding human protein C and recombinant methods are claimed in U.S. Pat. No. 4,775,624. Plasmid pLPC used to express human protein C in human kidney 293 cells is disclosed in 4,992,373. RE37806 describes use of the BK enhancer sequence with a promoter sequence, e.g. the adenovirus 2 late promoter, to drive expression of Protein C in 293 cells. Recovery and purification methods for vitamin K-dependent proteins, e.g., Protein C, are described in U.S. 5,981,952. U.S. 6,037,322 and 6,268,337 claim methods for treating vascular disorders using activated protein C. U.S. 6,159,468 and 5,981,952 disclose stable lyophilized formulations of activated Protein C free of autodegradation products, including use of sugars and buffers. U.S. 4,992,37 discloses vectors, including plasmid pLPC (apparently used for manufacture of drotrecogin alfa). U.S. 4,775,624 describes the activation of drotrecogin alfa. See the manufacture section for further information about 4,992,373 and 4,775,624. U.S. 6,242,192 includes descriptions of recombinant Protein C produced in 293 cells with novel glycosylation patterns.

On June 2002, ARIAD Pharmaceuticals filed a patent infringement suit against Lilly regarding Xigris [and also rolaxifene (Evista) for osteoporosis]. The suit was filed within hours of the granting of U.S. patent 6,410,516, assigned to the Massachusetts Institute of Technology (MIT); Whitehead Institute for Biomedical Research; and Harvard University (with two Nobel Prize winners among the inventors). This patent has been exclusively licensed by ARIAD. The patent concerns NF-kappaB protein, NF-kappaB signaling pathways and, to a limited extent (e.g., no actual examples) related NF-kappaB inhibitors. NF-kappaB is believed to play a role in a wide range of diseases, e.g., cancer, osteoporosis and bacterial infections , including sepsis. The patent was filed in 1986, after the inventors discovered the NF-kappaB protein, and took 16 years to make it through the patent office. The inventors demonstrated the presence but did not purify the NF-kappaB protein, and were unable to sequence and clone the gene (before another group cloned the full gene in 1991). The patent does not include any examples of pharmaceutical agents capable of inhibiting NF-kappaB, other than a single protein called “inhibitor of kappa B” or IKB (suggesting that the patent may encounter problems with enablement and written description requirements).

U.S. 6,410,516 has a life of 17 years after issuance (to 2019). ARIAD particularly asserted against Lilly that Xigris infringes claim 14, “ A method for reducing bacterial lipopolysaccharide-induced expression of cytokines in mammalian cells, which method comprises reducing NF-kappaB activity in the cells so as to reduce bacterial lipopolysaccharide-induced expression of said cytokines in the cells.” Claim 15 comparably claims reduction of tumor necrosis factor-alpha (TNF) expression. Lilly has published papers showing that Xigris inhibits NF-kappaB activity (J. Biol. Chem. 276: 11199-11203, 2001; Crit. Care Med. 30:S288-S293, 2002), with drotrecogin alfa (activated) reducing bacterial lipopolysaccharide-induced expression of cytokines through its inhibition of NF-kappaB.

Many have expressed concern about the broadness of 6,426,198, which has 203 claims. For example, Claim 1 states “A method for inhibiting expression, in a eukaryotic cell, of a gene whose transcription is regulated by NF-kappaB, the method comprising reducing NF-kappaB activity in the cell such that expression of said gene is inhibited.” This and other claims may be interpreted as covering anything that blocks the activity of the NF-kappaB. ARIAD argues that it covers therapeutics that inhibit NF-kappaB (even if this mechanism is not the molecule’s primary biological activity) discovered after its 1986 filing date, and that the patent’s breadth befits the pioneering nature of this discovery.

