Data-driven Temporal Prediction of Surgical Site Infection.

Analysis of data from Electronic Health Records (EHR) presents unique challenges, in particular regarding nonuniform temporal resolution of longitudinal variables. A considerable amount of patient information is available in the EHR - including blood tests that are performed routinely during inpatient follow-up. These data are useful for the design of advanced machine learning-based methods and prediction models. Using a matched cohort of patients undergoing gastrointestinal surgery (101 cases and 904 controls), we built a prediction model for post-operative surgical site infections (SSIs) using Gaussian process (GP) regression, time warping and imputation methods to manage the sparsity of the data source, and support vector machines for classification. For most blood tests, wider confidence intervals after imputation were obtained in patients with SSI. Predictive performance with individual blood tests was maintained or improved by joint model prediction, and non-linear classifiers performed consistently better than linear models.

Concomitant inhibition of Janus kinase 3 and calcineurin-dependent signaling pathways synergistically prolongs the survival of rat heart allografts.

The cytoplasmic localized Janus tyrosine kinase 3 (Jak3) is activated by multiple cytokines, including IL-2, IL-4, and IL-7, through engagement of the IL-2R common gamma-chain. Genetic inactivation of Jak3 is manifested as SCID in humans and mice. These findings have suggested that Jak3 represents a pharmacological target to control certain lymphoid-derived diseases. Using the rat T cell line Nb2-11c, we document that tyrphostin AG-490 blocked in vitro IL-2-induced cell proliferation (IC(50) approximately 20 microM), Jak3 autophosphorylation, and activation of its key substrates, Stat5a and Stat5b, as measured by tyrosine/serine phosphorylation analysis and DNA-binding experiments. To test the notion that inhibition of Jak3 provides immunosuppressive potential, a 7-day course of i.v. therapy with 5-20 mg/kg AG-490 was used to inhibit rejection of heterotopically transplanted Lewis (RT1(l)) heart allografts in ACI (RT1(a)) recipients. In this study, we report that AG-490 significantly prolonged allograft survival, but also acted synergistically when used in combination with the signal 1 inhibitor cyclosporin A, but not the signal 3 inhibitor, rapamycin. Finally, AG-490 treatment reduced graft infiltration of mononuclear cells and Stat5a/b DNA binding of ex vivo IL-2-stimulated graft infiltrating of mononuclear cells, but failed to affect IL2R alpha expression, as judged by RNase protection assays. Thus, inhibition of Jak3 prolongs allograft survival and also potentiates the immunosuppressive effects of cyclosporin A, but not rapamycin.

Inhibition of C-raf expression by antisense oligonucleotides extends heart allograft survival in rats.

BACKGROUND: C-raf is a well-characterized serine/ threonine (Ser/Thr) protein kinase that is involved in the transduction of multiple signals of T cells. We demonstrate that the inhibition of C-raf mRNA expression prolongs heart allograft survival. METHODS: Three 20-mer C-raf antisense oligonucleotides, each with identical sequences, were synthesized with different chemical modifications: one as a uniform phosphorothioate oligodeoxynucleotide (PS oligo), a second with a PS backbone and 2'-methoxyethyl (ME) substitutions at the 2'-sugar positions in the first and last five nucleotides, and a third with a mixed PS and phosphodiester (PD) backbone and ME modifications on the first and last five nucleotides. RESULTS: Both ME-modified C-raf antisense oligos were at least 5-fold more effective than the PS C-raf antisense oligo in blocking C-raf mRNA expression in two cell lines. Similarly, each of the ME C-raf antisense oligos produced better heart allograft survival rates than did PS C-raf oligo. Furthermore, although the combination of PS C-raf antisense oligo with sirolimus (SRL) acted synergistically to extend heart allograft survival, the effect was potentiated by either of the ME-modified oligos. CONCLUSIONS: C-raf inhibition extends heart allograft survival, and ME-modification potentiates antisense activity.

ICAM-1 antisense oligodeoxynucleotide improves islet allograft survival and function.

