Impact of C-Peptide Preservation on Metabolic and Clinical ......2013/09/23 · 09/03/13 Impact of C-Peptide Preservation on Metabolic and Clinical Outcomes in the Diabetes Control

09/03/13

Impact of C-Peptide Preservation on Metabolic and Clinical Outcomes

in the Diabetes Control and Complications Trial

John M. Lachin ScD1, Paula McGee MS

1, Jerry P. Palmer MD

2 for the DCCT/EDIC

Research Group*

1The Biostatistics Center, The George Washington University, Rockville Maryland

2Department of Medicine, University of Washington, Seattle, Washington; and the

Department of Medicine, DVA Puget Sound Health Care System, Seattle, Washington

*A complete list of the members of the DCCT/EDIC Research Group is provided in the

supplementary appendix published in New England Journal of Medicine, 2011;

365:2366-76.

For submission to Diabetes.

Words in Abstract: 198 < 200 max

Words in text: 3977 < 4000 max

References:

Tables: 3

Figures 3 (8 panels)

Page 1 of 47 Diabetes

Diabetes Publish Ahead of Print, published online October 2, 2013

2

Abstract

The DCCT established that a stimulated C-peptide concentration ≥0.2 nmol/L at

study entry among subjects with up to 5 years diabetes duration is associated with

favorable metabolic and clinical outcomes over the subsequent 7 years of follow-up.

Herein we further examine the association of both fasting and stimulated C-peptide

numerical values with outcomes. In the intensive treatment group, for a 50% higher

stimulated C-peptide on entry, such as from 0.10 to 0.15 nmol/L, HbA1c decreased by

0.07 HbA1c% (0.8 mmol/mol, p=0.0003); insulin dose decreased by 0.0276 U/kg/d

(p<0.0001); hypoglycemia risk was decreased by 8.2% (p<0.0001); and the risk of

sustained retinopathy was reduced by 25% (p=0.0010), all in unadjusted analyses. Other

than HbA1c, these effects remained significant after adjusting for the HbA1c on entry.

While C-peptide was not significantly associated with the incidence of nephropathy, it

was strongly associated with the albumin excretion rate. The fasting C-peptide had

weaker associations with outcomes.

As C-peptide decreased to non-measureable concentrations, the outcomes

changed in a nearly linear manner, with no threshold or breakpoint. While preservation of

stimulated C-peptide at ≥0.2 nmol/L has clinically beneficial outcomes, so also does an

increase in the concentration of C-peptide across the range of values.

Page 2 of 47Diabetes

3

The concentration of C-peptide in response to a stimulus (mixed meal or glucagon

injection) has been established as a valid and reliable measure of residual beta cell

function (1, 2). C-peptide is co-secreted with insulin in an equimolar ratio and has a much

longer half-life.

In the Diabetes Control and Complications Trial (3), at study entry, the fasting

and stimulated C-peptide values were correlated (r = 0.83); values among adolescents

were significantly less than those among adults; and values for those with 1-5 y duration

were significantly greater than those with 5-15 y duration. Among subjects with 1-5 y

duration, over an average 6.5 years of follow-up, C-peptide responders with stimulated

C-peptide values ≥ 0.2 nmol/L had significantly lower values of HbA1c and blood

glucose and were receiving lower doses of insulin than the non-responders with C-

peptide<0.2 nmol/L (3).

During the DCCT, subjects assigned to intensive therapy had a 57% lower risk of

their C-peptide response falling to < 0.2 nmol/L at a subsequent annual assessment

compared to those receiving conventional treatment (4). Within the intensive group, the

C-peptide responders had lower risks of retinopathy progression and of severe

hypoglycemia.

Also within the intensive treatment group, the mean level of HbA1c over the

average of 6.5 years of DCCT follow-up was significantly lower by 0.4% among the C-

peptide responders than non-responders (7.0 vs. 7.4%, 53 vs. 57 mmol/mol, p < 0.0001).

In addition, there was a substantial reduction in the risk of nephropathy, but the effect

was not significant, likely due to the small number of events in this subset (N=49). In a

separate analysis, Steffes et al. (5) found that entry concentrations of C-peptide as low as


4

0.04 nmol/L were significantly related to a reduced risk of both retinopathy and

nephropathy.

In this report we describe the association of the concentrations of C-peptide at

study entry, as well as the prior classification of responders and non responders, with

DCCT outcomes that include HbA1c, insulin dose, risk of progression of retinopathy,

nephropathy and neuropathy, and the incidence of hypoglycemia. For each we also

evaluate whether there is a generally increasing (or decreasing) association over the entire

range of C-peptide values versus a break-point.

Methods

C-peptide was measured at the initial eligibility screening visit (3). Specimens

were collected fasting (basal) and at 90 minutes (stimulated) following ingestion of a

liquid mixed meal, Sustacal®. C-peptide was measured centrally by radioimmunoassay

with the Novo M1230 antibody with a lower limit of detection of 0.03 nmol/L (3, 4).

Subjects with 1-5 years duration (n=855) were eligible for entry into the DCCT

with stimulated C-peptide <= 0.5 nmol/L, most (n = 657) of whom had measurable

concentrations. Those with 5 to 15 years duration (n = 586) were required to have values

≤ 0.2 nmol/L, most (n = 436) of whom had non-measurable concentrations (4). Eligible

subjects 13-39 years of age were randomly assigned to receive intensive versus

conventional therapy. The primary prevention cohort included 726 subjects with 1-5

years duration, no pre-existing retinopathy and normal albuminuria. The secondary

intervention cohort included 715 subjects with 1-15 years duration, pre-existing mild

retinopathy, and albumin excretion rate (AER) up to 200 mg/24 h (6).


5

Since the majority of subjects with diabetes duration ≥ 5 years had non-

measurable C-peptide at entry, analyses herein are based on the 855 subjects with 1 – 5 y

duration and stimulated C-peptide ≤ 0.5 nmol/L at DCCT entry from both the primary

(713) and secondary (142) cohorts. Of these, 303 subjects were classified as C-peptide

responders based on stimulated values between 0.2 – 0.5 nmol/L (138 intensive, 165

conventional), and the remaining 552 subjects were classified as non-responders with

stimulated values < 0.2 nmol/L (274 intensive, 278 conventional) (4).

Other DCCT methods have been extensively described (6). Briefly, HbA1c was

measured by high pressure liquid chromatography (HPLC) and insulin dose assessed

(units/kg of body weight) at eligibility screening and every three months during follow-

up. Retinopathy was assessed centrally based on seven field fundus photographs obtained

at study entry and every 6 months during follow-up. Timed collection of AER was

measured on entry and annually. The primary retinopathy outcome was three or more

step sustained progression on the final Early Treatment Diabetic Retinopathy Study

(ETDRS) scale of retinopathy severity from the level on entry or pan-retinal laser therapy

for retinopathy. Nephropathy was defined as an AER > 40 mg/24 h (microalbuminuria or

worse). Confirmed clinical neuropathy required the presence of both definite clinical

neuropathy by the neurologist’s examination and an abnormal nerve conduction study

consistent with a distal symmetrical polyneuropathy (7). Severe hypoglycemia was

defined as an episode that required the assistance of another person.

A local smoothing function (LOESS) assessed the linearity of the trend in the

relationship of log C-peptide with other outcomes (8). Normal errors longitudinal

regression models were employed to assess C-peptide effects on quantitative outcomes


6

(HbA1c, insulin dose) over time using an unstructured covariance matrix (9). The log(C-

peptide) improved the linearity of the associations and better satisfied the assumptions of

approximately normally distributed and homoscedastic residuals. Thus, the change in

HbA1c or insulin dose per 50% higher C-peptide on entry, is obtained as log(1.5β), where

β is the coefficient for log(C-peptide) in the model. Owing to small sample sizes within

treatment groups observed at years 8 and 9 of follow-up (a result of staggered patient

entry), longitudinal analyses of quantitative outcomes (e.g. HbA1c) only include data

through year 7, with a mean follow-up of 5.3 years.

