Posters –
Tuesday May 23, 2006 8:00AM Hepatitis C
HCV
Treatment Issues and Early Drug Development
S. Seiwert; S. W. Andrews; H. Yang; H. Tan; B.
Marafino; R. Rieger; R. B. Franklin; J. Pheneger; P. A. Lee; Y. Jiang; A. L.
Kennedy; S. M. Wenglowsky; M. R. Madduru; G. A. Doherty; K. R. Condroski; C.
Lemieux; L. Pieti Opie; F. Sullivan; N. Neitzel; G. P. Hingorani; J. Otten; B.
Brandhuber; G. Vigers; J. A. Josey; L. M. Blatt
Current
treatment options for chronic HCV infections provide sustained virologic
response rates of ~50%, indicating novel therapeutic approaches are needed. We
therefore embarked on a rational drug design campaign to produce inhibitors of
the HCV NS3/4A protease. ITMN 191 emerged from this discovery effort and was
nominated as a preclinical candidate. Here we describe the preclinical
characterization of ITMN 191.
In
biochemical assays using HCV NS3/4A protease domains derived from genotypes 1b,
1a, 2, or 3, the EC50 value of ITMN 191 is <300 pM, 400 pM, 400
pM, and 12.4 nM, respectively. Its EC50 value against full-length
genotype-1b protease is 900 pM. ITMN 191 retains subnanomolar to low-nanomolar
potency against the NS3/4A variants at positions A156 and D168 that confer
resistance to other experimental NS3/4A inhibitors. In a genotype-1b replicon
system EC50 = 2.1 nM and EC90 = 5.6 nM. ITMN 191 is 97.9%
bound by human serum protein, and its replicon potency is shifted modestly by
human serum. ITMN 191 does not inhibit a panel of selected serine proteases, a
broader ligand panel, or hERG ion channel. Its CC50 value is >50
μM. ITMN 191 does not inhibit CYP isoforms 1A2, 2C19, 2C9, or 2D6 at 10
μM; it inhibits roughly 33% of the activity of the 3A4 isoform at this
concentration. Use of CYP inhibitors show ITMN 191 to be metabolized by
multiple CYP isoforms. After 120 min incubation with rat, dog, human, or monkey
hepatocytes, 81%, 73%, 35%, or 19% of input compound remained intact, respectively.
After a regimen in which a human equivalent dose of 290 mg BID PO was
administered for 7 days, compound trough liver levels in rats and
cynomolgus monkeys were 390-fold and 52-fold the EC90 value,
respectively, after the last dose in the study. Plasma exposure was linear with
dose in both species to at least 300 mg/kg.
In
conclusion, ITMN 191 is a highly potent, orally absorbed inhibitor of the
NS3/4A protease found in the liver of rats and cynomolgus monkeys at levels
predictive of human efficacy. Based on these data, ITMN 191 has been nominated
for preclinical development and is currently undergoing IND-enabling
toxicologic assessment.
K. R. Condroski; H. Zhang; S.
Seiwert; J. A. Ballard; B. A. Bernat; B. Brandhuber; H. Colwell; D. Smith; G.
Vigers; S. W. Andrews; A. L. Kennedy; Y. Jiang; S. M. Wenglowsky; J. A. Josey;
L. M. Blatt
A
need exists for novel treatments of chronic hepatitis C since the current
standard of care, PEG IFN-α-2 + ribavirin, results in a sustained
virologic response in ∼50% of patients. Recent advances in the field have
identified the HCV serine protease NS3/4A as a promising target for drug
discovery. Developing potent inhibitors of NS3/4A has been challenging from a
medicinal chemistry standpoint due to its relatively flat and largely
featureless binding site. Additionally, rapid error-prone replication leading
to multiple active genotypes and quasi-species has slowed the drug discovery
process for HCV NS3/4A inhibitors. In our effort to design potent inhibitors of
HCV NS3/4A protease we used an iterative structure-based design approach
powered by protein crystallography that guided the rational design of analogs
leading to the effective optimization of the P2 moiety, the linker, the P1’
moiety and contacts to the NS3/4A catalytic triad. By gaining potency from
optimized contacts in each region of the target, it was possible to achieve
exceptionally potent NS3/4A inhibitors that maintain potency across genotypes
and mutant strains. These efforts resulted in identification of ITMN 191, a
potent, selective HCV NS3/4A protease inhibitor. Analysis of a crystal
structure of an ITMN 191–NS3/4A protease complex led to an understanding of the
structural factors responsible for the subnanomolar potency of ITMN 191 in
NS3/4A of genotype 1 and its significant potency against NS3/4A derived from
HCV genotypes 2 and 3. In addition, these studies inform the differences in
activity observed for prototypical peptide-based protease inhibitors such as
VX-950 and SCH 503034. In conclusion, these studies rationalize the potency
characteristics of the preclinical candidate ITMN 191.