Expression of intercellular adhesion molecule-1 (ICAM-1) and its ligand, leukocyte function antigen-1 (LFA-1), after pancreatic islet transplantation may affect both nonspecific and alloantigen-specific phases of graft destruction. We examined the effects of ICAM-1/LFA-1 blockade on the survival of islet allografts. Fresh C57BL/10 (H2h) pancreatic islets were transplanted under the renal subcapsular space (KC) or embolized into the liver after portal vein (PV) injection to C3H (H2k) mice. Recipients remained untreated or were treated for 7 days by i.p. administration of: ICAM-1 antisense phosphorothioate oligodeoxynucleotide (oligo) alone; anti-1CAM-1 (alphaICAM-1) monoclonal antibody (mAb) alone: alphaLFA-1 mAb alone; ICAM-1 oligo/alphaLFA mAb combination; alphaICAM-1 mAb/alphaLFA-1 mAb combination; or control oligo IP-8997 or IP-1082. In some experiments, donors were pretreated with ICAM-1 oligo. Inhibition of single ligand with 5.0 mg/kg ICAM-1 oligo (25.1 +/- 10.3), 100 microg/daily alphaICAM-1 mAb (24.2 +/- 8.0 days), or 50 microg/daily alphaLFA-1 mAb (42.8 +/- 25.9 days) prolonged the survivals of KC islet allografts in comparison with untreated controls (11.9 +/- 1.0 days; all p < 0.01). However, dual ICAM-1/LFA-1 blockade with either ICAM-1 oligo/alphaLFA-1 mAb (78.3 +/- 16.5 days) or (alphaICAM-1 mAb/aLFA-1 mAb (65.2 +/- 31.3 days) was the most effective therapy. Although pretreatment of donors with ICAM-1 oligo alone was ineffective (12.2 +/- 0.8 days; NS), a combination of donor pretreatment and recipient treatment started 1 day prior to grafting with ICAM-1 oligo (39.2 +/- 14.0 days) was more effective than the recipient treatment alone (24.6 +/- 8.8 days). Furthermore, ICAM-1/LFA-1 blockade improved islet function as evaluated by glucose tolerance test, and decreased inflammation in comparison with untreated controls. Similar in vivo results were obtained following PV administration of islet allografts. Thus, ICAM-1/LFA-1 blockade prolongs the survival of pancreatic islet allografts and improves their early function.

Perfusion of kidneys with unformulated "naked" intercellular adhesion molecule-1 antisense oligodeoxynucleotides prevents ischemic/reperfusion injury.

BACKGROUND: We have previously shown that phosphorothioate intercellular adhesion molecule (ICAM)-1 antisense oligodeoxynucleotide (oligo) IP-9125 blocks the expression of rat ICAM-1 mRNA in rat L2 cells. A single ex situ perfusion of grafts with unformulated IP-9125, suspended in Euro-Collins solution, prolonged the survival of kidney allografts in rats. The present experiments examined whether perfusion of kidneys with unformulated IP-9125 prevents ischemic/reperfusion injury. METHODS: Kidneys were perfused ex situ with 2 ml of Euro-Collins solution without or with IP-9125 and exposed to 30-min cold (4 degrees C storage time) and 30-min warm (anastomosis time) ischemia. Kidneys were then transplanted to syngeneic nephrectomized recipients. RESULTS: Within 24 hr after transplantation, the glomerular filtration rate values were reduced by almost 60% to 0.49+/-0.14 ml/min from 1.20+/-0.27 ml/min in normal kidneys (P<0.001). Kidney perfusion with 10 mg of either IP-12140 (0.41+/-0.07 ml/min) or IP-13944 (0.47+/-0.07 ml/min) control oligo was ineffective. In contrast, perfusion with 10 mg of IP-9125 significantly improved kidney function (0.8+/-0.18 ml/min; P<0.005), whereas the lower doses of 2 mg (0.47+/-0.13 ml/min; NS) or 4 mg (0.54+/-0.04 ml/min; NS) had no significant effect. The glomerular filtration rate results were confirmed by measurements of blood creatinine (CR) levels at 24 hr after grafting: untreated recipients had a twofold higher CR value (0.70+/-0.14 mg/dl) compared with normal controls (0.65+/-0.07 mg/dl; P<0.001). Although perfusion with 10 mg of control IP-12140 (0.80+/-0.14 mg/dl) or IP-13944 (0.65+/-0.07 mg/dl) did not affect CR levels, perfusion with 10 mg of IP-9125 (0.45+/-0.07 mg/dl) lowered CR levels. The Western blots or reverse transcription-polymerase chain reaction experiments performed in kidney transplants within 24 hr after grafting showed that 10 mg of IP-9125 (but not control IP-12140) reduced the expression of ICAM-1 protein and ICAM-1 mRNA, respectively. CONCLUSIONS: Perfusion of grafts with unformulated ICAM-1 antisense oligo specifically reduces intragraft ICAM-1 protein expression and prevents ischemic/reperfusion injury.

