Diabetologia

, Volume 52, Issue 9, pp 1745–1754

Insulin glargine use and short-term incidence of malignancies—a population-based follow-up study in Sweden

  • J. M. Jonasson
  • R. Ljung
  • M. Talbäck
  • B. Haglund
  • S. Gudbjörnsdòttir
  • G. Steineck
Article

DOI: 10.1007/s00125-009-1444-2

Cite this article as:
Jonasson, J.M., Ljung, R., Talbäck, M. et al. Diabetologia (2009) 52: 1745. doi:10.1007/s00125-009-1444-2

Abstract

Aims/hypothesis

In the light of a report suggesting that insulin glargine may increase cancer occurrence, the EASD asked us to perform this study.

Methods

We followed 114,841 individuals who had a prescription dispensed for insulin between 1 July and 31 December 2005. From 1 January 2006 to 31 December 2007, we noted the occurrence of malignancies. Seven different nationwide registers were used to obtain information on insulin exposure, outcome and possible confounders; these were linked using the unique personal identity number assigned to every Swedish resident.

Results

After adjustment for age and, when appropriate, sex, users of insulin glargine alone (no other types of insulin), compared with users of types of insulin other than insulin glargine, had an RR of 1.99 (95% CI 1.31–3.03) for breast cancer, 0.93 (95% CI 0.61–1.40) for gastrointestinal cancer, 1.27 (95% CI 0.89–1.82) for prostate cancer and 1.07 (95% CI 0.91–1.27) for any type of malignancy. Adjustment for age, smoking, BMI, age at onset of diabetes, age at birth of first child, cardiovascular disease and oestrogen use gave an RR for breast cancer of 1.97 (95% CI 1.29–3.00). The 95% CIs crossed 1.0 for the RR calculated in all analyses of users of insulin glargine in combination with other types of insulin.

Conclusions/interpretation

In Sweden, during 2006 and 2007, women using insulin glargine alone (no other types of insulin) had an increased incidence rate of breast cancer as compared with women using types of insulin other than insulin glargine. This result may be due to a random fluctuation; the possibilities for examining validity are limited, and no statistically significant results were obtained for any other individual cancer site or for the outcome ‘all malignancies’. No definitive conclusions regarding a possible causal relationship between insulin glargine use and the occurrence of malignancies can be drawn from the results of this study.

Keywords

Cancer incidence rate Diabetes mellitus Insulin Insulin glargine 

Abbreviations

ATC

Anatomical Therapeutic Chemical

ICD

International Classification of Diseases

Introduction

Since the introduction of non-human insulins, in vitro data indicating proliferative and anti-apoptotic activity have raised concern [1, 2]. Sweden offers favourable conditions for the evaluation of adverse effects of drug use in the general population: the Prescribed Drug Register gives near-complete coverage of individuals for whom insulin has been prescribed, and the Cancer Register contains, by and large, all new cases of cancer. A personal identity number, unique for each Swedish resident, allows the information from different registers on a particular person to be linked [3]. In the light of a report suggesting that insulin glargine (A21Gly,B31Arg,B32Arg human insulin) may increase cancer occurrence [4], the EASD asked us to perform the present study. We have investigated the incidence rates of breast cancer, gastrointestinal cancer, prostate cancer and any type of malignancy associated with the use of insulin glargine and compared these with the rates associated with the use of other types of insulin.

Methods

We used the unique personal identity number assigned to each Swedish resident (about 9 million in total) to link together information from seven population-based registers (Fig. 1) [3].
Fig. 1

The registers used, and the variables retrieved from each register

The Prescribed Drug Register, the Cancer Register, and the Causes of Death Register, all maintained by the National Board of Health and Welfare were used to obtain information on targeted person-time and outcome. We retrieved variables reflecting potential confounding factors [5] from the Swedish National Diabetes Register (maintained by the local health authorities), the Prescribed Drug Register, the National Patient Register [6], and the Medical Birth Register [7] (all maintained by the National Board of Health and Welfare). Information on educational level was extracted from the National Education Register [8], which is maintained by Statistics Sweden. By law, local health authorities must report all new cases of cancer to the Cancer Register, all in-patient information to the National Patient Register, all births to the Medical Birth Register, and all deaths to the Causes of Death Register [9].

The Swedish Prescribed Drug Register contains details of all the prescriptions dispensed in Sweden [10]. Updated monthly, there are presently around 100 million prescriptions dispensed each year. Before 1 July 2005, the personal identity number was not recorded in the register. We therefore had to start recruiting subjects for observation from 1 July 2005.