The Cox proportional hazards regression model (10) was employed to assess

associations with the risk of an event over time (retinopathy progression, nephropathy),

and its generalization, the multiplicative intensity model, likewise for recurrent

hypoglycemia. A logistic regression model assessed the association with neuropathy at 5

years of follow-up. Here as well the log(C-peptide) improved the fit of the models. Since

the model describes the log(risk) as a function of the log(C-peptide), the percentage

change in risk per a 50% higher C-peptide is computed as 100*(1.5β -1).

The stimulated response cut-point of 0.2 nmol/L was the 68th

percentile of the

distribution of stimulated C-peptide values. The corresponding cut-point for classification

of the fasting C-peptide response was 0.075 nmol/L. There was 83% agreement between

the classification of fasting versus stimulated C-peptide response versus not, with a

chance-corrected kappa index (10) of 0.63 (95% CI 0.55, 0.70) showing good agreement

between the two classifications.

Statistical test results were not adjusted for multiple tests of significance. Nominal

p ≤ 0.05 is employed to declare statistical significance.


7

Results

Unless stated otherwise, all results refer to the stimulated C-peptide value

obtained at DCCT eligibility screening. Patient characteristics within intensive and

conventional treatment groups have been previously described (1, 4). Table 1 shows the

characteristics of the subjects with 1-5 years duration (mean 2.6 y) classified as

responders versus non-responders separately within the two treatment groups. Baseline

characteristics were similar within the intensive and conventional groups on entry. In the

intensive group, the C-peptide responders (n=138) versus non-responders (n=274) tended

to be in the primary prevention cohort and have shorter duration of diabetes, lower

HbA1c, lower insulin dose, and less severe complications.

The table also presents the numbers of principal outcome events. Tests of

treatment group differences are presented in subsequent tables.

Intensive Therapy, 1-5 years Duration

Among those in the intensive treatment group with 1-5 years diabetes duration on

entry (from both the primary and secondary cohorts), model-free LOESS estimates of the

trend of the association of the stimulated log(C-peptide) concentration with HbA1c,

insulin dose and hypoglycemia (Figure 1A, B and D) all showed an approximately linear

association for log(C-peptide) > -2.5 or a C-peptide concentration > 0.082 nmol/L. Below

this value the relationships were flat. Model diagnostics indicated that the log

transformed values still provided better fit than did the untransformed values. There was

a strongly linear association between the log of the risk of retinopathy progression and

the log(C-peptide), affirming the assumption of the proportional hazards model over the


8

range of C-peptide. Thus, the log(C-peptide) was used in all quantitative analyses. The

risk of retinopathy did not show a flattening but continued to decrease below a stimulated

C-peptide of 0.082 nmol/L (Figure 1C).

HbA1c. The relationship of C-peptide, either qualitatively (responders versus

non-responders) or quantitatively, with HbA1c did not vary significantly over time (C-

peptide by time interaction p=0.34 and p=0.13, respectively). Thus, the associations are

described over the 7 years of follow-up from longitudinal regression models.

The mean HbA1c during 7 years of follow-up among C-peptide responders (Table

2, Figure 2A) was 6.9% (52 mmol/mol) versus 7.5% (58 mmol/mol) among non-

responders, an average HbA1c% reduction of 0.6% (6.6 mmol/mol, p < 0.0001). In

addition, HbA1c% also decreased by 0.07% (0.8 mmol/mol) per 50% higher C-peptide (p

= 0.0003) across the range of C-peptide values, such as comparing two subjects with C-

peptide values of 0.1 versus 0.15 or 0.2 versus 0.3: each with a 50% increase. These

observations were obtained from a longitudinal mixed model that was minimally adjusted

only for the year of the follow-up HbA1c assessment. The results were equivalent after

also adjusting for both primary/secondary cohort and duration of diabetes.

Even though the entry HbA1c among responders was lower than that among non-

responders (8.3% versus 9.4%, 67 versus 79 mmol/mol, Table 1), in a further longitudinal

regression model, the average HbA1c over the entire follow-up period remained

significantly lower among responders than non-responders when adjusted for cohort and

diabetes duration, and also when adjusted for the HbA1c on entry (7.1% vs. 7.4%, 51

vs.57 mmol/mol, p = 0.0007), similar to the unadjusted analysis (Table 2). The

association of the entry C-peptide (quantitatively) with the mean HbA1c during follow-


9

up was also statistically significant in the unadjusted analysis and when adjusted for

cohort and duration, with HbA1c decreasing by 0.07 HbA1c% (0.8 mmol/mol) per 50%

higher C-peptide (p = 0.0003). However, the effect on HbA1c was diminished and no

longer significant after also adjusting for the baseline HbA1c (p = 0.09) (Table 2). Thus,

some of the association of entry C-peptide with the mean HbA1c during follow-up is

explained by the effect of the C-peptide on the entry HbA1c.

Insulin Dose. At DCCT entry, the intensive treatment group had a mean insulin

dose of 0.50 Units per kg body weight per day (U/kg/d) among responders versus 0.70

U/kg/d among non-responders (p < 0.0001) (4). Thus, the lower HbA1c among C-peptide

responders versus non-responders in the intensive treatment group was achieved with

significantly less exogenous insulin. The minimally-adjusted mean insulin dose during

follow-up among C-peptide responders versus non-responders (Table 2) was 0.63 versus

0.78 U/kg/d, an average reduction of 0.15 U/kg/d (p < 0.0001). Insulin dose also

decreased by 0.0276 U/kg/d per 50% higher C-peptide (p < 0.0001). Results were similar

after also adjusting for cohort and duration, and also after adjusting for the entry HbA1c,

again indicating that the association of entry C-peptide with insulin dose was not

explained by the association between entry C-peptide and HbA1c. Further, when adjusted

for the entry insulin dose, the difference between responders and non-responders

remained significantly different (p = 0.04), whereas the relationship with quantitative C-

peptide was no longer significant (Table 2). Neither the qualitative nor quantitative C-

peptide association with insulin dose was significant when adjusted for both insulin dose

and HbA1c on entry, indicating that some of this C-peptide effect was mediated by its

association with these other values on entry.


10

Further, the association of C-peptide, both qualitatively and quantitatively, with

insulin dose varied significantly over time (i.e. a C-peptide by time interaction). Insulin

dose among responders rose over time, secondary to the further decline in C-peptide

during follow-up as previously shown (4) (Figure 3A). The decrease in insulin dose per

50% increase in C-peptide also declined over time (Figure 3B).

Meal and basal mean insulin doses were lower among C-peptide responders than

non responders (0.37 vs 0.45 and 0.34 vs 0.43 U/kg/d respectively, both p <0.0001), and

the slope of the association with the quantitative C-peptide concentrations was similar

(decrease in insulin dose of 0.0148 and 0.0156 U/kg/d per 50% higher C-peptide, both p

<0.0001), each in minimally adjusted models.

Retinopathy. Table 3 presents the association between the quantitative entry

concentration of C-peptide in the intensive treatment group with the risk of retinopathy

progression during follow-up. The Table also shows the results of the previously

published analyses (1) comparing responders versus non-responders. The mean C-peptide

among responders (0.317 nmol/L) was approximately 4 fold greater than that among non-

responders (0.079 nmol/L). This translated into significant reductions in the risks of

retinopathy by 58% and sustained retinopathy by 79% with no adjustments for other

factors. The risk reductions were diminished slightly after adjustment for the presence or

absence of retinopathy on entry and the entry level of HbA1c, and the difference in

progression (but not sustained progression) remained statistically significant.

When examined as a quantitative effect, the patterns were similar. For a 50%

higher C-peptide on entry (e.g. 0.45 versus 0.30 nmol/L) the risk of retinopathy was


11

significantly reduced by 12% and that of sustained retinopathy by 25%. Adjustment for

retinopathy status and HbA1c on entry had negligible effect on these results.

The appendix describes how to compute the estimated risk reduction associated

with other percentage differences in stimulated C-peptide on entry.

Nephropathy. Only 49 cases of nephropathy (microalbuminuria or worse) were

observed. Even though the risk was reduced by 43% among responders versus non-

responders, the difference was not significant. The risk reductions for nephropathy using

quantitative C-peptide are smaller than those for retinopathy (and non-significant), and

were reduced further after adjustment for factors on entry.