Prophylaxis of acute renal allograft rejection using FTY720 in combination with subtherapeutic doses of cyclosporine.

BACKGROUND: In rodent transplant models, FTY720 exerts a synergistic affect with cyclosporine (CsA) to prolong allograft survival. The present experiments sought to test this combination in subhuman primates. METHODS: Cynomolgus monkeys were transplanted with kidney allografts that were incompatible in mixed lymphocyte culture reactions. The animals were treated with daily intramuscular injections of CsA using doses selected to maintain whole blood trough concentrations at therapeutic values between 40 and 200 ng/ml. The 4 experimental groups included CsA without or with 0.1, 0.3, or 1 mg/kg/day FTY720 delivered daily by intravenous bolus injection. Therapeutic effects were suggested both by the graft histology of biopsy within the first 10 posttransplant days and by the length of host survival. RESULTS: Whereas recipients treated with CsA alone rejected kidney allografts at a median survival time of 8.5 days (n=4), those treated with either 0.1 or 0.3 mg/kg/day FTY720 in addition to CsA showed significant prolongation of kidney allograft survival to 71 days (n=3; P<0.04) or 63 days (n=5; P<0.05), respectively. The hosts in the 1.0 mg/kg/day FTY720 group survived 48 days, with 2 of 5 recipients succumbing at 9 or 17 days postgraft, suggesting possible complications caused by overimmunosuppression. Biopsies of the 0.1 mg/kg/day FTY720 group on posttransplant day 7 documented mild to moderate rejection (grade I), indicated by multiple focal areas of tubular destruction. The histology results of transplants in the 0.3 or 1 mg/kg/day FTY720 group showed only minimal interstitial inflammatory infiltrates (borderline grade), with no evidence of tubular or arterial damage. Serum creatinine values among the animals in the 0.1 mg/kg/day FTY720 group showed increases in 2 of 3 recipients by day 20 and in the third by day 41 postgraft. Among the 0.3 mg/kg/day FTY720 group, 3 of 5 recipients maintained baseline creatinine values to 45 days postgraft; 1 recipient had stable kidney function for 120 days postgraft. CONCLUSIONS: Addition of FTY720 therapy to a subtherapeutic CsA immunosuppressive regimen delays the rejection of renal allografts in subhuman primates.

Protection against allograft rejection with intercellular adhesion molecule-1 antisense oligodeoxynucleotides.

BACKGROUND: We designed an antisense phosphorothioate oligodeoxynucleotide (oligo) to specifically inhibit the expression of rat intercellular adhesion molecule-1 (ICAM-1) mRNA (IP-9125). METHODS: IP-9125 oligo was delivered intravenously by osmotic pump alone or in combination with cyclosporine (CsA) to recipients in order to prevent the rejection of kidney or heart allografts. In additional experiments, kidney allografts were perfused with IP-9125 before grafting. RESULTS: IP-9125 inhibited ICAM-1 mRNA and ICAM-1 protein expression in rat aortic endothelial cells; scrambled controls IP-12140 and IP-13944 were ineffective. Untreated ACI (RT1a) recipients rejected Lewis (RT1l) kidney allografts at a mean survival time of 8.5+/-1.1 days. A 14-day intravenous administration of 2.5 mg/kg/day IP-9125 prolonged the survival of kidney allografts to 39.2+/-16.4 days; 5.0 mg/kg/day, to 43.0+/-17.5 days; and 10.0 mg/kg/day, to 50.4+/-21.6 days. In contrast, a scrambled control IP-12140 was not effective. A combination of 10 mg/kg/day IP-9125 and 1.0 mg/kg/day CsA delivered for 14 days synergistically extended kidney allograft survival times 88.5+/-7.5 days. In contrast, the combination of 10.0 mg/kg/day control IP-12140 with CsA was ineffective (20.7+/-3.2 days) when compared with CsA alone (20.2+/-4.0 days). Similar results were obtained for heart transplants in recipients treated with IP-9125 alone or in combination with CsA. Furthermore, in situ immunostaining showed that IP-9125 significantly reduced the expression of ICAM-1 protein in kidney allografts. Finally, perfusion of kidney grafts alone with 20.0 mg per 2 ml of IP-9125 protected kidney allografts from rejection (37.5+/-7.5 days; P < 0.001), whereas perfusion with 20 mg per 2 ml of control IP-12140 was ineffective (12.6+/-5.0 days). CONCLUSIONS: Rat ICAM-1 IP-9125 oligo inhibits ICAM-1 protein expression in vitro and in vivo as well as blocks allograft rejection when used for pretreatment of donors, graft perfusion, or postoperative treatment of recipients.