Experts are divided on whether the discovery is sufficient to merit a patent that lays claim to any agent acting on the pathway. The patent raises questions of whether a patent on a pathway/mechanism can cover therapeutics (e.g., Xigris) for which its in vivo mechanism(s) of activity remain unelucidated; covers therapeutics (e.g., Xigris) discovered well before the pathway was discovered; and, the converse, i.e., whether the existence of a therapeutic (e.g., Xigris) functioning through a pathway invalidates a patent on the pathway by making it insufficiently novel. Simply stated, the case depends on whether NF-kappaB inhibition is new and patentable, or whether it is a mechanism of action that inherently occurs whenever an NF-kappaB inhibitor is used. Lilly has Xigris-specific patent filing priority back to 1985; and notes that ARIAD can’t reach back and claim a prior invention (in the U.S., with its first-to-invent patent system), and that the very existence of Xigris (and Evista) invalidates the patent. Also, well before Xigris, drugs, nutraceuticals and herbs were available, e.g., ibuprofen, aspirin, caffeic acid, garlic, capsaicin and vitamin E, with activity now known to involve inhibition of NF-kappaB.

On May 4, 2006, a U.S. District Court of Boston jury trial ruled that Lilly’s two products, Xigris and Evista, infringed claims 80, 95, 144, and 145 of U.S. concerning NF-kappaB signaling pathways and related inhibitors, and that those claims were valid and enforceable. Lilly was ordered to pay $65.2 million in back royalties to ARIAD, and will also have to pay a 2.3% royalty on future U.S. sales Xigris and Evista through the life of the patent (into 2019). Lilly is appealing the verdict.

On August 2, 2006, the U.S. patent office (PTO) issued its initial Office Action (reexamination) regarding 6,410,516. This upheld as patentable ~20% of the claims of the patent and rejected the other claims. This action is a routine and expected step in the re-examination procedure and does not represent a final ruling.

In July 2007, a U.S. District Court judge ruled in favor of the plaintiffs (ARIAD and MIT) in their lawsuit against Lilly. The Judge found that the patent is valid and enforceable and that Lilly will have to pay the back and future royalties awarded by the jury. Many observers expect Lilly to prevail in the Court of Appeal for the Federal Circuit (CAFC), where a panels of knowledgeable judges will decide the case (although there are problems with the U.S., as described below).

Also in July 2007, U.S. patent office issued a final office action in its reexamination of 6,410,516, rejecting the patentability of certain claims of the patent, including those claims being asserted in the Lilly lawsuit and a separate lawsuit against Amgen, Inc. and Wyeth. ARIAD plans to respond promptly and, if necessary, will appeal the decision through the patent office and the courts. The patent remains valid and enforceable through the appeals process, which is expected to take two or more years.

On April 3, 2009, Eli Lilly won an U.S. Court of Appeals for the Federal Circuit ruling (Ariad Pharmaceutical v. Eli Lilly, 08-1248, U.S. Court of Appeals for the Federal Circuit ) that eliminated the $65.2 million and royalties on sales verdict won by Ariad Pharmaceuticals Inc. over royalties on Xigris. The Court found the Ariad patent's four claims in dispute to be invalid, because the patent failed to adequately describe the invention or explain how others could replicate its work. The Court rejected a Lilly argument that the patent should be deemed unenforceable because the inventors had misled the patent office in obtaining the patent. Ariad will likely appeal this verdict, and may still attempt to assert other claims of the patent against Lilly.

The Oklahoma Medical Research Foundation has received U.S. 5,695,993, “Cloning and regulation of an endothelial cell protein C/activated protein C receptor,” and other patents concerning cloning and uses of Protein C. Lilly had licensed the foundation’s inventions, but refused to pay the royalties after Xigris reached the market. The dispute was resolved amicably, presumably, with the foundation receiving a share of royalties or other payment(s).

