Search

Chemokines and their receptors have been implicated in the development of diabetic nephropathy. To determine whether the risk of diabetic nephropathy is influenced by two functional polymorphisms in the regulated upon activation normal T-cell expressed and secreted (RANTES) receptor gene (CCR5), we recruited patients with type 1 diabetes, including 496 case subjects with overt proteinuria or end-stage renal disease and 298 control subjects with normoalbuminuria. Male carriers of the 59029G allele, which is associated with diminished expression of CCR5 on the surface of immunocompetent cells, had significantly higher risk of developing diabetic nephropathy than noncarriers (OR [95% CI] 1.9 [1.2-3.0]). Similarly, male carriers of the 32-bp deletion, which causes truncation of the protein, had significantly higher risk of diabetic nephropathy than noncarriers (2.3 [1.3-4.2]). Combining both polymorphisms, three haplotypes were distinguished: one nonrisk haplotype carrying the 59029A allele and the 32-bp insertion and two risk haplotypes carrying the 59029A allele with the 32-bp deletion and carrying the 59029G allele with the 32-bp insertion. The distribution of these haplotypes differed significantly (P0.00001) in men with and without diabetic nephropathy but was not associated with diabetic nephropathy in women. In conclusion, two functional polymorphisms in CCR5 that decrease expression of the RANTES receptor on immunocompetent cells are associated with increased risk of diabetic nephropathy in type 1 diabetes, but only in men.

Because of the role of insulin receptor substrate-1 in insulin action, the insulin receptor substrate-1 gene is a candidate gene for noninsulin-dependent diabetes mellitus (NIDDM). Modest associations between NIDDM and a GGG--AGG single base substitution (corresponding to a glycine--arginine amino acid substitution) in codon 972 of the gene have been found, but none reached statistical significance. To examine further how large a proportion of NIDDM cases could be caused by the mutation, we performed a stratified analysis combining the results from the 6 earlier studies and those from our panel of 192 unrelated NIDDM subjects and 104 healthy controls. In addition, we looked for a possibility that the codon 972 mutation plays a role only in the presence of certain conditions. Genomic DNA samples obtained from NIDDM cases and healthy controls were genotyped using a PCR-restriction fragment length polymorphism protocol modified for genomic DNA. The GGG--AGG substitution was found in 5.7% of the diabetic subjects (11 of 192) and 6.9% of the controls (7 of 104). The difference between groups was not statistically significant, and it was not different from the results of other studies. The Mantel-Haenszel summary odds ratio across all studies was 1.49 (P0.05; 95% confidence intervals, 1.01-2.2). This summary odds ratio is consistent with a small proportion of NIDDM cases (approximately 3%) being caused by the mutation. Exploratory subgroup analyses on our panel suggested a clustering of NIDDM, the codon 972 mutation, and overweight, raising the hypothesis that the mutation may predispose to NIDDM only in the presence of excess body weight.

In the present study, we aimed to determine the frequency of a significant reduction in urinary albumin excretion and factors affecting such reduction in patients with type 1 diabetes and microalbuminuria.The study included 386 patients with persistent microalbuminuria, indicated by repeated measurements of urinary albumin excretion (estimated on the basis of albumin-to-creatinine ratios) in the range of 30 to 299 microg per minute during an initial two-year evaluation period. Subsequent measurements during the next six years were grouped into two-year periods, averaged, and analyzed for regression of microalbuminuria, which was defined as a 50 percent reduction in urinary albumin excretion from one two-year period to the next.Regression of microalbuminuria was frequent, with a six-year cumulative incidence of 58 percent (95 percent confidence interval, 52 to 64 percent). The use of angiotensin-converting-enzyme inhibitors was not associated with the regression of microalbuminuria. However, microalbuminuria of short duration, salutary levels of glycosylated hemoglobin (less than 8 percent), low systolic blood pressure (less than 115 mm Hg), and low levels of both cholesterol and triglycerides (less than 198 mg per deciliter [5.12 mmol per liter] and 145 mg per deciliter [1.64 mmol per liter], respectively) were independently associated with the regression of microalbuminuria. Patients with salutary levels of all modifiable factors had a hazard ratio for regression of 3.0 (95 percent confidence interval, 1.5 to 6.0), as compared with patients with no salutary levels of any modifiable factor.Frequent regression of microalbuminuria in patients with type 1 diabetes indicates that elevated urinary albumin excretion does not imply inexorably progressive nephropathy. Identification of the multiple determinants of the regression of microalbuminuria has implications for current theories about the mechanisms of early diabetic nephropathy.

Background: The T allele of the C677T polymorphism in the methylenetetrahydrofolate reductase ( MTHFR ) gene is associated with elevated plasma homocysteine levels, and it has been postulated to be a risk factor for the development of diabetic nephropathy. We examined this hypothesis in both a case-control and a follow-up study in individuals with type 1 diabetes. Methods: In the case-control study, the control group included 310 subjects with normoalbuminuria and diabetes duration of 15 years or greater, and the case group included 88 prevalent cases with end-stage renal disease (ESRD). The follow-up study included 235 subjects with overt proteinuria followed up for 6 years (on average), during which time ESRD developed in 69 subjects. DNA from each individual was genotyped for the C677T MTHFR polymorphism. Results: The frequency of TT homozygotes did not vary significantly among the four groups: 10% in controls, 15% in prevalent cases of ESRD, 13% in cases with new-onset ESRD, and 11% in those who remained proteinuric during follow-up ( P = 0.9, 6 df ). Similarly, frequency of the T allele varied little among the same groups (range, 33% to 36%; P = 0.9, 3 df ) During follow-up, 52 of 323 individuals with diabetic nephropathy died. Total mortality rates were 4.3/100 person-years in TT homozygotes, 2.4/100 person-years in CT heterozygotes, and 3.0/100 person-years in CC homozygotes ( P = 0.55, 2 df ). Conclusion: Using both a large case-control and a follow-up study, we found no evidence that the C677T MTHFR polymorphism has a significant role in the development of diabetic nephropathy in type 1 diabetes.