Immunosuppressive effects of FTY720 alone or in combination with cyclosporine and/or sirolimus.

BACKGROUND: We examined the ability of FTY720, a novel immunosuppressant that prolongs the survival of allografts in experimental animal models, to potentiate the immunosuppressive effects of cyclosporine (CsA) and/or sirolimus (SRL) in vitro and in vivo. METHODS: FTY720 alone (10-5000 ng/ml) or in combination with other drugs was added to human peripheral blood lymphocytes (PBLs) undergoing stimulation in vitro with phytohemagglutinin (PHA) or OKT3 monoclonal antibody. The combination index (CI) values were calculated to evaluate the nature of the interactions between FTY720 and CsA and/or SRL: CI values <1 reflect synergistic, CI=1, additive, and CI>1, antagonistic interactions. In addition, Wistar Furth (RT1u) rat recipients of Buffalo (RT1b) heart allografts were treated with FTY720 alone or in combination with other agents. FTY720 alone was also tested to block small bowel or liver allograft rejection in rats. RESULTS: FTY720 alone produced only modest inhibition of the proliferation of human PBL stimulated with PHA or OKT3 monoclonal antibody. In combination with CsA or SRL, however, FTY720 produced synergistic effects, namely, CI values of 0.58 and 0.36, respectively. A 14-day course of FTY720 (0.05-8.0 mg/kg/day) by oral gavage prolonged heart allograft survival in dose-dependent fashion. Although a 14-day oral course of CsA (1.0 mg/kg/day) alone was ineffective (mean survival time=7.0+/-0.7 vs. 6.4+/-0.6 days in treated vs. untreated hosts), treatment with a combination of 1.0 mg/kg/day CsA and 0.1 mg/kg/day FTY720 extended allograft survival to 62.4+/-15.6 days (P<0.001; CI=0.15). Similarly, a 14-day oral course of 0.08 mg(kg/day SRL alone was ineffective (6.8+/-0.6 days; NS), but the combination of SRL with 0.5 mg/kg/day FTY720 extended the mean survival time to 34.4+/-8.8 days (CI=0.28). The CsA/SRL (0.5/0.08 mg/kg/day) combination acted synergistically with FTY720 (0.1 mg/kg/day) to prolong heart survivals to >60 days (CI=0.18). CONCLUSIONS: FTY720 potentiates the immunosuppressive effects of CsA and/or SRL both in vitro (by inhibiting of T-cell proliferative response) and in vivo (by inhibiting allograft rejection).

Relative tissue distributions of cyclosporine and sirolimus after concomitant peroral administration to the rat: evidence for pharmacokinetic interactions.

The authors sought to determine the effect of concomitant peroral (PO) administration of cyclosporine (CsA) and sirolimus (SRL, rapamycin) on the tissue distributions of CsA and SRL in the rat. Groups of four adult male Wistar-Furth rats were treated for 14 days with 2.5, 5.0, or 10.0 mg CsA/kg x day. Other groups of four adult male Wistar-Furth rats were treated for 14 days with a 1-to-6.25 weight-to-weight ratio of SRL to CsA at SRL doses of 0.4, 0.8, or 1.6 mg/kg x day. Concentrations of CsA and SRL in homogenates of heart, intestinal, kidney, liver, lung, muscle, spleen, and testes were compared to those in whole blood (WB). There was a large, dose-dependent, distinctive distribution of CsA among rat tissues, as has previously been well documented. At a constant molar dose ratio, concomitant oral administration of SRL produced an approximately two-fold increase in the concentrations of CsA in rat tissues, although SRL did not change the CsA tissue-to-WB partition coefficients. Concomitant oral CsA administration produced dose-dependent increases in SRL tissue concentrations and decreases in the SRL tissue-to-WB partition coefficients. The increases in tissue and WB concentrations on coadministration of both agents may be explained either by an increase in absorption caused by competition between the two agents for binding sites on P-glycoprotein in the gut, a reduced rate of metabolism, or to an as yet unidentified elimination mechanism. The dose-independent and unchanged CsA tissue-to-WB partition coefficients suggest that SRL does not affect the equilibrium of CsA between the central and tissue compartments, namely the tissue uptake or intracellular binding. Altered values of the SRL tissue-to-WB partition coefficients suggest that, under the conditions studied, CsA disturbs the equilibrium of SRL between the central and tissue compartments.