Neuropathy. Only 20 cases of neuropathy were observed at 5 years of follow-up.

Entry C-peptide responders had a non-significant 65% lower odds of confirmed clinical

neuropathy that remained unchanged after adjustment for factors on entry.

Hypoglycemia. Responders had a 45% lower risk of an episode of severe

hypoglycemia than did non-responders, and this risk reduction was increased after

adjusting for other factors. Per 50% increase in C-peptide the risk of hypoglycemia was

decreased by 8.2%, and remained unchanged after adjustment. All associations were

highly significant.

Associations with Fasting C-peptide.

The Appendix presents like analyses using the fasting C-peptide values which

show smaller non-significant differences in HbA1c between the fasting responders and

non-responders and a weaker association with the quantitative fasting value (Table A.2).

A similar analysis showed a slightly smaller difference in insulin dose among fasting C-

peptide responders versus non-responders (0.11 U/kg/d) than in the stimulated


12

comparison above, though still significant (p < 0.0001); and the association with the

quantitative fasting values was nearly identical as that for the stimulated values, each

minimally adjusted. Fasting C-peptide showed smaller non-significant associations with

retinopathy than did stimulated C-peptide, and smaller but significant associations with

sustained retinopathy (Table A.3). Interestingly, fasting C-peptide responders versus non-

responders had a greater reduction in nephropathy risk (146%, p = 0.016) than did

stimulated responders versus non-responders (43%, Table 2).

The associations of fasting C-peptide with neuropathy were similar to those of

stimulated C-peptide and non-significant. While fasting responders had significantly

lowered risk of hypoglycemia than did responders, there was no significant association of

quantitative fasting C-peptide with hypoglycemia risk.

Conventional Therapy, 1-5 years Duration

Like analyses were also conducted among those with 1-5 years duration in the

conventional treatment group using the stimulated C-peptide on entry (see Appendix).

Without any adjustment, the mean HbA1c during follow-up differed little among

responders versus non-responders (9.2% versus 9.3%, p = 0.569), and there was no

association of the quantitative C-peptide concentration with HbA1c (Table A.4). Further,

the association of quantitative and qualitative C-peptide with HbA1c differed

significantly over time (interaction p=0.0006 and p=0.027, respectively). As shown in

Figure 2B, the mean HbA1c among non-responders remained fairly level over time

whereas that among responders increased, as responders on entry lost their stimulated

response.


13

Accordingly, there were no significant differences between stimulated responders

versus non-responders in other outcomes, as shown in Table 1 (Table A.5). Further,

when adjusted for other factors, there were significant associations of the C-peptide

concentration with both HbA1c and insulin dose, but with a positive, not inverse

association. This is in part a reflection of the above interaction (also present with insulin

dose) and the overall weak unadjusted associations.

5 – 15 Years Duration

Among those with 5 to 15 years duration, only those with stimulated C-peptide on

eligibility screening < 0.2 nmol/L (i.e. non-responders) were eligible for enrollment into

the DCCT. Among the 299 such subjects in the intensive group, only 66 subjects had a

measurable stimulated value above the lower limit of quantification (0.03 nmol/L) with a

maximum of 0.2 nmol/L and 25th

, 50th

, and 75th

percentiles of 0.05, 0.07, and 0.11

nmol/L. We compared these 66 subjects to the 233 with non-measurable values (see

Appendix). There were no differences in the HbA1c or insulin dose over time and no

significant differences in risks of retinopathy, sustained retinopathy, nephropathy or

neuropathy. However, there was a significant 22% reduction in the risk of severe

hypoglycemia among those with versus without measurable C-peptide values (p <

0.0001) that was unaltered after adjusting for the other factors.

Quantitative Outcomes

Further analyses were conducted to assess the association of C-peptide with each

of the raw outcome measures: the ordinal ETDRS score of retinopathy severity, the

albumin excretion rate (AER) and the O’Brien mean nerve conduction rank score. The

detailed statistical methods and results are presented in the online appendix.


14

Briefly, the patterns of associations with 3-step progression of retinopathy also

applied to an analysis of associations with the ETDRS scores after 4 years of follow-up.

However, the associations with AER were stronger than those with nephropathy.

Responders had a significantly lower mean AER than did responders by 22% (CI: 12,

34%, p = 0.0001). This difference was diminished but remained significant, even after

adjusting for both the baseline AER and HbA1c (14%, CI: 4, 24, p = 0.003).

Quantitatively as the C-peptide concentration increased, the AER was significantly lower

in unadjusted analyses with a 3% reduction in AER per 50% greater C-peptide (CI: 1, 5;

p = 0.002); and this effect persisted after adjustment for the baseline AER (1.8%, p =

0.03). However, the association was no longer significant when also adjusted for baseline

HbA1c (1.4%, p = 0.11).

The nerve conduction summary measure at 5 years was significantly better among

responders than non-responders in an unadjusted analysis, but not when adjusted for

baseline or HbA1c. There was no association with the quantitative C-peptide

concentration.

Discussion

Preservation of β-cell function as measured by C-peptide in patients with type 1

diabetes is known to result in improved metabolic control and reduced microvascular

complications (1). Prior analyses of DCCT data showed that 0.2 nmol/L could be

considered a meaningful cut point, above which patients demonstrate a strong clinical

benefit. In this report, we expand on this relationship by examining the association of the

quantitative C-peptide concentration with outcomes. Across the range of values, higher

amounts of secreted C-peptide were associated with lower HbA1c, lower daily insulin


15

dose, less severe hypoglycemia, and less risk of retinopathy. Most importantly, even

small incremental increases in C-peptide, particularly those measured below the

previously established cut point of 0.2 nmol/L, are shown to be clinically beneficial.

These results further describe the benefits of specific concentrationsof C-peptide beyond

those previously presented from the DCCT using the crude categories of <0.2 nmol/L

versus (> 0.2 -5) nmol/L (4), or the categories < 0.04, 0.04 to 0.2 and (> 0.2 -5) nmol/L

(5).

Analyses using model free local smoothing techniques showed that there was no

threshold or breakpoint in the relationship of outcomes with C-peptide. Linearity was

improved using the natural log transformation. There was a strong linear association with

risk of retinopathy over the range of log C-peptide values whereas for HbA1c, insulin

dose and hypoglycemia the relationship was flat for values less than approximately 0.08

nmol/L (Fig 1).

Over 7 years of follow-up there was a strong inverse association between higher

entry C-peptide and lower HbA1c that did not wane with time; i.e. the effect of a given

difference in entry C-peptide on the HbA1c values was similar at every year of follow-

up. This association persisted after adjusting for the entry HbA1c that was also inversely

correlated with the entry C-peptide. Thus, the effects of the entry C-peptide on HbA1c

over time were not explained by the association with the entry HbA1c.

The relationship with insulin dose was more varied. As with HbA1c there was a

strong inverse relationship between c-peptide concentrations and insulin dose. While the

difference among responders versus non-responders remained significant after adjusting

for the entry insulin dose, the association with the quantitative C-peptide concentration


16

was no longer significant. Further, the effect of C-peptide on total insulin dose waned

with time. Similar associations were observed for both basal and for meal-related insulin

doses.

The C-peptide concentrations also had an inverse relationship with the risk of

retinopathy progression that persisted after adjustment for the entry HbA1c and

retinopathy status. However, the associations of C-peptide concentrations with the risks

of microalbuminuria and of clinical neuropathy were less than that with retinopathy. As

stated previously (1), this is probably due to the small number of such cases (49 total

subjects for nephropathy and 20 with neuropathy). Further, the association with

nephropathy was reduced after adjustment for the levels of AER and HbA1c on entry,

indicating that some of the relationship of C-peptide with nephropathy was mediated by

the lower values of AER and HbA1c on entry, each of which is a function of these higher

concentrations of C-peptide.

As might be expected, the fasting concentrations of C-peptide had weaker

associations with all outcomes than did the stimulated values. However, the relatively

strong correlations between these measures suggest that, in the absence of beta-cell

stimulation conditions, fasting samples may yield some information on islet function.