To determine the distribution of the age at onset of diabetes (maturity-onset diabetes of the young-3 [MODY3]) and to identify determinants of the onset of diabetes in carriers of HNF-1alpha mutations.Extended families (n = 104) with type 2 diabetes inherited in a dominant pattern were recruited and screened for diabetes-causing mutations in HNF-1alpha.HNF-1alpha mutations cosegregated with diabetes in only 13 families, all with a mean age at onset35 years. Insulin secretion was diminished or absent in mutation carriers (n = 101), and diabetes developed in 65% by age 25 years and in 100% by age 50 years. If the mutation was inherited from the mother, diabetes onset was very young in those exposed to diabetes in utero; 57 +/- 8% were affected by age 15 years as compared with 0.0% in those not exposed (P7 x 10(-6)). By age 25 years, the difference was reduced (85 +/- 6 and 55 +/- 12%, respectively; P = 0.02). If the mutation was inherited from the father, diabetes developed in 52 +/- 8% by age 25 years. Age at diagnosis was shown to be highly heritable (h(2) = 0.47, P = 0.003). When parent of origin was included in the analyses, the magnitude of genetic contribution increased markedly (h(2) = 0.91).Mutations in HNF-1alpha accounts for diabetes in a small proportion of families with a dominant pattern of inheritance. Age at onset of diabetes in MODY3 families varied widely and was influenced by familial factors (including modifying genes) and parent of origin (whether a mutation carrier was exposed to diabetes in utero).

Genes play a role in many processes underlying late diabetic complications, but efforts to identify genetic variants have produced disappointing and contradictory results. Here, we evaluate whether the study designs and analytic methods commonly being used are optimal for finding susceptibility genes for diabetic complications. We do so by generating plausible genetic models and assessing the performance of case-control and family-based trio study designs. What emerges as a key determinant of success is duration of diabetes. This perspective focuses on duration of diabetes before complication onset and its influence on the ability to detect major and minor gene effects. It does not delve into the distinct effect of duration after complication onset, which can enrich case subjects with genotypes conferring survival advantage. We use clinically diagnosed nephropathy in type 1 diabetes to show how ignoring duration can result in considerable power loss in both case-control and family-based trio designs. We further show how, under certain circumstances, disregard for duration information can paradoxically lead to implicating nonrisk alleles as causative. Our results indicate that problems can be minimized by selecting case subjects with short diabetes duration and, to a lesser extent, control subjects with long duration or, perhaps, by adjusting for duration during analysis.

A polymorphism in the ecto-nucleotide pyrophosphatase/phosphodiesterase 1 gene (ENPP1) (previously known as PC-1), resulting in an amino acid change from lysine to glutamine at codon 121 (K121Q), is associated with insulin resistance. A small follow-up study of patients with type 1 diabetes and proteinuria found that renal function declines more rapidly in carriers of the Q variant than in noncarriers. To examine this finding further, we conducted a large case-control study and a family-based study. Genomic DNA was obtained from 659 patients: 307 with normal urinary albumin excretion despite diabetes duration of >15 years (control subjects) and 352 with advanced diabetic nephropathy, of whom 200 had persistent proteinuria and 152 had end-stage renal disease (ESRD). Individuals were genotyped for Q and K variants using a previously described protocol. The frequency of Q variant carriers was 21.5% in control subjects, 31.5% in subjects with proteinuria, and 32.2% in subjects with ESRD (P = 0.012). In a stratified analysis according to duration of diabetes, the risk of early-onset ESRD for carriers of the Q variant was 2.3 times that for noncarriers (95% CI, 1.2–4.6). The Q variant was not associated with late-onset ESRD. Similar findings were obtained in a family-based study. We conclude that carriers of the Q variant of ENPP1 are at increased risk for developing ESRD early in the course of type 1 diabetes.

Aims The goal was to estimate the sibling recurrence-risk ratio for Type 2 diabetes in families with diabetes occurring in middle age. Because diabetes aetiology involves environmental exposures and genetic susceptibility, we sought to identify determinants of the recurrence risk. Methods We surveyed patients diagnosed at ages 35–59 years (n = 563) to obtain information on the occurrence of diabetes in their relatives, particularly siblings (n = 1675). Age-specific prevalences of diabetes in the US population were used for comparison. Results The overall sibling recurrence-risk ratio for diabetes was low, about 1.8 in the Joslin families and even lower in three other studies that were reanalysed for comparison. In all studies, the diabetes risk in siblings of index cases without a history of diabetes in a parent was similar to that in the general population, suggesting that genetic factors contributed to the occurrence of diabetes in only a minority of these siblings. The fact that recurrence-risk ratios were elevated only in families with one or two diabetic parents indicates that susceptibility to Type 2 diabetes is transmitted primarily through an affected parent. In addition, the sibling recurrence-risk ratios were elevated even further in families with diabetes in both a parent and grandparent of the index case, and in siblings of non-obese index cases (percent ideal body weight

Evaluating the potential genetic components of complex disease will likely be aided through the use of dense polymorphism maps. Previously, we reported evidence for linkage with diabetic nephropathy on chromosome 3q in a region encompassing the type 1 angiotensin II receptor (AGTR1) gene. To further investigate any role for this gene in disease onset, we set out to design a dense polymorphism map spanning the AGTR1 locus for the purpose of association studies. Toward this goal, we have developed a technique for rapid identification of polymorphisms in long stretches of genomic DNA. This approach uses long-range PCR, DNA pooling, and transposon-based DNA sequencing. Using this technique, we efficiently validated and genotyped 18 polymorphisms spanning the 60.5-kb AGTR1 locus. Our panel of polymorphisms has an average spacing of 3.2 kb and an average minor allele frequency of 24%.

Resistin is a newly identified hormone secreted by adipocytes that inhibits insulin action on peripheral tissues. The aim of our study was to investigate whether genetic variability at this locus is associated with the risk of type 2 diabetes. By sequencing 32 subjects with type 2 diabetes, we identified 8 single nucleotide polymorphisms (SNPs) in the 5′-flanking region and introns of the resistin gene. Allele and genotype distributions were determined for all 8 SNPs in 312 cases with type 2 diabetes and 303 nondiabetic controls, all of Caucasian origin. No significant association with type 2 diabetes was found at any of the polymorphic loci. However, an interactive effect of genotype at SNP 6 (IVS2 + 181G→A) and obesity was a significant determinant of type 2 diabetes risk in this population. The relative risk of diabetes for the A/A genotype was 4.8 (95% confidence interval, 1.1–21.0) in individuals above the median for body weight, but only 0.7 (95% confidence interval, 0.2–2.1) in those below the median. This difference between relative risks was significant (χ2 = 4.5; P = 0.03). A similar, but much weaker, interaction with obesity was observed for SNPs in linkage disequilibrium with SNP6. In conclusion, resistin does not appear to be a major gene for type 2 diabetes. However, our data suggest a synergistic effect of sequence differences at the resistin locus and obesity on risk of type 2 diabetes. Further studies are needed to confirm this finding in other populations.

