at least once after the ﬁrst injection of study drug. The

per protocol population was deﬁned as all mITT subjects

in whom penile curvature was measured at week 36 and

who had PD-PRO responses at baseline and week 36,

and no protocol violations that would have affected the

efﬁcacy evaluation. A total of 11 patients discontinued

the study early due to withdrawal of consent (6), loss to

followup (1), AE (2) and protocol violation (2). Efﬁcacy

analysis was done using last observation carried for-

ward.

Subject demographics, baseline measurements, change

from baseline penile measurements, and PD-PRO and

IIEF responses were summarized with descriptive statis-

tics. Comparisons between treatment groups were based

on the change from baseline using ANOVA with ﬁxed

factors for drug (placebo or CCh), modeling (with/without)

and drug by modeling interaction. The linear relationship

between penile curvature measurements and PD-PRO re-

sponses was evaluated using the Pearson correlation co-

efﬁcient. The Fisher exact test was used to compare CCh

to placebo with and without modeling for the AE inci-

dence. All reported p values are 2 sided and were not

adjusted for multiple comparisons. Statistical analysis

was done with SAS®, version 9.1.

RESULTS

Study Population

A total of 147 subjects were randomized into 4

groups, including placebo with and without model-

ing, and CCh with and without modeling (

ﬁg. 1

). No

statistically signiﬁcant differences were observed

among the 4 treatment groups for any demographic

or baseline characteristic. Most subjects were be-

tween ages 45 and 64 years, and 78% and 54% did

not report a speciﬁc event of penile trauma and pain,

respectively (

table 1

). ED was self-reported by 44%

of subjects at baseline (

table 2

). Mean

SD PD

history was 3.0

2.7 (range 0.5 to 14.8) vs 2.1

1.3

years (range 0.6 to 6.1) in patients on CCh vs pla-

cebo with modeling and 3.0

3.5 (range 0.5 to 19.6)

vs 2.2

2.6 years (range 0.6 to 11.1) in those on

CCH vs placebo without modeling. No patient had a

ventral penile malformation.

mITT population

CCh (n=54)

Modeling (n=74)

Non-modeling (n=73)

Randomized subjects

(n=147)

Placebo (n=20)

CCh (n=57)

Placebo (n=16)

Allocation

Analysis

Randomization

CCh (n=54)

Placebo (n=20)

Placebo (n=16)

CCh (n=55)

Excluded from

analysis

(No

post-baseline

measurements

taken, n=2)

Figure 1.

Study ﬂow diagram

Table 1.

ITT population demographic and baseline characteristics

Modeling

No Modeling

CCh

Placebo

CCh

Placebo

No. pts

Mean

SD age (range)

57.4

7.4 (36–72)

56.6

6.4 (38–68)

56.6

8.2 (28–71)

53.9

7.5 (36–61)

No. age (%):

Less than 45

(5.6)

(5.0)

(8.8)

(12.5)

45–64

(77.8)

(85.0)

(78.9)

(87.5)

65 or Greater

(16.7)

(10.0)

(12.3)

Mean

SD degrees penile curvature (range)

54.7

15.2 (33–89)

51.9

15.9 (30–90)

54.1

15.1 (30–90)

48.9

14.3 (30–80)

Mean

SD wt (kg)

85.4

10.4

84.9

12.2

86.5

10.4

88.7

15.0

Mean

SD ht (cm)

178.6

7.8

175.3

7.6

179.1

6.8

179.1

7.4

No. Hispanic/Latino ethnicity (%):

Yes

(5.0)

(3.5)

(6.3)

(100.0)

(95.0)

(96.5)

(93.8)

No. race (%):

Black

(5.6)

(5.3)

(6.3)

White

(94.4)

(100.0)

(94.7)

(93.8)

No. alcohol use (%):

(11.1)

(25.0)

(22.8)

(25.0)

Yes

(85.2)

(60.0)

(68.4)

(68.8)

Previously

(3.7)

(15.0)

(8.8)

(6.3)

No. tobacco use (%):

(51.9)

(50.0)

(52.6)

(50.0)

Yes

(9.3)

(15.0)

(15.8)

(12.5)

Previously

(38.9)

(35.0)

(31.6)

(37.5)

PHASE 2B STUDY OF COLLAGENASE CLOSTRIDIUM HISTOLYTICUM

2270