Within the conventional group, subjects quickly lost their entry C-peptide

response during follow-up (4), and consequently the entry C-peptide did not have a

strong association with any outcome. However, loss of response was delayed in the

intensive group (4), providing a longer period to benefit from preservation of beta cell

function. Thus, the benefits of C-peptide preservation at baseline were largely observed

in the intensive treatment group.


17

Subjects with 5-15 years duration were required to have an eligibility C-peptide

<0.2 nmol/L, and there was generally no association between having a measurable value

(0.03 – 0.19 nmol/L) versus not (< 0.03 nmol/L) and outcomes, except for the risk of

hypoglycemia. Those with some measurable value had a significantly lower risk of

severe hypoglycemia, even after adjusting for other factors.

Another study (11) using a more recent C-peptide assay showed that in 20 T1D

patients with diabetes duration between 1-16 years, those with undetectable C-peptide

(<0.1 nmol/L) versus those with detectable C-peptide had less glycemic variability, were

more prone to hypoglycemia, and had impaired counter-regulation in response to

hypoglycemia. These results echo our findings that patients above 0.2 nmol/L are

protected in terms of hypoglycemia and that 0.2 nmol/L is not a good definition for

clinically significant residual insulin secretion.

One weakness of these DCCT data is that C-peptide was principally measured

during eligibility screening. An ancillary study also measured C-peptide annually among

those who entered the trial and retained C-peptide response (stimulated value > 0.2

nmol/L), but with no further testing if results fell below this threshold (4). Thus, it is not

possible to examine the longitudinal association of C-peptide concentrations over time

with outcomes over time. Nevertheless, the large cohort with precise assessments of

glycemia, insulin requirements, severe hypoglycemia, and microvascular outcomes

conclusively shows that each of these outcomes is associated with the amount of

endogenous insulin secretion, as measured by a simple stimulated C-peptide value. These

results have implications for future research into the factors associated with preservation


18

of beta cell function and the design of studies to evaluate the effectiveness of therapies

aimed at such preservation.

Acknowledgments

A complete list of participants in the DCCT/EDIC Research Group can be found in

the New England Journal of Medicine, 2011;365:2366-2376.

The DCCT/EDIC has been supported by U01 Cooperative Agreement grants (1982-

93, 2011-2016), and contracts (1982-2011) with the Division of Diabetes Endocrinology

and Metabolic Diseases of the National Institute of Diabetes and Digestive and Kidney

Disease, and through support by the National Eye Institute, the National Institute of

Neurologic Disorders and Stroke, the Genetic Clinical Research Centers Program (1993-

2007), and Clinical Translational Science Center Program (2006-present), Bethesda,

Maryland, USA. Industry contributors have had no role in the DCCT/EDIC study but

have provided free or discounted supplies or equipment to support participants’

adherence to the study: Abbott Diabetes Care (Alameda, CA), Animas (Westchester,

PA), Bayer Diabetes Care (North America Headquarters, Tarrytown NY) Becton

Dickinson (Franklin Lakes, NJ), CanAm (Atlanta, GA) , Eli Lilly (Indianapolis, IN),

Lifescan (Milpitas, CA) , Medtronic Diabetes (Minneapolis, MI) , Omron (Shelton CT),

OmniPod® Insulin Management System (Bedford, MA) , Roche Diabetes Care

(Indianapolis, IN) , and Sanofi-Aventis (Bridgewater NJ).

None of the authors declared any dualities of interest with regards to the current

study.


19

All authors participated in the design and conduct of the study. John Lachin and Paula

McGee wrote the manuscript with critical review and editing by Jerry Palmer. John

Lachin serves as the guarantor of this manuscript.

References:

1. Palmer JP, Fleming GA, Greenbaum CJ, Herold KC, Jansa LD, Kolb H, Lachin

JM, Polonsky KS, Pozzilli P, Skyler JS, Steffes MW. C-peptide is the appropriate

outcome measure for type 1 diabetes clinical trials to preserve beta-cell function:

report of an ADA workshop, 21-22 October 2001. Diabetes, 2004;53:250-64.

Erratum: Diabetes, 2004;53:1934.

2. Palmer JP. C-peptide in the natural history of type 1 diabetes.

Diabetes/metabolism research and reviews, 2009;25:325-328.

3. The Diabetes Control and Complications Trial (DCCT) Research Group: Effects

of age, duration and treatment of insulin-dependent diabetes mellitus on residual

beta-cell function: observations during eligibility testing for the Diabetes Control

and Complications Trial (DCCT). J Clin Endocrinol Metab 1987;65:30–36.

4. The DCCT Research Group: Effect of intensive therapy on residual β-cell

function in patients with type 1 diabetes in the Diabetes Control and

Complications Trial. Ann Intern Med 1998; 128:517–523.

5. Steffes MW, Sibley S, Jackson M, Thomas W: β-Cell function and the

development of diabetes-related complications in the Diabetes Control and

Complications Trial. Diabetes Care 2003;26:832–836.

6. The DCCT Research Group. The effect of intensive treatment of diabetes on the

development and progression of long-term complications in insulin-dependent

diabetes mellitus. The New England Journal of Medicine 1993;329:977-986.

7. The DCCT Research Group: Effect of intensive diabetes treatment on nerve

conduction in the Diabetes Control and Complications Trial. Ann Neurol

1995;38:869–880.

8. Cleveland WS and Grosse E. Computational Methods for Local Regression.

Statistics and Computing 1991;1:47–62.

9. Demidenko E. Mixed Models: Theory and Applications. Hoboken, New Jersey:

John Wiley & Sons, 2004.

10. Lachin JM. Biostatistical Methods: The Assessment of Relative Risks. Second

Edition. John Wiley and Sons, 2011.


20

11. Fukuda M, Tanaka A, Tahara Y, Ikegami H, Yamamoto Y, Kumahara Y, Shima

K. Correlation between minimal secretory capacity of pancreatic β-cells and

stability of diabetic control. Diabetes 1988;37:81-88.


21

Table 1. Characteristics on entry into the DCCT and outcome characteristics among

those with < 60 months duration of diabetes (N=855) classified as C-peptide

responders with Stimulated C-peptide ≥ 0.2 nmol/L on study entry versus non-

responders (Stimulated C-peptide < 0.2 nmol/L).

INTENSIVE CONVENTIONAL

Characteristics on entry

Responders

N=138

Non-

Responders

N=274

Responders

N=165

Non-

Responders

N=278

Age (years) (mean ± SD) 28 ± 7* 26 ± 7 28 ± 7* 26 ± 8

Duration (months) (mean ± SD) 26 ± 12* 34 ± 14 26 ± 14* 34 ± 15

Female (N, %) 70 (51%) 132 (48%) 62 (38%) 129 (46%)

Primary cohort (N, %) 121 (88%) 221 (81%) 141 (85%) 230 (83%)

AER mg/24 hr (geometric mean

x/÷ GSD) †

8.6 x/÷ 1.8 10.2 x/÷ 1.9 10.1 x/÷ 1.9 9.8 x/÷ 2.2

HbA1c (%) (mean ± SD) 8.3 ± 1.6* 9.4 ± 1.7 8.4 ± 1.7* 9.3 ± 1.7

HbA1c (mmol/mol) (mean ±

SD)

67 ± 17.5 79 ± 18.6 68 ± 18.6 78 ± 18.6

Total insulin dose (U/kg/day) 0.50 ± 0.19* 0.70 ± 0.24 0.53 ± 0.19* 0.70 ± 0.25

Stimulated C-peptide (nmol/L)

(mean ± SD)

0.317 ±

0.083

0.079 ± 0.054 0.309 ±

0.080

0.077 ±

0.051

Fasting C-peptide (nmol/L)

(mean ± SD)

0.130 ±

0.065

0.044 ± 0.025 0.132 ±

0.060

0.045 ±

0.029

Outcome characteristics in DCCT §

Nephropathy (AER > 40 mg/24

hr)

11 (8%) 38 (14%) 30 (18%) 53 (19%)

3-step retinopathy progression

(first observed)

15 (11%) 66 (24%) 57 (35%) 119 (43%)

3-step retinopathy progression

(sustained)

2 (1%) 19 (7%) 32 (19%) 64 (23%)

5-year confirmed neuropathy 3 (3%) 17 (8%) 17 (15%) 38 (18%)

Severe hypoglycemia

Total number of events

Rate/100 patient years

294

66.5

1100

35.0

105

20.7

294

10.6

* p-value for Reponders vs Non-Responders < 0.05, using Fisher’s exact and Wilcoxon.