R. Rieger; P. A. Lee; S. Seiwert;
S. W. Andrews; G. P. Hingorani; T. K. Pope; J. Pheneger; N. Neitzel; R. B.
Franklin; J. A. Josey; L. M. Blatt
Inhibition of the HCV serine
protease, NS3/4A, represents a promising new strategy for control of chronic
HCV infection. Given that viral replication of HCV is reported to occur
primarily in hepatocytes, achieving high drug concentrations in liver is
believed to be critical to the clinical success of protease inhibitors. In
addition, clinical trials with peptide-based NS3/4A protease inhibitors have
demonstrated a positive correlation between the magnitude of antiviral effect
and the trough concentration of drug in plasma, which is thought to be a
surrogate for trough concentrations of drug in liver. With the goal of
identifying potent NS3/4A protease inhibitors providing high liver
concentrations, particularly at trough, we analyzed a series of macrocyclic
peptidomimetic inhibitors for plasma pharmacokinetic (PK) parameters and tissue
concentrations (liver and heart) following a single oral dose of 30 mg/kg in
Sprague-Dawley rats. These data demonstrated that modest structural changes
resulted in radically different compound ratios between liver, plasma, and
heart. ITMN 191 provided an average liver Cmax of 3,098-fold the replicon EC90
value and much lower exposures in plasma (10-fold less) and heart (50-fold
less). ITMN 191 concentrations in plasma and liver were further examined in
multidose PK studies in rats, dogs, and monkeys. Following a 7-day PO BID
dosing regimen of 30 mg/kg in rats, average concentrations of compound in liver
at apparent Cmax and trough were 1,646-fold and 152-fold the EC90 value,
respectively, after the last dose on Day 7. A liver-to-plasma ratio of 2.4:1
was observed at apparent Cmax. Following a 7-day PO BID dosing regimen of 15
mg/kg dose in beagle dogs, average concentrations of compound in liver at
apparent Cmax and trough were 3,048-fold and 58-fold the EC90 value,
respectively, after the last dose on Day 7. A liver-to-plasma ratio of 51:1 was
observed at apparent Cmax. Lastly, following a 7-day PO BID dosing regimen of
15 mg/kg dose in cynomolgus monkeys, average concentrations of compound in
liver at apparent Cmax and trough were 238-fold and 52-fold the EC90 value,
respectively, after the last dose on Day 7. A liver-to-plasma ratio of 70:1 was
observed at apparent Cmax. In conclusion, the high Cmax and trough liver
concentrations of ITMN 191 across three preclinical species contributed to the
decision to nominate it for further development.
S. Brand; J. Dambacher; F.
Beigel; M. Storr; T. Olszak; K. Zitzmann; B. Goeke; C. J. Auernhammer; A. Kaul;
R. Bartenschlager; H. Diepolder
Background:
IL-28A and IL-29 belong to a novel
group of interferons named IFN-lambdas. Recently, we demonstrated antiviral
properties for these cytokines in intestinal epithelial cells (Am J Physiol
Gastrointest Liver Physiol 2005; 289:G960-8).
Aims:
The aim of this study was to analyze
receptor expression, signal transduction and antiviral functions, particularly
against hepatitis C virus (HCV) infection, mediated by IL-28A and IL-29 in hepatic
cells.
Methods:
The mRNA expression of IL-10R2 and
IL-28R1, the receptor subunits required for IL-28A and IL-29 signaling, was
determined by RT-PCR in hepatic cell lines. Lambda-interferon induced signal
transduction was analyzed by Western blotting using phospho-specific antibodies
against STAT proteins. The activation of the antiviral proteins was analyzed by
luciferase assays and Northern Blot analysis. The effect of IL-28A and IL-29 on
replication of HCV was analyzed in Huh-7 cells stably expressing HCV replicons
and firefly genes using luciferase assays. The IL-28A mRNA expression in HCV
infected human liver biopsy samples (n=10) and control liver biopsy samples
(n=19) was measured by quantitative PCR. To analyze regulation of IL-28 mRNA
expression in viral infection in vivo, C57/BL6 mice were infected i.v. with one
million pfu murine cytomegaly virus (MCMV) of the Smith strain for 45 hours.