Aims/hypothesis. Oxidative stress is a putative mechanism in the development of diabetic nephropathy. Paraoxonase gene 1 is an HDL-bound enzyme that protects tissues against oxidative damage. Three common polymorphisms of paraoxonase gene 1, T-107C in the promoter, Leu54Met and Gln192Arg, that modify paraoxonase activity have been associated with cardiovascular disease. This study aimed to find whether these polymorphisms also contribute to the development of diabetic nephropathy. Methods. The association between diabetic nephropathy and these three polymorphisms was examined in a case-control study. For this purpose, genomic DNA was collected from 188 patients with Type I (insulin-dependent) diabetes mellitus and diabetic nephropathy and from 179 unrelated patients with Type I diabetes but without diabetic nephropathy despite the duration of diabetes of 15 or more years. Results. The genotype and allele frequencies for each of the three polymorphisms (T-107C, Leu54Met and Gln192Arg) were similar in cases and control subjects. Conclusion/interpretation. The three polymorphisms in paraoxonase gene 1 that have been associated with serum levels of paraoxonase are not associated with diabetic nephropathy. We show that this genetically determined component of the antioxidant capacity of HDL does not play a critical part in the development of diabetic nephropathy. [Diabetologia (2000) 43: 1540–1543]

The goal of this study was to examine the association between known polymorphisms in the apolipoprotein E gene (APOE) and diabetic nephropathy (DN) in type 1 diabetes. We used both a case-control comparison and a family-based study design known as the transmission/disequilibrium test (TDT). For the case-control comparison, we collected DNA from 223 subjects with clinically diagnosed DN and 196 control subjects with normoalbuminuria and long-duration type 1 diabetes (> or = 15 years). For the family-based study, we obtained DNA from both parents of 154 DN subjects and 81 control subjects. The frequency of the epsilon2 allele of exon 4 of APOE was significantly higher in DN subjects than in control subjects. The risk of DN was 3.1 times higher (95% CI 1.6-5.9) in carriers of this allele than in noncarriers. In the family study, heterozygous parents for the E2 allele preferentially transmitted epsilon2 to DN offspring (64 vs. 36%, P < 0.03). Four additional polymorphisms (i.e., -491 A/T, -219 G/T, IE1 G/C, and APOCI insertion/deletion [I/D]) that flank the APOE locus were not associated with DN in either the case-control comparison or in the family-based study. In conclusion, the results of the case-control as well as the family-based study provide evidence that the epsilon2 allele of APOE increases the risk of DN in type 1 diabetes. The molecular mechanisms underlying this risk are unclear at present.

The helix-loop-helix (HLH) protein NEUROD1 (also known as BETA2) functions as a regulatory switch for endocrine pancreatic development. In mice homozygous for a targeted disruption of Neurod, pancreatic islet morphogenesis is abnormal and overt diabetes develops due in part to inadequate expression of the insulin gene (Ins2). NEUROD1, following its heterodimerization with the ubiquitous HLH protein E47, regulates insulin gene (INS) expression by binding to a critical E-box motif on the INS promoter. Here we describe two mutations in NEUROD1, which are associated with the development of type 2 diabetes in the heterozygous state. The first, a missense mutation at Arg 111 in the DNA-binding domain, abolishes E-box binding activity of NEUROD1. The second mutation gives rise to a truncated polypeptide lacking the carboxy-terminal trans-activation domain, a region that associates with the co-activators CBP and p300 (refs 3,4). The clinical profile of patients with the truncated NEUROD1 polypeptide is more severe than that of patients with the Arg 111 mutation. Our findings suggest that deficient binding of NEUROD1 or binding of a transcriptionally inactive NEUROD1 polypeptide to target promoters in pancreatic islets leads to the development of type 2 diabetes in humans.

OBJECTIVE: To investigate whether there are forms of early-onset autosomal-dominant type 2 diabetes that are distinct from typical maturity-onset diabetes of the young (MODY) and to characterize their phenotypic characteristics. RESEARCH DESIGN AND METHODS: The study included 220 affected subjects from 29 families in which early-onset type 2 diabetes occurred in multiple generations and was not linked to known MODY genes (MODY gene-negative families). All individuals underwent an oral glucose tolerance test and other clinical measurements aimed at investigating the underlying metabolic defect and the presence of diabetic complications. For comparison, 79 affected carriers of MODY3 (hepatocyte nuclear factor [HNF]-1 alpha) mutations were similarly examined. RESULTS: Subjects from MODY gene-negative pedigrees were diagnosed with diabetes at an older age (36 +/- 17 vs. 21 +/- 10 years, P = 0.0001) and were more frequently obese (52 vs. 18%, P = 0.0001) than MODY3 individuals. MODY gene-negative patients who were insulin treated required more exogenous insulin than did MODY3 subjects (0.7 +/- 0.4 vs. 0.45 +/- 0.2 U.kg-1.day-1, P = 0.04), despite similar C-peptide levels. Among subjects not treated with insulin, MODY gene-negative subjects had significantly higher serum insulin levels, both fasting (16.5 +/- 15 vs. 6.5 +/- 5 microU/ml, P = 0.027) and 2 h after a glucose load (53 +/- 44 vs. 11 +/- 10, P = 0.002). They also had higher serum triglycerides (P = 0.02), higher cholesterol levels (P = 0.02), more hypertension (P = 0.0001), and more nephropathy (P = 0.001). Differences persisted when families were matched for age at diagnosis. CONCLUSIONS: Our findings indicate the existence of forms of early-onset autosomal-dominant type 2 diabetes that are distinct from MODY and are frequently characterized by insulin resistance, similar to later-onset type 2 diabetes. Because of the Mendelian pattern of inheritance, the goal of identifying the genes involved in these forms of diabetes appears to be particularly feasible.

We investigated whether variability at the insulin receptor substrate (IRS)-2 locus plays a role in the etiology of early-onset autosomal dominant type 2 diabetes. By means of radiation hybrid mapping, we placed the human IRS-2 gene on 13q at 8.6 cRays from SHGC-37358. Linkage between diabetes and two polymorphic markers located in this region (D13S285 and D13S1295) was then evaluated in 29 families with early-onset autosomal dominant type 2 diabetes. Included were 220 individuals with diabetes, impaired glucose tolerance, or gestational diabetes (mean age at diabetes diagnosis 36 +/- 17 years) and 146 nondiabetic subjects. Overall, strongly negative logarithm of odds (LOD) scores for linkage with diabetes were obtained by multipoint parametric analysis (LOD score -45.4 at D13S285 and -40.9 at D13S1295). No significant evidence of linkage was obtained under the hypothesis of heterogeneity or by nonparametric methods. Fourteen pedigrees for which linkage could not be excluded (LOD score > -2.0) were screened for mutations in the IRS-2 coding region by dideoxy fingerprinting. However, no mutations segregating with diabetes could be detected in these families. These data indicate that IRS-2 is not a major gene for early-onset autosomal dominant type 2 diabetes, although a role of mutations in the promoter region cannot be excluded at this time.