By definition the responders and non-responders differed in the stimulated and fasting

C-peptide levels.

† From an analysis of the natural log(AER) and the results presented as a geometric mean

(exp(mean log(x)) and geometric standard deviation (exp(stddev(log(x))).

§ Crude percentages for these outcomes are presented that do not take account of the

follow-up time. Tests of significance are presented in subsequent tables using time-to-

event analyses.


22

Table 2. Adjusted mean difference over 7 years of DCCT follow-up between intensively

treated stimulated C-peptide responders versus non-responders with up to 5 years

duration (< 60 months), and the change per 50% increase in C-peptide obtained using

log(C-peptide) in a linear model* for HbA1c (as % and mmol/mol) and total insulin dose

U/kg/day.

Non-

Responders

Responders C-peptide quantitatively

LS Mean

(95% CI)

LS mean

(95% CI)

p < Change per 50% increase

in C-peptide

(95% CI)

p <

HbA1c %, mmol/mol

Unadjusted*, % and

mmol/mol

7.5 (7.34, 7.58)

58 (57, 59)

6.9 (6.76, 7.08)

52 (50, 54)

0.0001 -0.068 (-0.105, -0.032)

-0.75 (-0.35, -1.15)

0.0003

Adjusted for cohort &

duration, % and

mmol/mol

7.5 (7.35, 7.63)

58 (57, 60)

6.9 (6.74, 7.1)

52 (50, 54)

0.0001 -0.072 (-0.110, -0.033)

-0.78 (-0.36, -1.21)

0.0003

Also HbA1c on entry†,

% and mmol/mol

7.4 (7.24, 7.50)

57 (56, 58)

7.1 (6.90, 7.23)

54 (52, 56)

0.0007 -0.030 (-0.065, 0.004)

-0.33 (-0.71, 0.05)

0.0874

Total insulin dose U/kg/day

Unadjusted* 0.78 (0.75,

0.81)

0.63 (0.59,

0.67)

0.0001 -0.028 (-0.037, -0.019) 0.0001

Adjusted for entry

HbA1c†

0.77 (0.73,

0.80)

0.66 (0.61,

0.70)

0.0001 -0.020 (-0.029, -0.010) 0.0001

Adjusted for entry

insulin§

0.75 (0.72,

0.78)

0.71 (0.67,

0.74)

0.0391 -0.003 (-0.012, 0.005) 0.4442

Adjusted for both‡ 0.74 (0.71,

0.76)

0.72 (0.69,

0.76)

0.5441 0.002 (-0.007, 0.010) 0.7051

*Model including responder versus not and year alone, all year 1-7 values as repeated

measures.

†Model including primary versus secondary cohort, duration of diabetes, and HbA1c on

entry.

§ Model including primary versus secondary cohort, duration of diabetes, and insulin

dose on entry.

‡ Model including primary versus secondary cohort, duration of diabetes, insulin dose

and HbA1c on entry.


23

Table 3. Risk Reduction (RRd, %) of progression of microvascular complications in the

DCCT intensive treatment group per 50% higher C-peptide value, and also comparing C-

peptide responders (> 0.2 -5 nmol/L) versus non-responders (< 0.2 nmol/L), with no

adjustments, and also with adjustment for the complication status and HbA1c on entry.

Unadjusted

Adjusted for Entry

Status* and HbA1c

RRd, %

(95% CI)

RRd %

(95% CI)

Retinopathy Progression

> 3 Step

Responders vs non-resp §

p =

58% (26, 76)

0.0025

50% (12, 72)

0.0171

Per 50% higher C-peptide

p =

11.6% (4.1, 18.6)

0.0032

10.8% (2.8, 18.2)

0.0095

Sustained > 3 Step


p =

79% (9, 95)

0.0360

71% (-26, 93)

0.0947


p =

24.6% (10.7, 36.3)

0.0010

23.8% (8.8, 36.3)

0.0030

Nephropathy Progression


p =

43%‡ (-10, 71)

0.1052

27% (-46, 64)

0.3724


p =

9.4% (-1.2, 18.8)

0.0806

5.5% (-6.0, 15.7)

0.3347

Neuropathy† at 5 years

Responders vs non-resp

p =

65% (-22, 90)

0.0996

65% (-27, 90)

0.1101


p =

12.3% (-5.1, 26.9)

0.1555

12.0% (-6.0, 26.9)

0.1789

Severe Hypoglycemia


p =

45% (38, 52)

<0.0001

51% (44, 57)

<0.0001


p =

8.2% (6.4, 9.9)

<0.0001

8.9% (7.2, 10.6)

<0.0001

* Entry status is presence or absence of retinopathy on entry for analysis of retinopathy

and the log (AER) for nephropathy.


24

† Confirmed clinical neuropathy.

§ The comparison of responders and non-responders was previously published in Palmer

et al. (1).

‡ The HR = 64 in Palmer et al. (Reference 1) is a typographical error.


25

Figure Legends

Figure 1. Model-free (LOESS) estimates of the association of the log of the stimulated C-

peptide on study entry with A) the HbA1c value at year 1 of follow-up; B) the total

insulin dose at year 1; C) incidence of retinopathy progression; and D) incidence of

hypoglycemia. For C and D the log of the rate per 100 patient years of the event is plotted

within percentiles of the C-peptide distribution.

A. HbA1c (%)

B. Total Insulin Dose (U/kg)

C. Retinopathy progression

D. Severe Hypoglycemia

Figure 2. Mean HbA1c over follow-up with 95% confidence limits separately for

stimulated C-peptide responders versus non-responders in the intensive (A) and

conventional (B) treatment groups.

A. Intensive Group

B. Conventional Group

Figure 3. Association of stimulated C-peptide at study entry, qualitatively and

quantitatively, with the total insulin dose (U/kg/d) at each successive year of follow-up,

from longitudinal regression models. A): mean insulin dose among responder vs non-

responder; B): change in insulin dose U/kg/d per 50% lower stimulated C-peptide, each

with 95% confidence limits

A. Mean Insulin Dose

B. Change in Insulin Dose


Model-free (LOESS) estimates of the association of the log of the stimulated C-peptide on study entry with A) the HbA1c value at year 1 of follow-up; B) the total insulin dose at year 1; C) incidence of retinopathy

progression; and D) incidence of hypoglycemia. For C and D the log of the rate per 100 patient years of the

event is plotted within percentiles of the C-peptide distribution.

A. HbA1c (%)

101x101mm (300 x 300 DPI)


B. Total Insulin Dose (U/kg)

101x101mm (300 x 300 DPI)


C. Retinopathy progression

101x101mm (300 x 300 DPI)


D. Severe Hypoglycemia

101x101mm (300 x 300 DPI)


Mean HbA1c over follow-up with 95% confidence limits separately for stimulated C-peptide responders versus non-responders in the intensive (A) and conventional (B) treatment groups.

A. Intensive Group

101x101mm (300 x 300 DPI)


B. Conventional Group

101x101mm (300 x 300 DPI)


Association of stimulated C-peptide at study entry, qualitatively and quantitatively, with the total insulin dose (U/kg/d) at each successive year of follow-up, from longitudinal regression models. A): mean insulin dose among responder vs non-responder; B): change in insulin dose U/kg/d per 50% lower stimulated C-

peptide, each with 95% confidence limits

A. Mean Insulin Dose

101x101mm (300 x 300 DPI)


B. Change in Insulin Dose

101x101mm (300 x 300 DPI)


1

Supplementary Material Appendix

Impact of C-Peptide Preservation on HbA1c, Insulin Dose Requirements and

Microvascular Complications in the DCCT

John M. Lachin ScD, Paula McGee MS, Jerry P. Palmer MD for the DCCT/EDIC Research

Group

This Appendix provides the detailed description of: A: the association of the quantitative

relationship of the quantitative C-peptide with risk of outcomes, B: the association of fasting C-

peptide with the risk of outcomes, C: associations within the conventional treatment group, D:

associations among those with 5-15 years duration; and E: associations with quantitative or

ordinal raw outcome measures.