Several researchers (Drs. Crabtree and Plutzky) and the National Institutes of Health (NIH) were involved in a patent dispute concerning the cloning of the human Protein C gene. The researchers, then with NIH, isolated and provided a clone of the gene to Lilly (which claims not to have used it). In mid-2003, Lilly brought suit in federal court seeking a declaratory judgment that its patent was valid (that the NIH researchers were not inventors). The researchers filed a counter-suit to be included as inventors on Lilly’s relevant patents and receive related royalties. The NIH joined suit in 2004 against Lilly, but is claiming that it, not the researchers, owns the inventions (from which the researchers will, presumably, get a share of licensing income). In April 2006, a federal court judge ruled in favor of Lilly, that the NIH researchers had only provided a fragment of the Protein C gene. The researchers and NIH are expected to appeal this ruling.

The University of Washington, having received a patent related to Protein C, had brought an infringement suit against Lilly. This was resolved with the court ruling that both patents were valid, but covered different inventions, and were enoninfringing.

In March 2005, Lilly licensed its patents related to Xigris and sepsis to Biosite Inc. for development of a point-of-care diagnostic for assay of levels of protein C to identify patients likely to respond to Xigris.

Presumably, Lonza uses its glutamine synthetase (GS system) technology for expression of drotrecogin alfa. This includes 5,846,534, originally assigned to Celltech Group plc, transferred to Lonza Biologics plc, a subsidiary of Alusuise-Lonza Group, concerning use of the glutamine synthetase as a selectable marker for identification and amplification of transformed mammalian, e.g., CHO, cells. Related patents include U.S. 5,770,359 and 5,747,308. The technology is coassigned to the University of Glasgow (which presumably receives a share of royalties). The GS system involves use of glutamine synthetase as a dominant selectable marker for use in co-amplification of non-selected genes and in transforming host cell lines to glutamine independence. The GS gene is spliced into recombinant vectors as a marker along with the gene(s) for the desired protein, with only successfully transformed cells capable of producing their own GS and surviving in glutamine-deficient culture media.

Disease: Sepsis (septic shock) involves systemic inflammatory response to infection or trauma, associated with and mediated by the activation of a number of host defense mechanisms including the cytokine network, leukocytes, and the complement and coagulation/fibrinolysis systems. Disseminated intravascular coagulation (DIC), with widespread deposition of fibrin (clotting) in the microvasculature of various organs, is an early manifestation of sepsis. DIC is an important mediator in the development of the multiple organ failure syndrome, and contributes to the poor prognosis of sepsis patients.

The pathology of sepsis is ill-defined, and includes severe sepsis and septic shock. Sepsis may be caused by infections with bacteria (usually Gram negative), fungi, viruses, and other infections, as well as by non-infective stimuli such as multiple trauma, severe burns, and organ transplantation. Although mostly associated with Gram-negative bacterial infections, sepsis can be caused by progression of infections caused by any one of a multitude of pathogens at various infection sites.

Sepsis usually begins with tremor, fever, falling blood pressure, rapid breathing and heart beat, and skin lesions. Within hours or days, it can progress to spontaneous clotting in the blood vessels (DIC), severe hypotension, multiple organ failure, and death. Patients with severe sepsis often experience failures of various organ systems, including the circulatory system and kidneys, as well as bleeding and clotting. Most sepsis-related damage comes not from the source infection or injury, but from endotoxins binding to cells, such as monocytes/macrophages or endothelial cells, triggering them to produce various inflammation mediators, such as tumor necrosis factor-alpha (TNF-alpha), and various interleukins (IL-1, IL-6, and IL-8). Such excess production of cytokines can elicit sepsis.

A major problem in treating sepsis is that there are no specific working definitions of the disease and poor diagnostics, factors often resulting in what many consider to be needless or over-treatment for the disease. There are still no precise clinical measurements or surrogate markers for sepsis, and diagnostic tools are still relatively poor. The best available diagnostic methods involve blood cultures, which take days to provide results (diagnose systemic infections), while sepsis can kill in hours. Due to this, it is estimated that only one in 17 of the 300,000 neonates treated for sepsis in the annually in the U.S. actually has the condition.