In Type I (insulin-dependent) diabetes mellitus a genetic predisposition exists to nephropathy and is related to parental hypertension. Enhanced G-protein activation, a cellular phenotype observed in cultured cells from patients with essential hypertension, was recently documented in Type I diabetic subjects with nephropathy. This enhanced G-protein activation has been associated with a genetic variant in the G-protein β3 subunit, GNB3. A C→T polymorphism at position 825 in exon 10 is associated with G-protein activation, the T allele associated with enhanced activity. Furthermore the T allele was observed more frequently in a group with essential hypertension. In this report we have analysed the role of the C825T polymorphism in the predisposition to diabetic nephropathy in Type I diabetes. We have investigated the frequency of this polymorphism in a large case-control study and found no association of the T allele with diabetic nephropathy. Specifically carriage of the T allele as CT or TT was observed in 49 % of 200 Type I diabetic control subjects with normoalbuminuria (diabetes duration 24 years) compared with 53 % of 216 Type I diabetic subjects with nephropathy (overt proteinuria or end-stage renal failure). Within this group we have also examined the inheritance of C825T alleles in a family study and found no evidence for excess transmission of the T allele to Type I diabetic offspring with nephropathy (T allele transmitted to 51 % of nephropathy offspring, C allele transmitted to 49 % of nephropathy offspring, p = 0.79). In none of the Type I diabetic datasets examined was there any effect of genotype on variation in systolic or diastolic blood pressure. In conclusion we can find no evidence for the C825T polymorphism of the β3 G-protein subunit as a major gene in the susceptibility to diabetic nephropathy in Type I diabetes. [Diabetologia (1998) 41: 1304–1308]

In patients with diabetes, elevated urinary albumin excretion (UAE) indicates the existence of functional as well as morphological abnormalities in the kidney [1–3]. Therefore, measurement of UAE is used for the early diagnosis of diabetic kidney disease as well as for monitoring its natural course and the effectiveness of treatment. Quantitative and semiquantitative methods are used to measure UAE.

We investigated the contribution of polymorphisms in the angiotensin II type 1 receptor gene (AGTR1) to renal complications in an inception cohort of 152 insulin-dependent diabetic (IDDM) patients examined 15-21 years after diabetes onset. This nested case-control study included 79 normoal-buminuric control subjects and 73 cases with evidence of nephropathy ranging from microalbuminuria to overt proteinuria. Subjects were genotyped for two AGTR1 polymorphisms (T573→C and A1166→C), and an adjacent CA repeat microsatellite. Allele C1166 and the 140 bp allele of the microsatellite were more frequent among nephropathy cases than normoalbuminuric control subjects (0.322 vs 0.247, and 0.618 vs 0.521, respectively), but these differences were not statistically significant. Although not significant by themselves, the AGTR1 polymorphisms contributed significantly to the risk of diabetic nephropathy when accompanied by poor glycaemic control. Among patients with frequent severe hyperglycaemia during the first decade of diabetes, the relative risk of nephropathy among allele C1166 carriers was 12.1 (95 % CI: 3.7-39.8), whereas it was only 1.4 (95 % CI: 0.6-3.5) among allele A1166 homozygotes. The difference between relative risks was highly significant (x 2 = 8.25, p = 0.004 with 1 df). A similar pattern of higher risk of microalbuminuria, specifically among those carriers of allele C1166 who had poor glycaemic control was also found in an independent study of a cross-sectional sample of 551 IDDM individuals, although the effect was smaller in magnitude. We conclude that DNA sequence differences in the AGTR1 gene may modify the noxious effects of hyperglycaemia on the kidney. Allele C1166 carriers might especially benefit from nephropathy prevention programmes. [Diabetologia (1997) 40: 1293-1299]

The relationships of insulin secretion and insulin action to body weight are incompletely understood. Obesity is associated with reduced sensitivity to insulin and high fasting and postprandial serum insulin levels. However, it is unknown whether insulin secretion rises to compensate for insulin resistance or high insulin secretion promotes body weight gain and the development of insulin resistance. To shed light on this question, we examined weight gain over an interval of 16.7 ± 3.9 years (mean ± SD) in 107 glucose-tolerant offspring (48 men, 59 women) of two parents with NIDDM. The offspring had a baseline intravenous glucose tolerance test, at which time they were aged 32.9 ± 9.7 years, and only those who did not develop diabetes during the follow-up period were included. We estimated insulin sensitivity with the insulin sensitivity index from Bergman's minimal model of glucose disposal and acute insulin secretion from the incremental area under the insulin curve in the first 10 min of the intravenous glucose tolerance test. Weight-gain rate (g/year) was defined as the regression slope of each subject's body weight over time. High acute insulin secretion, young age, and low baseline percent ideal body weight (IBW) were each associated with a high rate of weight gain. After adjustment for differences in age and IBW, statistically significant effects of insulin sensitivity (P = 0.05) as well as acute insulin secretion (P = 0.001) were obtained. To estimate the effects of acute insulin secretion and insulin sensitivity on the average rate of weight gain (adjusting for age and IBW), the study group was stratified into four subgroups by dividing it at the medians of these two variables. Among those with low acute insulin secretion, weight-gain rate was the same regardless of whether insulin sensitivity was low or high (176 and 152 g/year, respectively). Among those with high acute insulin secretion, mean weight-gain rate was still rather low in those with low insulin sensitivity (271 g/year), but it was quite high in those with high insulin sensitivity (672 g/year; significantly higher than in all other subgroups). Therefore a high first-phase insulin response to intravenous glucose is a risk factor for long-term weight gain, and this effect is particularly manifested in insulin-sensitive individuals.