Murine IL-28 mRNA expression in liver tissue was measured by RT-PCR.
Results:
Both receptor subunits necessary for
IFN-lambda signaling (IL-10R2 and IL-28R1) are expressed in the hepatic cell
lines HepG2, Hep3B and Huh-7. The IL-28R receptor complex is functional
resulting in increased phosphorylation of STAT-1 and STAT-3 proteins following
IL-28A and IL-29 stimulation. Moreover, both cytokines activate the
transcription of the antiviral proteins MxA and 2’,5’-OAS. 100 ng/ml of IL-28A
and IL-29 decreased HCV replication in Huh-7 cells to 1.76 ± 0.08% and 1.38 ±
0.20%, respectively (control 100%, IFN-alpha 10 U: 0.31 ± 0.14%). IL-28A mRNA
expression was significantly higher in liver biopsy samples of HCV patients
than in samples of patients with non-viral hepatitis (36.7-fold increase;
p<0.05). Similarly, murine IL-28 mRNA expression was increased 3-fold in
liver tissue from MCMV infected mice compared to uninfected control mice
(p=0.003).
Conclusion:
Hepatic IL-28 mRNA expression is
elevated in human HCV and murine CMV infection and IFN-lambda mediated
signaling inhibits HCV replication in vitro. These results suggest that
IFN-lambda could play a role in the antiviral immune defense against HCV and
may have a therapeutic potential.
D. Dugourd; R. W. Siu; J. R. Fenn
Celgosivir
is an alpha glucosidase inhibitor that is being developed for the treatment of
Hepatitis C virus (HCV) infections in humans. The purpose of this study was to
evaluate the in vitro antiviral activity of celgosivir and its major
metabolite, castanospermine, when combined with current approved therapies
(ribavirin, interferon α2b, or both) in a surrogate model of HCV (Bovine
Viral Diarrhea Virus (BVDV)). Compounds alone or in combination were tested
against BVDV in infected Madin-Darby Bovine Kidney (MDBK) cells. Synergies were
analyzed using isobolograms and volume of synergy measurements (MacSynergy II™
software). The celgosivir-interferon α2b combination was significantly
more synergistic than the celgosivir-ribavirin combination (~3-fold), or the
ribavirin-interferon α2b combination (~2-fold). Similarly, the
castanospermine-interferon α2b double combination was more synergistic
than the castanospermine-ribavirin combination (~5-fold), or the
ribavirin-interferon α2b combination (~3.3-fold). The combinations of
celgosivir-interferon α2b or castanospermine-interferon α2b led to
significant decreases in the EC50s of celgosivir (up to >20-fold) and
castanospermine (up to >50 fold). The effective EC50s of celgosivir or
castanospermine were further reduced by the addition of ribavirin. The
cytotoxicity of the double and triple combinations was additive or less than
additive, indicating that combinations of celgosivir or castanospermine with
ribavirin and/or interferon α2b were generally less toxic than expected.
These results indicate that the combination of celgosivir with interferon
α2b or with interferon α2b and ribavirin may be effective in the
treatment of HCV.
S. Abe; R. Narita; T. Oto; A.
Tabaru; M. Otsuki
Background and Aims:
It
is well-known that the patients with HCV genotype 2a/2b have higher sustained
virological response than those with genotype 1b after interferon (IFN)
treatment in chronic hepatitis C (CHC). However, it is not clarified whether
there is a difference in activation of T lymphocytes among HCV genotypes after
IFN treatment. We evaluated the relationship between HCV genotypes and the host
immune response by examining the levels of serum soluble interleukin-2 receptor
(sIL-2R) that reflects activation of T lymphocytes.
Methods:
One
hundred and eleven CHC patients were included. Forty patients received IFN
monotherapy (group A): 6-10 million units (MU) of IFN-α2b daily for 14
days followed by 3 times per week (tiw) for a total of 24 weeks, whereas 71
patients received the combination therapy with IFN and ribavirin (RBV) (group
B): 6-10 MU of IFN-α2b daily for 14 days followed by tiw for a total of 24
weeks and 600 or 800 mg RBV per day. We measured serum sIL-2R levels in those
patients before (T0) and 2 weeks (T2) after the treatment.