A: Association of Quantitative C-peptide with Risk of Outcomes

For the primary DCCT outcome, sustained ≥ 3-step progression, a 10% higher C-peptide

was associated with a 6.37% lower risk. To facilitate calculations using other differences in C-

peptide, Appendix Table A1 presents the model coefficient and its 95% confidence limits. For

example, the change in risk with a 25% higher C-peptide at entry, such as 3.75 versus 3.0

pmol/ml, would be computed as 100*(1.25-0.69554

– 1) = -14.4 or a 14.4% risk reduction. A 50%

higher C-peptide would be associated with a change in risk of 100*(1.5-0.69088

– 1) = -24.6 or a

24.6% risk reduction.

Likewise, in order to achieve a given risk reduction, such as a 30% reduction, the

required C-peptide increase would be computed as [1 - 0.3](1/β)

= 1.670 or a 67% difference in C-

peptide.

B: Fasting C-peptide

Tables A.2 and A.3 present analyses in the intensive treatment group of the qualitative

and quantitative associations of the entry fasting C-peptide with outcomes. The results are

described in the text of the main paper.

C: Conventional Treatment Group

Tables A.4 and A.5 present like analyses among those with 1-5 years duration within the

conventional treatment group. The results are described in the text of the main paper.

D: 5 – 15 Years Duration


2

Tables A.6 and A.7 present analyses among those with 5-15 years duration on entry in

the intensive treatment group. Those with any detectable levels are compared to those with non-

detectable levels. The results are described in the text of the main paper.

E: Quantitative and Ordinal Outcome Measures.

Further analyses examined the association between C-peptide at baseline and the detailed

measurements of each outcome: the severity of retinopathy using the ordinal ETDRS scale over

7 years of follow-up, the magnitude of albuminuria using the AER (mg/24 h) and the O’Brien

rank score of the severity of neuropathy on a nerve conduction examination at 5 years of follow-

up. The methods used for each analysis and the results follow.

Table A.8 describes the average overall characteristics during DCCT follow-up of the

855 subjects with 1-5 years duration. The geometric mean AER mg/24 h over the 7 years of

follow-up was somewhat less among responders versus non-responders within the intensive

group with little difference in the conventional group. For descriptive purposes the ETDRS

scores at year 4 of follow-up are presented (thereafter the N observed declines). In the intensive

group the majority of the responders (61%) remain free of retinopathy compared to 42% among

non-responders who also had higher fractions at higher levels of retinopathy severity. Within the

intensive group the O’Brien composite nerve conduction score was higher (less severe) among

responders in the intensive group than non-responders whereas values within the conventional

group were worse among both responders and non-responders, with no difference. Additional

analyses were conducted within the intensive treatment group alone.

Retinopathy. The severity of retinopathy was assessed using the Early Treatment

Diabetes Retinopathy Study (ETDRS) final scale of retinopathy severity (1). Explicit

descriptions of the steps along the scale are provided in prior DCCT analyses (2). The Wei-

Lachin multivariate rank test (3) with the estimate of the Mann-Whitney difference (4) was used

to test the difference between responders and non-responders in the distribution of the EDTRS

scores using the scores from the annual follow-up assessments through DCCT year 7. The Mann-

Whitney parameter is the probability that a non-responder has a higher (worse) score than a

responder minus the probability that a responder has a worse score than a non-responder. The

difference in probabilities is used owing to the discrete nature of the scores. The parameter is

zero when the distributions are the same in the two groups and is > 0 when the scores among

non-responders are higher (worse) than responders.


3

The analysis was conducted with no adjustment. Then a stratified analysis (5) was

conducted that adjusted for membership at baseline in the primary (no retinopathy) versus

secondary (mild retinopathy) cohort. Then a further analysis was stratified by cohort and the

quartiles of the baseline HbA1c within each cohort to provide a joint adjustment for the baseline

retinopathy level and HbA1c. The following are the unadjusted and adjusted results.

The Mann-Whitney Difference in the probability of a worse ETDRS retinopathy

score over 7 years of follow-up for non-responders versus responders.

Mann-Whitney Difference

95% Confidence Limits

p =

Unadjusted 0.131 0.052, 0.211 0.0013

Adjusted for Primary vs

Secondary cohort 0.112 0.042, 0.182 0.0018

Adjusted for cohort and

baseline HbA1c. 0.114 0.003, 0.226 0.0446

Unadjusted, the probability that a non-responder would have worse retinopathy is 0.131 greater

than the probability that a responder would have worse retinopathy, that is highly significant.

The result is virtually identical after adjustment for baseline cohort and remains significant after

adjusting for baseline cohort and HbA1c.

To examine the association of the quantitative C-peptide value with the ETDRS score, we

fit a quantile (median) regression model (6) to the ETDRS scores at year 4 as shown in Table

A.8 as a function of the log C-peptide and the results were expressed as the number of steps

change in the median on the ETDRS scale per 50% increase in C-peptide. The results are shown

in the following table.

Steps change in the median on the ETDRS scale at 4 years of follow-up per 50%

increase in C-peptide Steps

Lower

95% Confidence

Limits

p =

Unadjusted -0.15 -0.23, -0.07 0.0001

Adjusted for Primary vs

Secondary cohort -0.16 -0.23, -0.08 0.0001

Adjusted for cohort and

baseline HbA1c. -0.029 -0.07, 0.010 0.15


4

The median number of steps on the ETDRS scale is reduced by about 0.15 steps per 50% higher

level of C-peptide that is highly significant both unadjusted and then adjusted for baseline

cohort. However, there is no meaningful change in the median number of steps as the C-peptide

increases when also adjusting for HbA1c.

Nephropathy. The difference in the log AER between responders and non-responders,

and the percentage change per 50% increase in C-peptide were assessed in a longitudinal mixed

model adjusted for the follow-up year using values up to 7 years of follow-up. The following

table presents the mean percentage change in AER for responders versus non-responders over

the 7 years of follow-up

Mean percentage difference in AER over 7 years of follow-up for Responders versus

non-responders Percent

Difference

95% Confidence

Limits

p =

Unadjusted -22.2% -11.6, -33.8 0.0001

Adjusted Baseline AER -16.5% -7.3, -26.5 0.0003

Adjusted Baseline AER

and baseline HbA1c. -13.8% -4.5, -23.9 0.0030

Unadjusted, the AER among responders was 22% less than that of non-responders over the 7

years of follow-up. Adjustment for the baseline AER and also for HbA1c diminished this

difference somewhat but it remained significantly different.

A like analysis was conducted using the log C-peptide values and the results are

presented as the percentage change in AER per 50% increase in C-peptide in the following table

Percentage change in AER over 7 years of follow-up per 50% increase in C-peptide

Percent change 95% Confidence

Limits

p =

Unadjusted -2.9% -4.6, -1.2 0.0011

Adjusted Baseline AER -1.8% -3.4, -0.2 0.0260

Adjusted Baseline AER

and baseline HbA1c. -1.4% -2.95, 0.27 0.1015


5

Unadjusted, the AER decreased by about 3% per 50% higher C-peptide over the 7 years of

follow-up. Adjustment for the baseline AER diminished this effect somewhat but it remained

significantly different. Adjustment for HbA1c diminished the effect further that was not

significant.