The Swedish Cancer Register was set up in 1958, and since that time every clinician, pathologist and cytologist in Sweden must notify the National Board of Health and Welfare of each person who has been diagnosed with a new primary malignancy. The Cancer Register includes primary malignancies and certain benign tumours and precancerous lesions [11]. A comparison with death certificates revealed the rate of non-reporting to the National Cancer Registry to be less than 2% during the late 1970s [12], and in a comparison with the National Patient Register, the rate of non-reporting to the National Cancer Registry was estimated to be 3.7% in 1998 [13].

Launched as a quality assurance register in 1996, the National Diabetes Register includes clinically relevant information [14]. Trained physicians and nurses report the data collected during visits to hospital outpatient clinics and primary healthcare centres via the internet or via clinical record databases. The National Education Register is updated annually with information on the highest formal education achieved.

Ethical considerations

The Swedish National Board of Health and Welfare is a government agency and may, in accordance with Swedish law, use population-based registers to follow and analyse health and social conditions among the general population. Data were made available to us in such a way that individuals could not be identified.

Targeted person-time

We have studied all 114,841 individuals who were aged 35–84 years old at the end of 2005, had at least one prescription dispensed for insulin (Anatomical Therapeutic Chemical [ATC] code A10A) between 1 July and 31 December 2005, and who were alive at the start of follow-up (1 January 2006). We studied first diagnosis of a primary malignancy as an outcome measure, excluding individuals who received this diagnosis at any time between 1 January 1958 and 31 December 2005. That is, a subject with a record of having been diagnosed with any type of malignancy was excluded from the analyses of outcomes including ‘all malignancies’, and men who had a record of having been diagnosed with prostate cancer before 2006 were excluded from the analyses of the outcome ‘prostate cancer’. We followed the subjects from 1 January 2006 to 31 December 2007. The number of person-years of follow-up for each individual was from 1 January 2006 to death or loss to follow-up (censorship) or the outcome being analysed or study end. Consequently, the number of observed person-years varies according to the particular type of malignancy studied.

Categories of insulin use

Information on exposure to insulin and analogues was obtained from prescriptions dispensed between 1 July and 31 December 2005. Individuals registered as having had at least one prescription dispensed for insulin glargine (ATC code A10AE04), but no prescriptions dispensed for other types of insulin (ATC code A10A) were classified as using insulin glargine alone (no other types of insulin). Having a prescription dispensed for both insulin glargine and another type of insulin classified the individual as a user of insulin glargine and other types of insulin. Having a prescription dispensed for insulin but not for insulin glargine, classified the individual as a user of types of insulin other than insulin glargine.

Outcomes

We studied five malignancy outcomes; death from any cause and acute myocardial infarction were also used as endpoints. Following the routines for the Cancer Register, the outcome ‘all malignancies’ included a carcinoid tumour, a granulosa cell tumour, a thymoma, an adamantinoma, a chordoma, a transitional cell papilloma of the urinary tract, a hormonally active tumour from at least one endocrine gland (except the thyroid), an enterochromaffin or a neuroendocrine tumour. We also included precancerous lesions, including gastrointestinal polyps with suspected malignancy, bronchial adenomas, carcinoma in situ of the breast, fibro-adenoma with suspected malignancy, adenoma phyllodes, precancerous endometrial lesions, hydatidiform moles of placental tissue, ovarian cystadenomas of borderline type, histologically benign tumours of the central nervous system and meningomas. In situ cases were included with malignant tumours as a second outcome. We defined outcomes for three different anatomical areas: ‘breast cancer’ (International Classification of Diseases, 10th revision [ICD-10] code C50), ‘prostate cancer’ (ICD-10 code C61) and ‘gastrointestinal cancer’ (ICD-10: codes C16–C20). For these three areas we only included tumours that were histopathologically classified as adenocarcinoma (WHO/HS/CANC/24.1 histology code 096).

Individuals registered with any type of malignancy between 1 January 1958 and 31 December 2005 were excluded from the analysis when ‘any type of maligancy’ was the outcome. When studying breast cancer among women we excluded those who had previously been diagnosed with breast cancer; similar exclusions were made when studying prostate cancer and gastrointestinal cancer. We retrieved the date of death from the Causes of Death register. Subjects who were not registered as dead and who did not have a prescription dispensed for any drug in 2008 were classified as having been lost to follow-up. The date for loss to follow-up was set at 90 days after the last date for a dispensed prescription.