Results:
The
sustained response rates in genotype 2a/2b patients were significantly higher
than those in genotype 1b patients both in group A (77.8% vs 38.5%, P = 0.0146)
and B (88.9% vs 25.0%, P < 0.0001). In IFN-sustained responders, sIL-2R
levels at T2 were significantly higher than those at T0 both in group A and B
(P = 0.0049 and P = 0.0002, respectively). In IFN-nonresponders, sIL-2R levels
at T2 were not different from those at T0 (P = 0.0555) in group A, but were
significantly higher than those at T0 (P = 0.0143) in group B. In genotype 1b
patients, sIL-2R levels at T2 were not different from those at T0 (P = 0.1520)
in group A, but were significantly higher than those at T0 (P = 0.0274) in
group B. In genotype 2a/2b patients, sIL-2R levels at T2 were significantly
higher than those at T0 both in group A and B (P = 0.0025 and P < 0.0001,
respectively).
Conclusion:
These
findings suggest that the activation of T lymphocytes after IFN treatment
contributes to high sustained response rates, especially in patients with HCV
genotype 2a/2b.
E. Rubinchik; D. J. Erfle; J. J.
Clement; C. J. Pasetka
Celgosivir is a novel antiviral agent
currently in clinical development for the treatment of chronic hepatitis C
virus (HCV) infection. Non-clinical and clinical studies indicate that,
following oral administration, celgosivir (6-O-butanoyl castanospermine) is
readily converted to its primary metabolite castanospermine. Both celgosivir
and castanospermine inhibit alpha-glucosidase-I, an enzyme that is involved in
the processing of viral glycoproteins. In peripheral blood, celgosivir is
detected only sporadically for a short period after dosing while
castanospermine is present at significantly higher levels for a much longer
period. However, the extent of liver exposure to celgosivir has not been
conclusively established as the low peripheral blood levels may be the result
of first-pass metabolism. The objective of this study was to determine
celgosivir and castanospermine levels in the portal and caudal venous blood
during the first hour following oral celgosivir dosing.
Five male Sprague-Dawley rats with
portal vein catheters were administered celgosivir orally at 400 mg/kg. Blood
samples were collected from the portal and caudal veins at pre-determined
time-points. The heparinised blood was immediately extracted with an acetone/methanol
mixture. Celgosivir and castanospermine concentrations were determined using
high performance liquid chromatography coupled with mass spectrometric
detection.
Following oral dosing, celgosivir was
detected in both portal and caudal venous blood. Celgosivir concentrations in
the portal vein were on average 4.4-13.2 times higher than those in peripheral
circulation. The average maximum concentration (Cmax) of celgosivir in the
portal vein was 1,887 ng/mL with an area under the curve (AUC0-60 min)
of 1,003 ng.hr/mL. Comparatively, the portal vein Cmax of castanospermine was
105,630 ng/mL with an AUC0-60 min equivalent to 63,876 ng.hr/mL. The
average celgosivir concentration in portal blood was up to 3.4% of the total
amount of both analytes.
Based on the results described, it
was concluded that following oral celgosivir dosing the liver was exposed to
both celgosivir and castanospermine, with the majority of drug being converted
to castanospermine prior to liver exposure. This conversion most likely
occurred in the gastrointestinal tract.
D. Jensen; A. Di Bisceglie; N.
Gitlin; B. Freilich; K. Reddy; V. Feinman; S. Arora; P. Marcellin
The REPEAT study is currently
investigating the efficacy/safety of Peg-IFNα-2a (40KD) plus RBV in
previous non-responders to Peg-IFNα-2b (12KD)/ribavirin. Here, we report
results of a protocol-planned interim efficacy/safety analysis after 12 weeks
of re-treatment.