Neuropathy. Nerve conduction assessments were conducted in the complete cohort at 5

years of follow-up and included measurement of 10 components: median motor amplitude,

median motor conduction velocity, median motor F-wave latency, median sensory amplitude,

median sensory conduction velocity, peroneal amplitude, peroneal conduction velocity, peroneal

F-wave latency, sural amplitude, and sural conduction velocity. For each component, the

fractional rank was computed for each participant such that higher values represented worse

nerve conduction. For the two latencies, the descending fractional ranks were used so that higher

values were worse. The mean fractional rank (O’Brien score) was then used in a Mann-Whitney

analysis (8). These scores take values between 0 and 1. An adjusted Mann-Whitney analysis was

obtained from an analysis of the residuals after fitting a quantile (median) regression model of

the 5 year O’Brien scores regressed on either the baseline O’Brien score, or the baseline score

and HbA1c. The results are interpreted as above for the analysis of the retinopathy ETDRS

scores.

The following are the unadjusted and adjusted results.

The Mann-Whitney Difference in the probability of a worse O’Brien nerve

conduction score at 5 years of follow-up for non-responders versus responders.

Mann-Whitney Difference

95% Confidence Limits

p =

Unadjusted 0.158 0.027, 0.288 0.018

Adjusted for baseline

O’Brien score 0.080 0.054, 0.213 0.25

Adjusted for baseline

O’Brien score and

baseline HbA1c.

0.066 0.068, 0.200 0.34

Unadjusted, the probability that a non-responder would have worse neuropathy is 0.158 greater

than the probability that a responder would have worse neuropathy. After adjusting for the

baseline neuropathy score alone, or the baseline score and HbA1c, the difference between

responders versus non-responders is diminished.


6

Quantile (median) regression models were then used to assess the change in the median

O’Brien score at 5 years as a function of the log C-peptide unadjusted and then adjusted. The

change in the median rank per 50% increase in C-peptide is presented in the following table

Change in the median O’Brien score at 5 years of follow-up per 50% increase in C-

peptide

Change 95% Confidence

Limits

p =

Unadjusted 0.003 -0.007, 0.012 0.58

Adjusted Baseline

O’Brien score 0.002 -0.003, 0.007 0.38

Adjusted Baseline

O’Brien score and

baseline HbA1c.

0.001 -0.003, 0.006 0.55

There is no association between the quantitative C-peptide value and the O’Brien score.

References:

1. Early Treatment Diabetic Retinopathy Study Research Group. Fundus photographic risk

factors for progression of diabetic retinopathy: ETDRS report number 12.

Ophthalmology 1991; 98:Suppl:823-33.

2. The Diabetes Control and Complications Trial Research Group. The effect of diabetes

therapy on the progression of diabetic retinopathy in insulin-dependent diabetes mellitus:

the Diabetes Control and Complications Trial. Arch Ophthalmol 1995;113:36-51.

3. Wei LJ and Lachin JM. Two-sample asymptotically distribution-free tests for incomplete

multivariate observations. J. Amer. Statist. Assoc., 79, 653-661, 1984.

4. Thall PF and Lachin JM. Assessment of stratum-covariate interactions in Cox's

proportional hazards regression model. Statistics in Medicine 1986; 5:73-83.

5. Lachin, JM. Some large sample size distribution-free estimators and tests for multivariate

partially incomplete observations from two populations. Statistics in Medicine

1992;11:1151-1170.

6. Koenker, R. and Bassett, G. W. Regression Quantiles, Econometrica 1978;46: 33–50.

7. Demidenko E. Mixed Models: Theory and Applications. Hoboken, New Jersey: John

Wiley & Sons, 2004.

8. O'Brien PC. Procedures for comparing samples with multiple endpoints. Biometrics.

1984;40: 1079-1089.


7

Table A.1. Coefficient (β) for the effect of the log of the C-peptide level (pmol/ml) on risk of

progression of microvascular complications in the DCCT intensive treatment in a Proportional

Hazards model with no adjustments, adjustment for the entry complication status, and with

adjustment for the entry HbA1c.

Unadjusted

Adjusted for Entry

Status * and HbA1c

β (95% CI) β (95% CI)

Retinopathy

> 3 Step Progression

-0.30456

(-0.50728, -0.10185)

-0.28257

(-0.49621, -0.06893)

Sustained > 3 Step

Progression

-0.69544

(-1.11057, -0.28031)

-0.66947

(-1.11153, -0.22741)

Nephropathy

AER > 40 mg/24 h

-0.24201

(-0.51351, 0.029491)

-0.13921

(-0.42205, 0.14363)

* Entry status is presence or absence of retinopathy on entry for analysis of retinopathy,

and the log (AER) on entry for nephropathy.


8

Table A.2. Adjusted mean difference over 7 years of DCCT follow-up between intensively

treated fasting C-peptide responders (≥ 0.075pmol/mL) versus non-responders with up to 5 years

duration (< 60 months), and the change per 50% increase in C-peptide obtained using log(C-

peptide) in a linear model*. A. HbA1c %; B. Total insulin dose U/kg/day.

A. HbA1c %.

Fasting Non-

Responders

Fasting

Responders

Fasting C-peptide

quantitatively

LS Mean

(95% CI)

LS mean

(95% CI)

p < Change in HbA1c

insulin per 50%

increase in C-

peptide

(95% CI)

p <

Unadjusted* 7.3 (7.21, 7.46) 7.2 (7.0, 7.31) 0.0576 -0.0299 (-0.0419,

-0.0178)

0.0001

(adj for cohort +

duration only)

7.4 (7.23, 7.53) 7.2 (7.02, 7.34) 0.0707 -0.0648 (-0.1157,

-0.0139)

0.0130

Adjusted† 7.3 (7.14, 7.40) 7.3 (7.12, 7.43) 0.9602 0.0032 (-0.0179,

0.0136)

0.5513

B. Total insulin dose U/kg/day.

Fasting Non-

Responders

Fasting

Responders

Fasting C-peptide

quantitatively

LS Mean

(95% CI)

LS mean

(95% CI)

p < Change in insulin

per 50% increase

in C-peptide

(95% CI)

p <

Unadjusted* 0.7655 (0.7370,

0.7940)

0.6626 (0.6244,

0.7008)

0.0001 -0.0299 (-0.0419 -

0.0178)

0.0001

Adjusted‡ 0.7224 (0.6942,

0.7506)

0.7476 (0.7131,

0.7820)

0.2180 0.00317 (-.00725

0.013586)

0.5513

*Model including responder versus non-responders and year alone, all year 1-7 values as

repeated measures.

†Model including primary versus secondary cohort and duration of diabetes, and HbA1c on

entry.

‡ Model including primary versus secondary cohort and duration of diabetes, insulin dose and

HbA1c on entry.


9

Table A.3. Risk Reduction (RRd %) of progression of microvascular complications in the DCCT

intensive treatment group per 50% higher fasting C-peptide value, and also comparing fasting C-

peptide responders (> 0.075 pmol/L) on entry versus non-responders (< 0.075 pmol/L), with no

adjustments, and also with adjustment for the entry complication status and HbA1c.

Unadjusted

Adjusted for Entry

Status* and HbA1c

RRd %

(95% CI)

RRd %

(95% CI)


> 3 Step


p =

58% (-3, 59)

0.0685

43% (-13, 135)

0.1596


p =

8.8% (-1.9, 18.5)

0.1045

8.7% (-2.2, 18.5)

0.1141

Sustained > 3 Step


p =

247% (2, 1081)

0.0464

190% (-15, 892)

0.0905


p =

28.6% (9.5, 43.6)

0.0052

27.0% (7.6, 42.3)

0.0090



p =

146% (19, 411)

0.0156

217% (-1, 333)

0.0538


p =

12% (-2, 24)

0.0958

8.6% (-6.3, 21.5)

0.2419



p =

60% (-24, 87)

0.1135

59% (-29, 87)

0.1267


p =

16% (-7.5, 35)

0.1644

16% (-8.5, 35)

0.1839

Severe Hypoglycemia


p =

45% (38, 52)

<0.0001

47% (40, 53)

<0.0001


p =

9.8% (7.5, 12.1)

<0.0001

10.8% (8.5, 13.0)

<0.0001




10

Table A.4. Adjusted mean difference over 7 years of DCCT follow-up between conventionally

treated stimulated C-peptide responders (≥ 0.2pmol/mL) versus non-responders with up to 5

years duration (< 60 months), and the change per 50% increase in C-peptide obtained using

log(C-peptide) in a linear model*. A. HbA1c %; B. Total insulin dose U/kg/day.