To test the hypothesis that a gene (or genes) in the “MODY1 region” of the long arm of chromosome 20 contributes to the development of NIDDM, we conducted linkage studies in 29 extended Caucasian families in which many members were affected with NIDDM. A total of 498 individuals, including 159 NIDDM patients with an average age at diagnosis of 47 years, were genotyped for eight highly polymorphic microsatellite markers spanning a 31-cM region on chromosome 20q12–13.1. Using affected sib-pair analysis, we obtained evidence suggesting linkage between NIDDM and markers D20S119, D20S178, and D20S197 (allele sharing identical-by-descent [IBD], 0.56 for all three; P = 0.005, P = 0.009, and P = 0.004, respectively). Multipoint nonparametric linkage (NPL) analysis also showed evidence for linkage of NIDDM with the same three markers. The evidence for linkage was much stronger (allele sharing IBD by affected sibpairs, 0.64 [P < 0.0001]; maximum NPL score, 3.3 [P = 0.009]) in the 14 families whose average age at diagnosis of NIDDM was above the median (47 years) for all families. In these 14 families, one particular allele of the microsatellite D20S197 was transmitted from heterozygous parents to NIDDM offspring more frequently than expected (P < 0.01). This indicates that the marker allele and the disease allele are in linkage disequilibrium, implying that they are in close proximity. Consequently, the recently identified MODY1 gene (hepatocyte nuclear factor 4) is an unlikely candidate gene for NIDDM in our families, since it is located about 8 cM centromeric of D20S197. In conclusion, we have identified a new region on chromosome 20q that contains one or more NIDDM genes distinct from the recently identified MODY1 gene.

OBJECTIVES: Registration practices were evaluated as the initial phase of a validation study of the Register of the Massachusetts Commission for the Blind. METHODS: Massachusetts eye care providers were surveyed to determine factors associated with nonreporting of legal blindness to the commission. RESULTS: Among ophthalmologists, factors associated with nonreporting were small practice size and practicing for 5 years or less in Massachusetts. Among optometrists, factors included small practice size and unawareness of the Massachusetts reporting law. CONCLUSIONS: Information should be disseminated to eye care providers, legally blind patients, and the public to ensure registration and sustain it.

To evaluate familial factors in the development of diabetic nephropathy in insulin-dependent diabetes mellitus (IDDM) we examined concordance for diabetic nephropathy in families with multiple IDDM siblings. Families (n = 110) were identified through Joslin Clinic patients (probands) with a sibling having IDDM. To be eligible, the probands' and siblings' ages at IDDM diagnosis were less than 21 years, and IDDM duration was more than 15 years for probands and more than 10 years for siblings. Mean post-pubertal diabetes duration was 23 years for probands (n = 110) and 21 years for siblings (n = 125). Nephropathy history was determined by medical record review for deceased patients and those with persistent proteinuria or end-stage renal disease to ascertain the date of onset of persistent proteinuria. For patients without documented nephropathy, the albumin/creatinin ratio was measured in multiple urine samples. The cumulative incidence of persistent proteinuria according to post-pubertal duration of IDDM was determined by life-table analysis. For probands and siblings combined, the cumulative incidence of advanced diabetic nephropathy after 30 years of IDDM was 35%, but the risk in siblings varied according to the proband's renal status. The cumulative risk in siblings after 25 years of IDDM (post-puberty) was 71.5% if the proband had persistent proteinuria but only 25.4% if the proband did not (p < 0.001). A difference of nearly 50% in the risk to IDDM siblings, depending upon the IDDM proband's renal status, is consistent with a major gene effect that predisposes an individual with IDDM to develop advanced diabetic nephropathy.

The objective of this study was to determine the prevalence of stages of diabetic nephropathy, defined by the albumin/creatinine ratio (AC ratio) in repeated measurements in random urine samples. Over a 30-month interval, 1613 patients with Type I diabetes (IDDM) (aged 15 to 44 yr, IDDM duration 1 to 39 yr), and 218 healthy control subjects provided multiple urine specimens. AC ratios measured in urine samples taken 5 months apart were highly reproducible (Spearman r = 0.83). A criterion for the boundary between normoalbuminuria and microalbuminuria was obtained by searching for a cutpoint that optimized agreement between serial specimens on individuals. The result was lower in men than women: 17 as compared with 25 micrograms/mg. These two values corresponded to the 95th percentiles of the respective distributions of the AC ratio in healthy control subjects. Also these sex-specific cutpoints, when converted to albumin excretion rates, became almost equal: 30 and 31 micrograms/min. Microalbuminuria appeared early in the course of IDDM (6% of those with only 1 to 3 yr of diabetes) and then increased rapidly during two intervals, the first and third decades, before leveling off at 52%. By that time the cumulative risk of overt proteinuria had risen to 27%. Determinations of the AC ratio in random urine samples are easily obtained and are reliable indices of elevated urinary albumin excretion (microalbuminuria) in IDDM. The pattern of occurrence of microalbuminuria according to duration of IDDM suggests that there may be two subsets of diabetic nephropathy, one appearing early and the other late. Patients with microalbuminuria and 25 yr of postpubertal IDDM have low risk of progression to advanced diabetic nephropathy.

Purpose To determine prospectively whether cardiovascular autonomic neuropathy is a risk factor for proliferative diabetic retinopathy. Methods A five-year follow-up study of 88 diabetic persons was performed at a center providing primary and specialized care for diabetes. Participants were white, insulin-dependent patients with diabetes of 15 to 21 years' duration. The primary end point was the presence of proliferative diabetic retinopathy, seen either on fundus photography or ophthalmologic examination. Cardiovascular autonomic neuropathy was measured at baseline by using a standard protocol. Results Fourteen patients developed proliferative diabetic retinopathy during follow-up. One measure of cardiovascular autonomic neuropathy, the 30:15 ratio, the heart rate variation at the 30th beat compared with that at the 15th beat, was lower among patients with proliferative diabetic retinopathy (P = .0049) The risk of proliferative diabetic retinopathy in patients with an abnormal cardiovascular autonomic neuropathy index was 2.59, although the estimate was not statistically significant because of the small number of patients who developed proliferative diabetic retinopathy. Conclusions This study provides prospective evidence that cardiovascular autonomic neuropathy is associated with proliferative diabetic retinopathy. In addition to ocular determinants of proliferative diabetic retinopathy, systemic risk factors also should be considered when examining patients with diabetes mellitus.