Methods:
Non-responders to ≥12 weeks of
approved doses of Peg-IFNα-2b (12KD)/ribavirin who remained HCV RNA
positive throughout treatment were eligible. 950 patients were randomized to 1
of 4 groups: Peg-IFNα-2a (40KD) 360 μg/week for 12 weeks then 180
μg/week for a further 60 or 36 weeks (Arms A and B, respectively);
Peg-IFNα-2a (40KD) 180 μg/week for 72 or 48 weeks (Arms C and D,
respectively). All 942 treated patients received RBV 1000/1200 mg/d. HCV RNA
was measured by quantitative (≥600 IU/mL) or qualitative (≥50
IU/mL) PCR assay. Data from Arms A+B (Peg-IFNα-2a [40KD] fixed-dose
induction), and Arms C+D (Peg-IFNα-2a [40KD] standard dose) were combined.
Results
(Table):
Baseline characteristics were similar
in the 2 groups. At week 12, a significantly greater proportion of patients
treated with fixed-dose induction Peg-IFNα-2a (40KD) had an early
virological response (EVR; HCV undetectable, HCV non-quantifiable, or ≥2-log10
drop in HCV RNA), compared with standard-dose Peg-IFNα-2a (40KD). No
substantial increase was seen in the incidences of adverse events (AEs) and
serious AEs in patients treated with fixed-dose induction vs standard-dose
Peg-IFNα-2a (40KD).
Conclusion:
Over the first 12 weeks of
retreatment, standard doses (180 μg/week) of Peg-IFNα-2a (40KD) plus
RBV are associated with an EVR rate of 45% in previous non-responders to
Peg-IFNα-2b (12KD)/ribavirin. A fixed induction dose (360 μg/week) of
Peg-IFNα-2a (40KD) plus RBV was found to be even more effective than
standard doses at inducing an EVR (62%); moreover, the safety profile of
Peg-IFNα-2a (40KD) plus RBV did not appear to be compromised in patients
receiving fixed induction doses.
|
N (%)†
|
Peg-IFNα-2a
(40KD) dose during weeks 1−12 (+RBV
1000/1200mg/d) |
|
|
180
µg/week (n=469) |
360
µg/week (n=473) |
|
Male |
319 (68) |
297 (63) |
|
Mean age ± SD, yrs |
48.8 ± 8.8 |
48.3 ± 9.1 |
|
Caucasian/Black |
413 (88)/42 (9) |
418 (88)/39 (8) |
|
Genotype 1 / 2 or 3 |
425 (91) / 16 (3) |
429 (91) / 17 (4) |
|
Mean HCV RNA ± SD, x106 IU/mL |
4.9 ± 5.7 |
5.4 ± 6.6 |
|
Cirrhosis/advanced fibrosis |
133 (28) |
119 (25) |
|
Median duration of previous treatment,
weeks |
28 |
28 |
|
EVR‡ at week 12 |
210 (45) |
291 (62)* |
|
Peg-IFNα-2a (40KD) dose modified or
discontinued |
63 (13) |
88 (19) |
|
≥1 AE/serious AE |
430 (92)/19 (4) |
441 (93)/10 (2) |
Abstract
T1801 – Impact of Therapy of Chronic Hepatitis C (CHC) on Quality of Marital
Relationships.
M.
O'Brien; L. S. Rosenthal; E. Lebovics
BACKGROUND:
Impairment
of quality of life by therapy of CHC has been documented in studies utilizing validated
social science questionnaires. Specific impact on close valued relationships
such as marital or significant others has not been examined.
METHODS:
Questionnaires
were sent to 445 consecutive patients treated with either interferon-alpha 2b
or pegylated interferon plus ribavirin. Questions addressed communication
within the relationship, sexual intimacy, ability to share household
responsibilities, quality of shared leisure time, arguing, and overall quality
of the marriage prior to treatment initiation, during treatment, and after
treatment completion. Responses were scored on a scale of 1 to 5, with 5 being
the highest functioning. Also, patients were asked whether they separated or
divorced during or shortly after their treatment and if this was attributable
to therapy.
RESULTS:
Of
114 respondents, 15 patients (13.1%) were either separated (n=10; 8.8%), or
divorced (n=5; 4.4%) during or shortly after treatment. Six of the 10 who
separated and 2 of the 5 who divorced stated this was in part secondary to CHC
therapy. Mean scores for all parameters assessing aspects of the marital
relationship significantly decreased from baseline to during therapy
(p<0.001 for each) and returned to baseline after therapy. 35% of patients
had no impairment in communication, 49% had a drop of 1-2 points, and 12% a
drop of 3-4 points. 30% reported no change in sexual intimacy, 50% had a drop
of 1-2 points, and 19% a drop of 3-4 points. 27% reported no change in sharing
household responsibilities, 48% had a drop of 1-2 points, and 23% a drop of 3-4
points. 29% reported no change in quality leisure time spent with spouse, 32%
had a drop of 1-2 points, and 16% a drop of 3-4 points. 49% reported no change
in frequency of tense arguments with significant other, 37% had a drop of 1-2
points, and 7% a drop of 3-4 points. 53% reported no impairment in overall
quality of marriage, 37% had a drop of 1-2 points, and 7% a drop of 3-4 points.