In 2001, sepsis was reported to be the 13th leading cause of death in the U.S.; and it is the single most common cause of death in non-coronary intensive care units. In April 2001, a study by Datamonitor estimated there are 1.8 million cases of sepsis each year in the major pharmaceutical markets (U.S., Europe and Japan), with mortality rates reaching 40% in the elderly. CDC/DHHS has reported that about 750,000 people are diagnosed sepsis in the U.S. each year, and an estimated 30% die from it, despite treatment with intravenous antibiotics and supportive care. In a 10-year study of sepsis cases released in late 2001, there were 684,000 cases in 1998, up 23.3% from 1988, with a 17.4% mortality rate in 1998. The incidence of sepsis continues to rise in developed countries, with U.S. increases of 1.5% per annum expected until 2020 (likely related to general aging of the population).

Trials: In a pivotal, placebo-controlled, multi-center, randomized, 28 day clinical trial in nearly 1,700 patients, Xigris reduced the relative risk of death (overall mortality) by 6% (from 31% to 25%), while 3.5% of Xigris-treated patients (compared to 2% of placebo recipients) experienced an increased risk of serious bleeding events. Among patients at highest risk of death, mortality was reduced 13% (from 44% to 31%). Many consider these results to be indicative of only marginal efficacy. Xigris did not lower mortality rates in patients who were less severely ill. Because activated Protein C (aPC) interferes with blood clotting, the most serious side effect associated with Xigris therapy is bleeding, including bleeding that may cause stroke. Patients at high risk of bleeding were excluded from the trial, as were severely ill patients with preexisting conditions not related to sepsis making them likely to die within the study period.

In patients with severe sepsis, Xigris infusions of 12 µg/kg/hr to 30 µg/kg/hr rapidly produce steady state concentrations (Css) of drotrecogin alfa proportional to infusion rates. In the majority of patients, plasma concentrations of aPC fell below the assay’s quantization limit of 10 ng/mL within 2 hours after stopping infusion. Plasma clearance of aPC in patients with severe sepsis is approximately 50% higher than that in healthy persons.

In trials in patients with severe sepsis, Xigris infusions of 48 or 96 hours produced dose dependent declines in D-dimer and interleukin-6 (IL-6). Compared to placebo, Xigris-treated patients experienced more rapid declines in D-dimer, PAI-1 levels, thrombin-antithrombin levels, prothrombin F1.2, IL-6, more rapid increases in Protein C and antithrombin levels, and normalization of plasminogen. As assessed by infusion duration, the maximum observed pharmacodynamic effect of Xigris on D-dimer levels occurred at the end of 96 hours of infusion for the 24 µg/kg/hr treatment group.

Results of a large study reported in the July 15, 2003, issue of Clinical Infectious Diseases found Xigris to provide significant short and long-term survival advantages for older patients with severe bacterial infection. Patients treated with Xigris had absolute risk reductions in 28-day and in-hospital mortality of 15.5% and 15.6%, respectively, compared with placebo recipients. Resource use and patient disposition compared favorably. The only adverse effect was a slight, but not statistically significant, increase in the incidence of serious bleeding in the Xigris group, with no age-related adverse effects. An accompanying editorial noted, “The survival advantages with reduced relative incidence of recurring disease and a lack of age-related adverse effects suggest that it is appropriate for the infectious disease community to prescribe DAA [Xigris] as a treatment for older patients who are at high risk for severe bacterial infections.”

In April 2005, Lilly halted a Phase IIIb trial (F1K-MC-EVBP) of Xigris for treatment of sepsis in ~400 children after it was shown to not be significantly more effective than placebo, with interim analysis showing it “highly unlikely” that the trial would meet the primary endpoint. Also, there was an increase in the rate of CNS bleeding in the Xigris group vs. placebo. Xigris in not approved for pediatric use, so this will not affect sales, no applications were planned based on this trial, and the trial was conducted by Lilly at the request of FDA (which is seeking to extend to pediatric populations use of therapeutics originally approved for adult use).