The angiotensin I-converting enzyme (ACE) gene (17q23) is a candidate gene for essential hypertension and related diseases, but investigation of its role in human pathology is hampered by a lack of identified polymorphisms. Currently, a 287-bp insertion/deletion (I/D) RFLP in intron 16 represents the only one known. Additional polymorphisms for the ACE gene would make most families informative for linkage studies and would allow haplotypes to be assigned in association studies. To increase the information provided by the ACE gene, we used a sensitive screening technique, denaturing gradient gel electrophoresis (DGGE) blots, to identify polymorphisms and combined this with gene counting to identify haplotypes. Five independent polymorphisms, restriction fragment melting polymorphisms (RFMPs), were identified by four probes (encompassing half of the ACE cDNA) in digests produced by three restriction enzymes (DdeI, RsaI, and AluI). One RFMP has three alleles while the others have two alleles. In a sample of 67 unrelated control subjects, minor allele frequencies ranged from 0.12 to 0.49. A significant level of linkage disequilibrium was found for all pairs of markers. The four most informative RFMPs, taken in combination, define 24 potential haplotypes. Based on gene counting, 11 of the 24 are rare or nonexistent in this population, and the estimated heterozygosity of the remaining 13 haplotypes approaches 80%. Under these conditions for the ACE locus, phase-unknown genotypes could be assigned to haplotype pairs in unrelated subjects with reasonable certainty. Thus, using DGGE blot technique for identifying numerous DNA polymorphisms in a candidate locus, in combination with gene counting, one can often identify DNA haplotypes for both related and unrelated study subjects at a candidate locus. These markers in the ACE gene should be useful for clinical and epidemiologic studies of the role of ACE in human disease.

In search of genetic determinants of susceptibility to diabetic nephropathy, we examined the association between DNA sequence differences at the locus of angiotensin I-converting enzyme (ACE) and renal complications in 151 insulin-dependent diabetes mellitus (IDDM) patients with a diabetes duration of 16–21 years. This nested case-control study included 77 normoalbuminuric control subjects (albumin excretion rate

Objective To explore epidemiologic evidence for the relationship between dietary intake and glycemic control. Design/subjects We examined usual dietary intake, assessed by a food frequency questionnaire, from a random sample (n = 136) of patients who had had insulin-dependent diabetes mellitus for 15 to 21 years. Results In men, absolute intakes of energy, carbohydrate, protein, fat, and dietary fiber were positively correlated with hemoglobin A 1 (HbA 1 ) ( P r s ) were .28, .22, .28, 34, and .25, respectively. In women, the correlations were weaker and not significant; r s values were .18, .13, .17, .19, and .16, respectively. When these nutrients were expressed as a percentage of energy or as an amount per 1,000 kcal, only percentage of energy from fat showed a significant association with HbA 1 —but only in men ( r s =.23 for men and .02 for women). Adjustment for body weight, insulin dose, and physical activity using multiple regression analysis did not change the relationship between HbA 1 and intakes of energy and fat. Sucrose and alcohol intakes did not show any association with HbA 1 . Conclusions Among men with insulin-dependent diabetes mellitus, the major dietary correlates of poor glycemic control were high intake of energy and percentage of energy from fat. Further investigation is needed to confirm this relationship of energy, fat, and glycemic control.

Almost all patients with type I and many with type II diabetes develop proliferative retinopathy. This entity consists of two components: new blood vessels on the optic disc (NVD), which frequently lead to visual loss, and new blood vessels elsewhere on the retina (NVE), which do not pose such a serious threat to vision. This study examined determinants of neovascularization specifically on the optic disc in eyes with severe nonproliferative retinopathy. The study eyes were under surveillance as the untreated control eyes of participants in the Diabetic Retinopathy Study. During the 5-year follow-up period, NVE developed in almost all of the eyes, whereas the cumulative incidence of NVD in these same eyes was 64% and varied according to several factors. The risk of NVD in a study eye was increased if the contralateral treated eye had NVD rather than NVE or severe nonproliferative retinopathy (odds ratio [OR], 6.1; P.0001). It was also increased if the study eye had, at the baseline examination, soft exudates and intraretinal microvascular abnormalities (OR, 5.7; P = .002) or soft exudates alone (OR, 4.0; P = .04). Nephropathy and poor glycemic control were each associated with a two-fold increase in risk but neither was statistically significant. Eyes of individuals over 40 years of age were protected from the development of NVD (OR, 0.5; P.05). The findings of this study support the hypothesis that, in patients with diabetes, the development of NVD is determined by different factors than the development of NVE.

The goal of this review was to assess the magnitude of coronary artery disease (CAD) mortality and its determinants in insulin-dependent diabetes mellitus (IDDM) patients with persistent proteinuria. By reanalyzing data from two previously published studies of patients with nephropathy, it was found that these patients had extremely high CAD mortality rates in comparison with IDDM patients without proteinuria, but only after the age of 35 yr. In addition, the risk of CAD death was associated with high serum cholesterol levels but was unrelated to systemic blood pressure, smoking habits, and obesity. Further studies of the determinants of CAD in patients with IDDM and proteinuria are urgently needed. Except for efforts to lower serum cholesterol, it is not known whether any other measure can be undertaken to reduce the extremely high mortality due to CAD that afflicts IDDM patients with persistent proteinuria, in particular those patients whose renal failure might have been "successfully" postponed by antihypertensive therapy.

A limiting factor in the study of genetic determinants of human disorders is the availability of informative DNA markers. In this report, we describe an application of the denaturing gradient gel blot method for detecting high-frequency DNA sequence polymorphisms in the human insulin receptor locus. Using two restriction enzymes and cDNA probes for the insulin receptor, we found five DNA polymorphisms. The probe that contained exons 4-10 of the insulin receptor gene detected two two-allelic polymorphisms in HinfI digests, one at denaturant concentrations of 38%/39% and the other at 46%/48%. The probe that contained exons 14-22 detected three two-allelic polymorphisms in Sau96I digests, the first at denaturant concentrations of 34%/35%, the second at 38%/39%, and the third at 46%/47%. All these DNA polymorphisms segregated in families in a Mendelian fashion, and the allelic distribution for each of them did not deviate from Hardy-Weinberg equilibrium. The identified polymorphisms were in linkage equilibrium and provided sufficient genetic information to determine parental haplotypes at the insulin receptor locus in small two-generation families. The denaturing gradient gel blot method is a very sensitive technique for identifying sequence polymorphisms in genomic DNA; its application will facilitate the search for genes involved in the development of many inherited disorders.

To identify risk factors for the development of cardiovascular autonomic neuropathy in patients with juvenile-onset type I diabetes mellitus.Cross-sectional examination of an inception cohort 15 to 21 years after the onset of diabetes.Outpatient diabetes clinic.Seventy-nine patients with type I diabetes who experienced onset of disease before 21 years of age and who were followed for 15 to 21 years.Autonomic nerve function was evaluated in all patients using deep breathing and tilt tests. On the basis of these tests, an index of cardiovascular autonomic neuropathy was derived and patients were classified as having intact, mildly impaired, or significantly impaired autonomic function.The group with significantly impaired function had a higher mean hemoglobin A1 at the time of examination than the group without impairment, yet the groups did not differ regarding glycemic control during the first decade of diabetes. The HLA-DR3/4 phenotype was present in more than 50% of the patients with significant autonomic dysfunction and conferred relative odds of 6.2 (95% CI, 1.7 to 23.3) for the development of autonomic neuropathy when compared with other HLA-DR phenotypes. Sex, percent ideal body weight, and smoking did not have a statistically significant effect on the development of autonomic neuropathy.The development of cardiovascular autonomic neuropathy in type I diabetes mellitus is strongly associated with the HLA-DR3/4 phenotype. Thus, genetic predisposition may play an important role in the development of this complication.