A. HbA1c %.

Stimulated Non-

Responders

Stimulated

Responders

Stimulated C-peptide

quantitatively

LS Mean

(95% CI)

LS mean

(95% CI)

p < Change in HbA1c

insulin per 50%

increase in C-

peptide

(95% CI)

p <

Unadjusted* 9.3 (9.15, 9.47) 9.2 (9.03, 9.45) 0.5683 -0.0046 (-

0.05578, 0.04657)

0.8600

(adj for cohort +

duration only)

9.3 (9.14, 9.54) 9.3 (9.01, 9.49) 0.5247 -0.0069 (-0.0608,

0.0470)

0.8022

Adjusted† 9.2 (9.08, 9.41) 9.4 (9.23, 9.63) 0.0901 0.0645 (0.0209,

0.10806)

0.0039


Stimulated Non-

Responders

Stimulated

Responders

Stimulated C-

peptide

quantitatively

LS Mean

(95% CI)

LS mean

(95% CI)

p < Change in insulin

per 50% increase

in C-peptide

(95% CI)

p <

Unadjusted* 0.6871

(0.6634, 0.7108)

0.6161

(0.5854, 0.6468)

0.0003 -0.044199 (-

0.017921,

0.00535)

0.2905

Adjusted‡ 0.6548

(0.6347, 0.6749)

0.6856

(0.6611, 0.7101)

0.0243 0.03068

(0.001307,

0.012441)

0.0159

*Model including responder versus not and year alone, all year 1-7 values as repeated measures.


entry.


HbA1c on entry.


11


conventional treatment group with duration < 5 years per 50% higher stimulated C-peptide value

on entry, and also comparing stimulated C-peptide responders (> 0.2 pmol/L) versus non-

responders (< 0.2 pmol/L), with no adjustments, and also with adjustment for the entry

complication status and HbA1c.

Unadjusted

Adjusted for Entry

Status* and HbA1c

RRd %

(95% CI)

RRd %

(95% CI)


> 3 Step


p =

13% (-27, 37)

0.4110

-13% (-58, 20)

0.4919


p =

2.9% (-3.1, 8.6)

0.3359

-1.6% (-8.5, 4.8)

0.6314

Sustained > 3 Step


p =

6% (-46, 39)

0.7946

-25% (-96, 20)

0.3218


p =

3.3% (-4.9, 10.7)

0.4215

-2.5% (-12.0, 6.2)

0.5812



p =

9% (-48, 44)

0.7143

-2% (-68, 37)

0.9262


p =

7.8% (-1.1, 15.8)

0.0840

4.8% (-4.6, 13.4)

0.3079



p =

23% (-43, 59)

0.4093

13% (-65, 54)

0.6746


p =

1.4% (-10.9, 12.4)

0.8094

1.4% (-14.8, 10.4)

0.8250

Severe Hypoglycemia


p =

38% (22, 50)

<0.0001

44% (30, 56)

<0.0001


p =

5.5% (2.1, 8.9)

0.0017

7.8% (4.4, 11.1)

<0.0001




12

Table A.6. Adjusted mean difference over 7 years of DCCT follow-up between intensively

treated patients with a detectable stimulated C-peptide (> 0.03pmol/mL) versus those

undetectable with 5-15 years duration (≥ 60 months). A. HbA1c %; B. Total insulin dose

U/kg/day.

A. HbA1c %.

Stimulated

undetectable

Stimulated

Detectable

LS Mean

(95% CI)

LS mean

(95% CI)

p <

Unadjusted* 7.3 (7.15, 7.39) 7.2 (6.96, 7.39) 0.4414

(adj for cohort +

duration only)

7.5 (7.12, 7.81) 7.2 (6.84, 7.64) 0.0675

Adjusted† 7.4 (7.04, 7.68) 7.2 (6.87, 7.60) 0.2618


Stimulated

undetectable

Stimulated

Detectable

LS Mean

(95% CI)

LS mean

(95% CI)

p <

Unadjusted* 0.7187

(0.6909, 0.7465)

0.7363

(0.6842, 0.7883)

0.5573

Adjusted‡ 0.7631

(0.6797, 0.8233)

0.8021

(0.7272, 8771)

0.0923

*Model including responder versus not and year alone, all year 1-7 values as repeated measures.


entry.


HbA1c on entry.


13


intensive treatment group comparing those with a detectable stimulated C-peptide (>

0.03pmol/mL) on entry versus those undetectable with 5-15 years duration (≥ 60 months), with

no adjustments, and with adjustment for the entry complication status and HbA1c.

Unadjusted

Adjusted for Entry Status*

and HbA1c

RRd (95% CI) RRd (95% CI)


> 3 Step

Detectable vs undetectable

p =

1.79 (1.01, 3.17)

0.0474

1.61 (0.90, 2.86)

0.1071

Sustained > 3 Step


p =

1.17 (0.56, 2.43)

0.6794

0.978 (0.47, 2.056)

0.9539



p =

0.96 (0.53, 1.75)

0.9036

0.78 (0.42, 1.43)

0.4178



p =

0.91 (0.35, 2.39)

0.8529

0.92 (0.35, 2.42)

0.8638

Severe Hypoglycemia


p =

22 (12, 32)

0.002

23 (12 33)

<0.0001




14

Table A.8. Outcome characteristics among those with < 60 months duration of diabetes

(N=855) classified as C-peptide responders with Stimulated C-peptide ≥ 0.2 pmol/mL on

study entry versus non-responders (Stimulated C-peptide < 0.2 pmol/mL).

INTENSIVE CONVENTIONAL

Responders

N=138

Non-

Responders

N=274

Responders

N=165

Non-

Responders

N=278

Ordinal or quantiative outcome characteristics§

AER (geometric mean x/÷ GSD)* 8.13 x/÷ 2.20 9.94 x/÷ 1.78 11.1 x/÷ 2.84 11.5 x/÷2.30

Year 4 Retinopathy levels

1 = 10, no retinopathy 73 (61%) 107 (42%) 58 (40%) 91 (35%)

2 = 20/10, very mild NPDR 27 (23%) 79 (31%) 35 (24%) 64 (25%)

3 = 20/20 18 (15%) 52 (20%) 28 (19%) 62 (24%)

4 = 35/<35, mild NPDR 1 (0.8%) 12 (5%) 16 (11%) 28 (11%)

5 = 35/35 0 3 (1%) 2 (1%) 9 (3%)

6 = 43/<43, Moderate NPDR(1) 0 2 (0.8%) 4 (3%) 3 (1%)

7 = 43/43, Moderate NPDR(1) 0 0 0 3 (1%)

8 = 47/<47, Moderate NPDR(2) 0 1 (0.4%) 0 0

9 = 47/47, Moderate NPDR(2) 1 (0.8%) 0 1 (0.7%) 0

Neuropathy O’Brien score at year 5

median (25,75 percentiles)

0.62

(0.50, 0.71)

0.56

(0.47, 0.66)

0.45

(0.35, 0.59)

0.47

(0.36, 0.58)

* From a longitudinal regression model of the natural log(AER) with a class effect for the 7

years of follow-up comparing responders versus non-responders, with separate models within

each treatment group. Results presented as a geometric mean (exp(mean log(x)) and geometric

standard deviation (exp(stddev(log(x)))

† Steps on the final ETDRS scale of retinopathy severity where X/<X designates that the worse

eye is at level X and the other eye at a lesser level, X/X designates that both eyes are at level X.

NPDR equals non-proliferative diabetic retinopathy. Values were missing for 36 intensive and

39 conventional patients that are excluded from denominator

§ The O’Brien mean fraction rank among the 10 nerve conduction components, a value 0.5 refers

to the median O’Brien score in the cohort. Lower O’Brien scores reflect increasing severity of

nerve conduction defects..


Documents

Impact of C-Peptide Preservation on Metabolic and Clinical ......2013/09/23 · 09/03/13 Impact of C-Peptide Preservation on Metabolic and Clinical Outcomes in the Diabetes Control