Cytokine mRNA expression in tolerant heart allografts after immunosuppression with cyclosporine, sirolimus or brequinar.

We sought to examine the impact of the preferential activation of Th2 cells on the induction and maintenance of a tolerant state in heart allograft rat recipients treated with a short course of cyclosporine (CsA), sirolimus (SRL) or brequinar (BQR). A quantitative polymerase chain reaction (PCR) method was used to measure the levels of cytokine mRNAs, namely interferon (IFN)-gamma and interleukin (IL)-2 in T helper 1 (Th1) cells and IL-4, IL-5 and IL-10 in Th2 cells. Our main findings were that on day 5 postgrafting allografts from untreated recipients had increased levels of IFN-gamma (216 +/- 119 fg), IL-2 (449 +/- 75 fg), IL-4 (6.2 +/- 1.3 fg), IL-5 (34.8 +/- 9.3 fg) and IL-10 (1554 +/- 184 fg) mRNAs compared with normal hearts. CsA reduced the levels of IFN-gamma, IL-2, IL-5 and IL-10, but not IL-4, mRNAs. SRL did not affect the expression of cytokine mRNAs. BQR decreased the levels of IFN-gamma, IL-2 and IL-10, but not IL-5 or IL-4 mRNAs. Compared with grafts from untreated recipients, those from CsA- or BQR-treated tolerant hosts (day 100) displayed undetectable IL-2 mRNA levels, and reduced levels of IFN-gamma, IL-4 and IL-10 mRNAs. In fact, the patterns of cytokine mRNA expression in grafts from CsA- and BQR-treated tolerant hosts were similar to those of normal hearts. Grafts from SRL-treated tolerant hosts merely showed slightly increased Th2 cell activity. In conclusion the selective activation of Th2 cells is not absolutely required for induction or maintenance of tolerance.

Induction of tolerance toward rat cardiac allografts by treatment with allochimeric class I MHC antigen and FTY720.

BACKGROUND: The combination of FTY 720, a novel immunosuppressant, and allochimeric class I MHC proteins bearing donor-type amino acid (aa) epitope substitutions for host-type sequences induces tolerance of Wistar Furth (WF; RT1.Au) heart allografts in ACI (RT1.Aa) recipients. METHODS: Allochimeric alpha(1h)l58-80-RT1.Aa proteins were produced by substituting the allogeneic nucleotide sequence encoding 10 aa residues unique to the alpha1 helical (alpha1h) region of RT1.Al Lewis (Asp58, Arg62, Glu63, Gln65, Lys66, Gly69, Asn70, Asn73, Ser77, and Asn80) for native RT1.Aa residues. The RT1.Au and the RT1.Al haplotypes share four of these aa (Arg62, Glu63, Gln65, and Gly69). A baculovirus/Spodoptera frugiperda insect cell system was used to express the alpha(1h)l58-80-RT1.Aa proteins. RESULTS: The addition of a 3-day oral gavage of 0.05 mg/kg/day FTY720 to a single portal vein injection of 10 microg alpha(1h)l58-80-RT1.Aa protein induced permanent acceptance of WF heart allografts in 16 of 26 ACI recipients (>100 days); the alpha(1h)l58-80-RT1.Aa protein alone only modestly prolonged WF heart survival (13.8+/-0.8 days). The same tolerogenic protocol did not prolong the survival of third-party Brown Norway (RT1.An) heart allografts (14.3+/-2.5 days) compared with FTY720 alone (14.0+/-2.3 days; NS). Tolerant ACI recipients bearing primary WF heart allografts for more than 100 days accepted second WF hearts, but promptly rejected third-party Brown Norway heart grafts (9.3+/-1.5 days). The tolerant state was transferred to irradiated ACI rats (400 rad) with either purified T cells (4-10 x 10[7]) or serum (1-2 ml) from tolerant hosts, and was not broken by daily intraperitoneal injections of interleukin-2 (1000 U/day; 7 days). CONCLUSIONS: The combination of allochimeric protein with FTY720 induces transplantation tolerance, a state that may be associated with the appearance of donor-specific regulatory factors.