CONCLUSIONS:
· CHC therapy impairs marital relations and may
contribute to separation and divorce.
· Patients should be counseled prior to initiation of
therapy of these potentially life altering effects, and appropriate care taken
to prevent such outcomes.
· Future studies should address whether these results
differ from a population matched for psychosocial conditions.
P.
Kwo; I. Jacobson; R. Brown; B. Freilich; B. Afdhal; J. Santoro; S. Becker; A.
Wakil; D. Pound; E. Godofsky; R. Strauss; D. Bernstein; S. Flamm; N. Bala; V.
Arraya; M. Davis; H. Monsour; J. Vierling; F. Regenstein; V. Balan; M.
Dragutsky; M. Epstein; R. Herring; L. H. Griffel; C. Brass
Background:
Registration trials using PEG IFN and ribavirin,
conducted by investigators experienced in the treatment of CHC have
demonstrated SVR rates of 54-56%. Whether these SVR rates can be replicated in
clinical practice in the community remains unclear. Aim: To determine
investigator factors that affect SVR rates in patients receiving PEG IFN
alfa-2b 1.5 ug/kg/week with ribavirin for HCV treatment.
Methods:
Patients in academic or community settings with CHC
were randomized to PEG IFN alfa-2b 1.5 ug/kg once weekly plus RBV 800 mg/day or
RBV based on weight. Treatment was 48 weeks for patients with HCV G1, patients
with G2 or 3 were randomized to 24 or 48 weeks. Follow-up for all patients was
24 weeks. HCV RNA was determined by PCR. 25 academic and community investigators,
all having participated in CHC registration trials or having extensive
experience in HCV therapy served as regional PIs . High enrollment sites (≥
25 patients enrolled) had selected data verified by study monitor. Analyses
performed by logistic regression analysis. Results: 225 sites enrolled 4913
patients with overall SVR of 43.6%., 923/4913 (18.8%) were treated at regional
PI sites, 3990/4913 were at non-regional PI sites(81.2%). 3114/4913 patients
were treated at monitored sites (63.4%). 1800/4913 patients (36.6%) dropped out
(DO).
Conclusions:
1. Patients within regional PI sites are more like to
achieve SVR and less likely to drop out than patients within non-regional PI
sites
2. Within non-monitored sites (<25 patients
enrolled), patients within regional PI sites are more like to achieve SVR and
less likely to drop out than patients within non-regional PI sites
3. No differences were seen in SVR or DO for monitored
and non-monitored sites.
Research Support: Schering Plough
|
SVR all
patients |
SVR
regional PI patients |
SVR
non-regional PI patients |
OR |
p-value
|
|
2140/4913(43.56%) |
445/923 (48.21%) |
1695/3990 (42.48%) |
1.261 |
0.0016 |
|
SVR all patients |
SVR monitored sites |
SVR non-monitored sites |
|
|
|
2140/4913 (43.56%) |
1372/3114 (44.06%) |
768/1799 (42.69%) |
1.057 |
0.3518 |
|
SVR all monitored patients |
monitored regional SVR |
monitored non-regional SVR |
|
|
|
1372/3144 (43.64%)) |
351/763 (45.88%) |
1021/2351 (43.43%) |
1.11 |
0.2121 |
|
SVR all non-monitored patients |
SVR regional non-monitored patients |
SVR non-regional non-monitored patients |
|
|
|
768/1799 (42.69%) |
94/160 (58.75%) |
674/1639 (41.12%) |
2.039 |
<0.0001 |
|
DO all patients |
DO regional PI patients |
DO non-regional PI patients |
|
|
|
1800/4884(36.86%) |
312/921 (33.88%) |
1488/3963(37.55%) |
0.852 |
0.0376 |
|
DO all patients |
DO monitored sites |
DO non-monitored sites |
|