Predisposition to hypertension: Risk factor for nephropathy and hypertension in IDDM. Less than a quarter of the patients with juvenile-onset IDDM develop diabetic nephropathy during the first 20 years of diabetes. To study the determinants of this complication, we selected patients who had come with newly diagnosed IDDM to the Joslin Clinic between 1967 to 1972, and we examined them in 1986 to 1988, that is, 15 to 21 years after onset of diabetes. Using a case control design we compared three groups of cases, that is, advanced nephropathy (TV = 43), only microalbuminuria (TV = 41), and hypertension alone (N = 17), with a group of controls who remained normoalbuminuric and nor-motensive despite the long duration of IDDM (N = 61). In comparison with controls, patients with advanced nephropathy had more parents with hypertension (odds ratio 3.8), higher V max values of Na/Li countertransport in red blood cells (odds ratio 10.0 for the highest tertile), and higher mean arterial pressure during adolescence and early adulthood (odds ratio 3.1 for those above the median). They also had significantly poorer glycemie control during their first 12 years of diabetes. Patients with hypertension alone were similar to those with advanced nephropathy with regard to markers of predisposition to hypertension but differed from them with regard to glycemie control, having the best glycemie control of all the study groups. Patients who developed only microalbuminuria during 15 to 21 years of IDDM (some of whom will progress to overt proteinuria later) did not differ significantly from controls with regard to predisposition to hypertension. In conclusion, predisposition to hypertension is a major risk factor for the development of advanced diabetic nephropathy and essential hypertension during the first 20 years of IDDM.

White diabetic patients are at high risk of developing coronary artery disease (CAD). The natural history of CAD in insulin-dependent (ID) and noninsulin-dependent (NID) diabetes mellitus (DM) is reviewed to gain insight into the mechanisms responsible for the development of premature or accelerated atherosclerosis in diabetic patients. In both IDDM and NIDDM, the risk of CAD increases with lengthening duration of diabetes; the risk, however, does not grow as a constant multiple of the nondiabetic risk of CAD, suggesting that the cumulative exposure to diabetes plays a significant role as a risk factor for CAD only in a subset of patients. This is consistent with the hypothesis that the diabetic milieu has an impact on the progression of atherosclerotic lesions but not on their initiation. This hypothesis is corroborated further by the observation that CAD does not occur in diabetic patients in populations with a low risk of CAD among nondiabetic patients. The component of the diabetic milieu responsible for promotion of atherosclerotic lesions is unknown. There is evidence, however, of a direct or indirect role of hyperinsulinemia in this process.

We thank Bakker et al. (1) for their concern about the factors affecting risk of end-stage renal disease (ESRD) and their appreciation of the issues relating to survival bias. In the original report (2), the Cox proportional hazards regression model of ESRD risk according to carnosinase genotype was adjusted for sex, duration of diabetes, age at examination, age at diagnosis of diabetes, and use of ACE inhibitor treatment. As reported, there was no effect of genotype on the risk. …

Context A common allele on chromosome 9p21 has been repeatedly associated with increased risk of coronary artery disease (CAD) in the general population. However, the magnitude of this effect in the population with diabetes has not been well characterized. Objective To examine the association of the 9p21 variant with CAD in individuals with type 2 diabetes and evaluate its interaction with poor glycemic control. Design, Setting, and Participants (1) Case-control study of 734 type 2 diabetes patients (322 with angiographically diagnosed CAD and 412 with no evidence of CAD) who were recruited between 2001 and 2006 at the Joslin Clinic, Beth Israel Deaconess Medical Center; and (2) independent cohort study of 475 type 2 diabetes patients from the Joslin Clinic whose survival status was monitored from their recruitment between 1993 and 1996 until December 31, 2004. Participants for both studies were genotyped for a representative single-nucleotide polymorphism at 9p21 (rs2383206) and characterized for their long-term glycemic control by averaging multiple hemoglobin A 1c (HbA 1c ) measurements taken in the years before study entry. Main Outcome Measures For the case-control study, association between single-nucleotide polymorphism rs2383206 and CAD defined as angiographically documented stenosis greater than 50% in a major coronary artery or a main branch thereof was assessed and for the cohort study, cumulative 10-year mortality was documented. Results Individuals who were homozygous for the risk allele were significantly more frequent among case than control participants (42.3% vs 28.9P = .0002). This association was unaffected by adjustment for cardiovascular risk factors, but the effect of the risk genotype was significantly magnified (adjusted P for interaction = .048) in the presence of poor glycemic control (worst tertile of the distribution of HbA 1c at examination). Relative to the CAD risk for patients with neither a 9p21 risk allele nor poor glycemic control, the CAD odds for participants having 2 risk alleles but not poor glycemic control were increased 2-fold (odds ratio [OR], 1.99; 95% confidence interval [CI], 1.17-3.41), whereas the odds for study participants with the same genotype and with poor glycemic control were increased 4-fold (OR, 4.27; 95% CI, 2.26-8.01). The interaction was stronger (adjusted P = .005) when a measure of long-term glycemic control (7-year average rather than most recent HbA 1c ) was used with ORs of 7.83 (95% CI, 3.49-17.6) for participants having 2 risk alleles and a history of poor glycemia and 1.54 (95% CI, 0.72-3.30) for participants with the same genotype but without this exposure. A similar interaction between 9p21 variant and poor glycemic control was observed with respect to cumulative 10-year mortality in the cohort study (43.6% in patients with 2 risk alleles and poor glycemic control, 23.1% in individuals with only the 2 risk alleles, 30.0% in individuals with only poor glycemic control, and 31.6% in individuals with neither factor, P for interaction, = .036). Conclusion In this study population, the CAD risk associated with the 9p21 variant was increased in the presence of poor glycemic control in type 2 diabetes.

To the Editor: Since we found evidence of a threshold for the effect of hyperglycemia on the risk of microalbuminuria in patients with insulin-dependent diabetes mellitus (IDDM),1 we used published data to examine whether there was a similar pattern for diabetic retinopathy. In the Diabetes Control and Complications Trial (DCCT),2 the incidence of the development or progression of retinopathy in the group assigned to intensive treatment fluctuated around 2.0 per 100 person-years in the nine deciles of the hemoglobin A1c distribution below 8.5 percent and rose to 7.0 per 100 person-years in the highest decile, consistent with a threshold . . .