In the Sept. 29, 2005, issue of the New England Journal of Medicine results were reported from a post-marketing study showing that Xigris does not benefit septic patients at low risk of death, with Xigris efficacy comparable to that of placebo. The trial had been halted early after an interim analysis of data showed little reduction in risk of mortality. The study originally intended to enroll 11,444. At the time of its termination, only 2,640 were enrolled, roughly split between the placebo and treatment groups. No statistically significant differences were observed between the two groups in either hospital deaths or deaths 28 days after the start of infusion therapy. Howeve, the rate of serious bleeding was greater in the treatment group than the placebo group during both the infusion and the 28-day study period. An accompanying editorial supported FDA’s approval of limited indications: for Xigris. The principle investigator stated, “My own interim conclusion is that the drug should be reserved for patients with septic shock criteria and organ failure, should be given by experts in the ICU, given very early in the course of disease, and withheld from patients with risks of bleeding or who are recently recovering from surgery.”

In March 2006, Lilly licensed its patents to Biosite Inc. for development of a point-of-care diagnostic for assay of levels of protein C to identify patients likely to respond to Xigris. Lilly plans to use this and other aPC tests in its Phase IIb RESPONSE trial of Xigris for severe sepsis, which is expected to start in the fourth quarter of 2006. Such a diagnostic, allowing better identification of patients likely to respond to Xigris, would improve its marketablity, cost-benefit, etc.

In the Oct. 16, 2006, issue of the New England Journal of Medicine, a group of investigators from the National Institues of Health (NIH) accused Lilly of too directly funding and influencing guidelines for use of Xigris, including a $1.8 million grant to the Surviving Sepsis Campaign, and using treatment guidelines as a tool to drive sales.

In Feb. 2007, Lilly announced plans for a new Phase III study of Xigris “to help clinicians better identify severe sepsis patients at high risk of death who are more likely to benefit from this novel therapy and to further clarify the drug’s benefit/risk profile…Advancements made in sepsis care over the past five years, and ongoing scientific questions surrounding appropriate patient selection for Xigris and about severe sepsis treatment in general make this an opportune time for a new Phase III placebo-controlled study of Xigris” The new trial followed discussions with the European Medicines Agency (EMEA), European Union, in the context of its fourth annual license reassessment for Xigris. The primary endpoint of the trial will be 28-day all-cause mortality.

Medical: Sepsis is a life-threatening condition. Treatment with Xigris entails serious risks and provides what some consider to be only marginal benefits. Xigris was approved by FDA only for the treatment of adult patients with severe sepsis who have an especially high risk of dying from sepsis, as measured by a scoring system based on their general health and the severity of their illness.

Market: Total worldwide (mostly in U.S.) sales of Xigris were $127.3 million in 2009; $166 million in 2008; $161 million in 2007; $144 million in 2007; ~$299 millioin in 2006; $214.6 million in 2005; $201.8 million in 2004; $160.4 million in 2003; and about $100 million in 2002.

Analysts with Friedman Billings and Ramsey (FRB) had projected sales of $303 million in 2006, and $365 million in 2007.

Since launch, sales of Xigris have been steadily increasing, but at a rate and with total sales very much lower than the blockbuster (>$1 billion/year) levels that had been expected.

The 2007 Average Wholesale Price (AWP) is $313.92/5 mg vial, with a Direct Price (discount price) of $261.60; and $1,255.68/20 mg vial, with a Direct Price of $1,046.40 (Red Book, 2007).

The cost for a typical 4-day course of treatment with Xigris is commonly reported to be $6,800. Hospitals have been reluctant to keep stocks of the drug in-house due to its price and shelf life; and reluctant to use it for patient treatment. Antibiotics are considered by many to be comparably effective and are relatively inexpensive for treatment of less severe sepsis patients.

Despite disappointing sales largely due to price considerations, Lilly does not plan to reduce the price, citing its value to patients and high cost of development and manufacture as determining the product’s cost. Lilly distributes Xigris directly to hospitals as needed, rather than through wholesalers.