Variables reflecting potential confounding factors

Sex and age were retrieved from the Prescribed Drug Register. We obtained data on age at onset of diabetes from the National Diabetes Register or estimated it from the time for first admission to hospital care with diabetes as the main diagnosis (ICD-8 code 250; ICD-9 code 250; ICD-10 codes E10–E14) from data in the Patient Register for 1969 to 2005. An age at onset of diabetes of less than 30 years, as recorded in the National Diabetes Register (primary choice) or by data from the Patient Register (secondary choice), defined an individual as having type 1 diabetes; an age at onset above 30 years defined an individual as having type 2 diabetes; the absence of information on age at onset defined the individual as having missing information on type of diabetes. The highest BMI reported to the National Diabetes Register from 2003 to 2005 for each individual was used as the value for BMI. We retrieved information on smoking habits from 2003 to 2005 from the National Diabetes Register. Anyone who reported smoking during 2005 was classified as a current smoker. Anyone who reported not smoking in 2005 but reported smoking in 2003 or 2004 was classified as a former smoker. Anyone who reported not smoking in 2003, 2004 and 2005 was classified as a non-smoker. A record of a prescription dispensed for an oestrogen or for metformin in the Prescribed Drug Register from 1 July to 31 December 2005 defined oestrogen and metformin use, respectively.

A record of at least one hospital admission with a main diagnosis of any cardiovascular disease (ICD-10 codes I00–I99) in the National Patient Register, during the period 1 July 2004 to 30 June 2005 (i.e. 1 year prior to definition of exposure), classified the individual as having cardiovascular disease.

Educational level refers to the highest attained educational level at the end of 2005. Educational level was classified into the following three categories, representing distinct levels in the Swedish educational system: (1) 9 years or fewer of schooling, equivalent to elementary school or less; (2) 10–12 years of schooling, equivalent to secondary school; and (3) more than 12 years, equivalent to university. Age at birth of first child (women only) was categorised into no children, <30 years, ≥30 years, and missing information. A large group of women (49%), mainly the older women in the study population, had information missing on childbearing.

Statistical methods

As a measure of the relative occurrence of malignancies, we used the incidence rate ratio. For example, we calculated the incidence rate of having been diagnosed with any type of malignancy among users of insulin glargine alone and compared this with the incidence rate among users of other types of insulin. We cite this measure of relative occurrence, the incidence rate ratio, as a relative risk. Poisson regression analyses were used to evaluate the association between the three groups of insulin users and malignancy outcome. These models were fitted with the logarithm of observed person-years as the offset and they also provided 95% CIs of the incidence rate ratio. When adjusting for potential confounding factors, we categorised the numeric variables as presented in Tables 1 and 4, and in order to avoid a substantial reduction of the number of subjects, we accepted ‘missing value’ as a single category in our main analyses. The Genmod procedure in the SAS statistical software package (SAS Institute, Cary, NC, USA) was used for the calculations.
Table 1

Baseline characteristics of the subjects

Characteristic

Insulin glargine alone

Insulin glargine and other insulins

Insulins other than insulin glargine

Total

n

%

n

%

n

%

n

%

Total

5,970

 

20,316

 

88,555

 

114,841

 

Type of diabetes

        

Type 1 diabetes

564

9.4

7,903

38.9

9,049

10.2

17,516

15.3

Type 2 diabetes

3,750

62.8

10,168

50.0

56,638

64.0

70,556

61.4

Missing

1,656

27.7

2,245

11.1

22,868

25.8

26,769

23.3

Sex

        

Male

3,273

54.8

11,427

56.2

50,403

56.9

65,103

56.7

Female

2,697

45.2

8,889

43.8

38,152

43.1

49,738

43.3

Age at baseline (years)

        

35–54

1,398

23.4

9,198

45.3

17,144

19.4

27,740

24.2

55–64

1,874

31.4

5,922

29.1

22,697

25.6

30,493

26.6

65–74

1,489

24.9

3,557

17.5

25,668

29.0

30,714

26.7

75–84

1,209

20.3

1,639

8.1

23,046

26.0

25,894

22.5

Age at onset of diabetes (years)

        

0–14

236

4.0

3,386

16.7

3,602

4.1

7,224

6.3

15–29

328

5.5

4,517

22.2

5,447

6.2

10,292

9.0

30–44

937

15.7

5,125

25.2

13,412

15.1

19,474

17.0

45–54

1,206

20.2

2,914

14.5

17,272

19.5

21,419

18.7

55–64

1,057

17.7

1,586

7.8

16,488

18.6

19,131

16.7

65–84

550

9.2

516

2.5

9,466

10.7

10,532

9.2

Missing

1,656

27.7

2,245

11.1

22,868

25.8

26,769

23.3

BMI (kg/m2)

        