Genetic factors contribute significantly to the development of diabetic nephropathy in patients with insulin-dependent diabetes mellitus. This report discusses some models of diabetic nephropathy that incorporate genetic susceptibility and presents strategies for identifying the responsible genes. To identify variation at a locus, newly developed methods are discussed that employ denaturing gradient gel electrophoresis to study sequence differences in both polymerase chain reaction-amplified DNA fragments and genomic DNA. These techniques are illustrated with studies of the angiotensinogen gene and the insulin receptor gene. In preliminary data from a comparison between individuals with and without diabetic nephropathy, no DNA sequence difference in that part of the angiotensinogen gene that codes for angiotensin I was found. However, with a probe corresponding to exons 7 and 8 of the insulin receptor gene and denaturing gradient gel electrophoresis of Rsal digestions of genomic DNA, different distributions of a DNA polymorphism were found in patients with fast as compared with slowly progressing nephropathy. The interpretation of this finding and the need for further studies are discussed. In conclusion, the advent of methods of molecular genetics makes possible studies on genetic determinants of diabetic nephropathy. However, more clinical and epidemiologic data are needed to find out how many genes are involved and how they interact with exposure to diabetes. Foremost, DNA from families with two or more siblings with diabetic nephropathy must be collected to permit the necessary genetic studies.

Studies have demonstrated that "antihypertensive" treatment with angiotensin-converting enzyme inhibitors (ACEI) may retard the progress of nephropathy in patients with insulin-dependent diabetes mellitus. To obtain an indication of the potential effect of ACEI treatment and as a guide to future research, the effects of screening and early ACEI treatment programs were estimated using cost-effectiveness models. The preliminary analysis suggests that the early treatment of insulin-dependent diabetes mellitus patients with ACEI is likely to be a very cost-effective use of health care resources. The cost-effectiveness ratio for screening and treatment at the stage of microalbuminuria ($7,900 to $16,500 per year of life saved) compares favorably with those of other medical life-saving interventions. Less-aggressive programs (screening followed by treatment at the stage of proteinuria) would improve life expectancy to a lesser extent but could save net health care costs as well as years of life. Although more exact and comprehensive cost-effectiveness analysis must await clinical trials, these illustrative results demonstrate the range of cost-effectiveness that can be expected from these programs and identify data needed for more decisive policy conclusions.

In Reply. — Parving's letter raises some important points that were not addressed in our article 1 regarding excess mortality associated with diuretic therapy in patients with diabetes. First, we did not cite the European "trials" 2 evaluating the effects of antihypertensive drugs for the reason that they were conducted among young patients, and it is well known that few of these patients die before age 35 years. When they do, it is mainly due to uremia. The patients in our study were over 35 years of age and were, therefore, in an age group where the risk of cardiovascular death is as great as the risk of uremic death. 3 The important clinical question is whether it is possible to protect the kidney with antihypertensive treatment and, at the same time, reduce the risk of cardiovascular death. There has been no randomized clinical trial addressing this question. Second, the

DIABETES arises through two etiologically distinct routes.1 Type I diabetes is the result of immunologically mediated destruction of the pancreatic beta cells, and usually requires replacement ther...

Prevalences of hypertension and orthostatic hypotension and their relationship to the microvascular complications of diabetes were assessed in 702 individuals aged 18–74 years, who had been selected as a representative sample of surviving patients with diabetes diagnosed at the Joslin Clinic between 1939 and 1965. In diabetes of short, long and very long duration, hypertension was 1.7, 1.9 and 2.1 times more frequent, respectively, than in the white U.S. population, regardless of gender. The excess frequency of hypertension in short duration diabetes suggests that some etiologic factor is shared by both conditions, while the magnification of the excess with increasing duration could be explained by an effect of diabetes on the kidney. Hypertension without accompanying proteinuria was not associated with retinopathy. Orthostatic hypotension was observed in 12% of the males and 13% of the females. The magnitude of the fall in systolic blood pressure was correlated with age, postprandial blood glucose, supine diastolic blood pressure, and the presence of retinopathy. Patients with proliferative retinopathy had the largest fall in systolic blood pressure.

The relationship between HLA-DR phenotype, refractive error, and risk of both nonproliferative and proliferative diabetic retinopathy was studied among 227 insulin-dependent diabetics participating in a case-control study of diabetic retinopathy. In the absence of myopia, risk of both proliferative and nonproliferative diabetic retinopathy was increased for HLA-DR phenotypes 4/0, 3/0, and X/X (HLA susceptible) compared to HLA-DR phenotypes 3/4, 3/X, and 4/X (HLA nonsusceptible). Odds ratios equalled 11.8 and 7.4 respectively. In the presence of myopia this increased risk associated with HLA status was abolished. Myopia decreased the risk of proliferative and nonproliferative diabetic retinopathy among HLA-susceptible individuals, odds ratio equalled .09 and .09, but had no protective influence among HLA-nonsusceptible individuals.

Between May, 1963 and December, 1966, 17 medical centers cooperated in two separate but integrated therapeutic trials of primary lung cancer. One study was of patients with lesions considered operable at the time of diagnosis, and the other of patients with initially inoperable cancer but who were considered potentially operable after radiotherapy. Patients operable at the time of diagnosis were randomly assigned to receive either immediate surgery (278 patients) or preoperative radiotherapy followed by surgery (290 patients). All but one were followed until death or 5 years of survival. Survival to each anniversary after randomization was almost identical for the two groups. At 5 years the survival rate was 14% after preoperative radiotherapy and 16% after immediate surgery. On the basis of the small standard error of the difference between these survival rates, a large advantage or a large disadvantage for preoperative radiotherapy is unlikely. Recurrence of cancer either locally or as distant metastasis was also similar in the two groups. Postoperative mortality was estimated to be 11% in the immediate surgery group, but cannot be estimated in a comparable fashion for the irradiated group. Certain postoperative complications were more frequent in the irradiated group, but survival during the first year was not affected. Out of 425 patients initially considered to be inoperable, 152 were considered resectable after radiotherapy. These patients were randomly assigned to have either a thoracotomy and resection of their cancer if possible (78 patients) or no surgery (74 patients). Survival to each anniversary after randomization was very similar. After 5 years the survival rate was 8% for the group assigned to surgery and 6% for the group assigned to no surgery. The difference has a standard error of 4%.