Synergistic mechanisms by which sirolimus and cyclosporin inhibit rat heart and kidney allograft rejection.

The studies presented herein examined the mechanism(s) whereby sirolimus (SRL) and cyclosporin (CsA) act synergistically to block allograft rejection. Combination index (CI = 1 reflects additive, CI > 1 antagonistic, and CI < 1 synergistic, effects) analysis documented potent synergism between SRL and CsA to block allograft rejection. Combinations of the two drugs produced synergistic prolongation of heart (CI = 0.001-0.2) or kidney (CI = 0.03-0.5) allograft survival at SRL/CsA ratios ranging from 1:12.5 to 1:200. Pharmacokinetic analysis of the individual drugs showed that CsA does not affect the blood levels of SRL, and SRL mildly increases the levels of CsA in SRL/CsA-treated rats. Quantitative polymerase chain reaction analysis was used to document that both subtherapeutic (1.0 mg/kg) and therapeutic (2.0 or 4.0 mg/kg) CsA doses inhibited the expression of interferon-gamma (IFN-gamma) (P < 0.03) and IL-2 (P < 0.003) mRNA produced by T helper (Th) 1 cells, as well as IL-10 (P < 0.001), but not IL-4 (NS) mRNA produced by Th2 cells. Contrariwise, all tested SRL doses (0.02, 0.04 or 0.08 mg/kg) did not affect cytokine mRNA expression. However, heart allografts from rat recipients treated with synergistic SRL/CsA doses displayed reduced levels of IFN-gamma (P < 0.01), IL-2 (P < 0.001) and IL-10 (P < 0.001) mRNA. Thus, because subtherapeutic doses of CsA reduce Th1/Th2 activity, thereby facilitating the inhibition of signal transduction by low does of SRL, the two agents act synergistically to inhibit allograft rejection.

Distribution of sirolimus in rat tissue.

OBJECTIVES: To examine the distribution of sirolimus (SRL, rapamycin), an immunosuppressive macrolide antibiotic, in the tissues of adult male Wistar-Furth rats following continuous intravenous infusion (CIVI) and repeated daily peroral gavage (PO). DESIGN AND METHODS: Animals received 14-day courses of SRL by either CIVI (0.04-0.4 mg/kg/day) or PO (0.4-1.6 mg/kg/day) administration. Samples of whole blood and homogenates of five solid organs (heart, kidney, liver, lung and spleen), and portions of intestinal, muscle and testicular tissues were prepared on day 13 of CIVI treatment or 24 hours after administration of the 14th PO dose. SRL concentrations were determined by high performance liquid chromatography with reference to calibration curves produced from SRL-spiked whole blood or tissue homogenates prepared from drug-free animals. RESULTS: Following PO but not CIVI administration, SRL concentrations in whole blood and all tissues increased linearly in relation to dose. SRL was extensively distributed among most tissues tested (tissue partitions coefficients of > 40 were observed in some cases). Comparatively, SRL whole blood concentrations were low. The ratio between the SRL whole blood concentrations after PO versus after CIVI administration (at like doses of 0.4 mg/kg/day) was 0.04. Therefore, we inferred that the oral bioavailability of SRL was low. CONCLUSIONS: The linear relationships between PO dose and SRL concentrations in whole blood and tissues may be attributed to the low oral bioavailability of SRL, which is indicated by the low levels of SRL observed in whole blood and tissues after PO administration. The nonlinear relationships between CIVI dose and SRL concentrations in whole blood and tissues may result because although whole blood depots may be saturated with SRL, the tissues continue to absorb SRL as the dose of SRL increases. Thus, because a high percentage of SRL is widely distributed into tissues stores, caution must be used when administering this drug in humans.