15≤BMI<25

660

11.1

5,080

25.0

9,154

10.3

14,894

13.0

25≤BMI<30

1,094

18.3

5,626

27.7

16,878

19.1

23,598

20.5

30≤BMI<35

686

11.5

2,242

11.0

11,320

12.8

14,248

12.4

35≤BMI≤50

374

6.3

821

4.0

5,898

6.7

7,093

6.2

Missing

3,156

52.9

6,547

32.2

45,305

51.2

55,008

47.9

Smoking

        

Non-smokera

2,287

38.3

11,697

57.6

36,891

41.7

50,875

44.3

Former smokerb

223

3.7

953

4.7

2,825

3.2

4,001

3.5

Current smokerc

346

5.8

1,515

7.5

4,430

5.0

6,291

5.5

Missing

3,114

52.2

6,151

30.3

44,409

50.1

53,674

46.7

aDefined as a person who reported not smoking in 2003, 2004 and 2005

bDefined as a person who reported smoking in 2003 or 2004, but reported not smoking in 2005

cDefined as a person who reported smoking in 2005

Results

Table 1 shows the characteristics of the study subjects at baseline. Of the 114,841 subjects followed, 5,970 (5.2%) were classified as users of insulin glargine alone, 20,316 (17.7%) were classified as users of insulin glargine in combination with other types of insulin, and 88,555 (77.1%) were classified as users of types of insulin other than insulin glargine. The majority (90.6%) of users of insulin glargine alone were classified as having type 2 diabetes or having information missing on type of diabetes. The same was true of users of types of insulin other than insulin glargine: a high percentage (89.8%) were classified as having type 2 diabetes or having information missing on type of diabetes. Compared with the other two groups, the group of users of insulin glargine in combination with other types of insulin had a lower mean age and a lower percentage (61.1%) were classified as having type 2 diabetes or having information missing on type of diabetes.

The 95% CIs of the adjusted RRs included 1.0 for malignancy outcomes other than breast cancer (Tables 2 and 3). The RR for breast cancer in women who used insulin glargine alone compared with those who used types of insulin other than insulin glargine was 1.91 (95% CI 1.25–2.89) when not adjusted, 1.99 (95% CI 1.31–3.03) when adjusted for age, and 1.97 (1.30–3.00) when adjusted for several variables (Table 2). The corresponding figures for women who used insulin glargine in combination with other types of insulin compared with those who used types of insulin other than insulin glargine were 0.92 (0.66–1.29), 1.10 (0.77–1.56) and 1.17 (0.81–1.68) (Table 3).
Table 2

Incidence rate and incidence rate ratio for cancer for users of insulin glargine alone compared with users of types of insulin other than insulin glargine

Cancer outcome

Types of insulin other than insulin glargine (reference category)

Insulin glargine alone

Incidence rate ratioa (95% CI)

No. of cases

Person-years

Incidence rate (per 1,000 person-years)

No. of cases

Person-years

Incidence rate (per 1,000 person-years)

Unadjusted

Adjusted for

Age and, when appropriate, sex

See footnote

All malignant tumours

2,199

148,804

14.8

149

10,323

14.4

0.98 (0.83–1.15)

1.07 (0.91–1.27)

1.06 (0.90–1.25)b

All malignant tumours and in situ tumours

2,353

144,499

16.3

156

10,009

15.6

0.96 (0.81–1.13)

1.05 (0.90–1.24)

1.04 (0.88–1.22)b

Breast cancer (women)

183

69,358

2.6

25

4,974

5.0

1.91 (1.25–2.89)

1.99 (1.31–3.03)

1.97 (1.30–3.00)c

Prostate cancer

432

91,845

4.7

32

6,037

5.3

1.13 (0.79–1.61)

1.27 (0.89–1.82)

1.26 (0.88–1.80)d

Gastrointestinal cancer

430

165,152

2.6

24

11,273

2.1

0.82 (0.54–1.23)

0.93 (0.61–1.40)

0.91 (0.61–1.38)b

aIncidence rate ratio is referred to as the RR in the text

bAdjusted for age, sex, BMI, smoking, age at onset of diabetes, and cardiovascular disease

cAdjusted for age, BMI, smoking, age at onset of diabetes, cardiovascular disease, and age at birth of first child

dAdjusted for age, BMI, smoking, age at onset of diabetes, and cardiovascular disease

Table 3

Incidence rate and incidence rate ratio for cancer for users of insulin glargine in combination with other insulins compared with users of types of insulin other than insulin glargine

Cancer outcome

Types of insulin other than insulin glargine (reference category)

Insulin glargine in combination with other insulins

Incidence rate ratioa (95% CI)

No. of cases

Person-years

Incidence rate (per 1,000 person-years)