Concerns about high costs and risks, limited benefits, and its restricted approved indications: are major factors restricting use of Xigris to the most severe (less common) cases of sepsis. Offering only incremental an increase in clinical efficacy (survival), potential serious risks, and high cost, many physicians and insurance companies simply avoid using Xigris, except for the most severe patients. In July 2002, a survey of over 600 members of the Society of Critical Care Medicine found Xigris to be the most rationed therapeutic/procedure in intensive care units, with 27% reported limiting use/rationing its use in the past year.

Xigris’ sales are far short of the blockbuster ($1 billion+) sales Lilly and many others had originally expected. Fewer than 15,000 prescriptions for Xigris were filled in 2002. Factors slowing the initial uptake of Xigris include delays by larger institutions in developing treatment protocols and adding Xigris to their formularies. Other factors include the inherent uncertainties associated with sepsis, lingering doubts over the efficacy and safety of Xigris, and its high price. Also, newer broad spectrum, potent antibiotics, e.g., fluoroquinolones, carbapenems and glycopeptides, are being effectively used more for sepsis treatment.

Xigris was projected by many analysts to have blockbuster (>$1 billion/year) sales potential for Lilly, and analysts predicted sales ranging from $300 to $500 million in 2002. During development of Xigris, its closest competitor, tifacogin, failed in late-stage clinical trials, leaving Xigris as the only new sepsis treatments in late-stage development. In March 2002, with marketing difficulties then apparent, Datamonitor predicted sales up to $273 million/year by 2005.

With many hospitals and health care organizations rationing Xigris due to perceived high cost-low benefit relationship, Lilly has been making efforts to improve Xigris’ image and increase its acceptance. As reported in a front page story in the Sept. 18, 2003, Wall Street Journal, Lilly has been conducting an “unconventional campaign...stoking a controversy over who should get expensive treatments in times of limited resources.” In late 2003, Lilly’s efforts included a public relations campaign entitled, “The Ethics, the Urgency and the Potential,” implying to some that it is unethical not to use Xigris. Lilly also funded a $1.8 million project called the “Values, Ethics & Rationing in Critical Care Task Force,” in which bioethicists and physicians from various U.S. medical schools evaluate the ethics of rationing certain pharmaceuticals and services, including Xigris. Also, Lilly has been lobbying for Medicare coverage of one-half or more the cost of Xigris.

In July 2002, Xigris became the first new pharmaceutical granted new technology status by the Centers for Medicare and Medicaid Services (CMS; formerly HCFA), the agency administering Medicare/Medicaid, under the provisions of the Benefits Improvement Act of 2000 (BIPA). Starting Oct. 2002, this allows hospitals using Xigris for treatment of Medicare patients with life-threatening sepsis to receive additional reimbursement, with CMS essentially paying 50% of the excess over the standard DRG payment. Special payments for new technologies will likely be made for two to three years, until guidelines are updated. Medicare coverage has been projected to add an additional $37 million/year in sales.

Some more extreme right-to-life (anti-abortion) organizations, such as Children of God for Life, have urge protests against Lilly and Xigris, because aPC is manufactured using the human embryonic kidney 293 cell line (HEK-293; originally derived from an aborted fetus), e.g., see the group’s press release issued Aug. 28, 2002.

In a letter published in the Oct. 19, 2006, edition of the New England Journal of Medicine, four researchers from the National Institutes of Health (NIH) accused Lilly of orchestrating the International Sepsis Forum’s medical guidelines for sepsis, which included Xigris as a valid treatment, charging that Lilly worked through medical societies to influence standards for treatment of sepsis, including substantially funding guidelines committees and associated “Surviving Sepsis Campaign” informational campaigns. The campaign, which began in Oct. 2002, allegedly failed to discuss evidence of increased risk of serious bleeding when using Xigris. One of the authors, Dr. P.Q. Eichacker, had voted against approval as an adviser the FDA advisory committeee that reviewed the original approval of Xigris. Lilly denied any role beyond funding. The published guidelines acknowledged that Lilly donated more than 90% of $861,000 in grants for the educational campaign and medical recommendations. Various guidelines committee members disagree on the role Lilly played.