No. of cases

Person-years

Incidence rate (per 1,000 person-years)

Unadjusted

Adjusted for

Age and sex

See footnote

All malignant tumours

2,199

148,804

14.8

341

36,892

9.2

0.63 (0.56–0.70)

0.95 (0.85–1.07)

1.02 (0.91–1.15)b

All malignant tumours and in situ tumours

2,353

144,499

16.3

360

35,902

10.0

0.62 (0.55–0.69)

0.94 (0.84–1.06)

1.00 (0.89–1.13)b

Breast cancer (women)

183

69,358

2.6

41

16,882

2.4

0.92 (0.66–1.29)

1.10 (0.77–1.56)

1.17 (0.81–1.68)c

Prostate cancer

432

91,845

4.7

58

21,862

2.7

0.56 (0.43–0.74)

0.98 (0.74–1.30)

1.05 (0.79–1.40)d

Gastrointestinal cancer

430

165,152

2.6

50

39,305

1.3

0.49 (0.36–0.65)

0.80 (0.59–1.08)

0.82 (0.61–1.12)b

aIncidence rate ratio is referred to as the RR in the text

bAdjusted for age, sex, BMI, smoking, age at onset of diabetes, and cardiovascular disease

cAdjusted for age, BMI, smoking, age at onset of diabetes, cardiovascular disease, and age at birth of first child

dAdjusted for age, BMI, smoking, age at onset of diabetes, and cardiovascular disease

As a consequence of these results for breast cancer we then used alternative specifications for the models for breast cancer, e.g. including age as continuous variable, deleting observations with missing values and replacing the assumption of Poisson distribution with negative binominal distribution. None of these alternative specifications had a substantial effect on the results. We also performed additional analyses restricted to women. The demographic data for those women who were not diagnosed with breast cancer from 1958 to 2005 are shown in Table 4; a pattern similar to that shown in Table 1 can be seen. We examined all available variables that may reflect one (or several) risk factor for breast cancer. The adjusted RRs varied between 1.96 and 2.00 for women using insulin glargine alone compared with those using types of insulin other than insulin glargine (Table 5).
Table 4

Baseline characteristics of women in the analyses of breast cancer

Characteristic

Insulin use during 2005

Total

Insulin glargine alone

Insulin glargine in combination with other insulins

Types of insulin other than insulin glargine

n

%

n

%

n

%

n

%

Total

2,595

 

8,649

 

36,532

 

47,776

100.0

Type of diabetes

        

 Type 1 diabetes

298

11.5

3,578

41.4

4,081

11.2

7,957

16.7

 Type 2 diabetes

1,566

60.3

4,148

48.0

22,831

62.5

28,545

59.7

 Missing

731

28.2

923

10.7

9,620

26.3

11,274

23.6

Age at baseline (years)

        

 35–54

646

24.9

3,856

44.6

6,735

18.4

11,237

23.5

 55–64

691

26.6

2,371

27.4

8,089

22.1

11,151

23.3

 65–74

633

24.4

1,581

18.3

10,416

28.5

12,630

26.4

 75–84

625

24.1

841

9.7

11,292

30.9

12,758

26.7

Age at onset of diabetes (years)

        

 0–14

136

5.2

1,678

19.4

1,772

4.9

3,586

7.5

 15–29

162

6.2

1,900

22.0

2,309

6.3

4,371

9.1

 30–44

397

15.3

1,937

22.4

4,939

13.5

7,273

15.2

 45–54

438

16.9

1,166

13.5

6,444

17.6

8,048

16.8

 55–64

452

17.4

749

8.7

6,934

19.0

8,135

17.0

 65–84

279

10.8

296

3.4

4,514

12.4

5,089

10.7

 Missing

731

28.2

923

10.7

9,620

26.3

11,274

23.6

BMI (kg/m2)

        

 15≤ BMI <25

321

12.4

2,482

28.7

3,931

10.8

6,734

14.1

 25≤ BMI <30

422

16.3

1,972

22.8

5,728

15.7

8,122

17.0

 30≤ BMI <35

275

10.6

981

11.3

4,703

12.9

5,959

12.5

 35≤ BMI ≤50

186

7.2

435

5.0

3,189

8.7

3,810

8.0

 Missing

1,391

53.6

2,779

32.1

18,981

52.0

23,151

48.5

Smoking

        

 Non-smokera

1,010

38.9

4,928

57.0

15,183

41.6

21,121

44.2

 Former smokerb

88

3.4

422

4.9

1,103

3.0

1,613

3.4

 Current smokerc

148

5.7

735

8.5

1,763

4.8

2,646

5.5

 Missing

1,349

52.0

2,564

29.6

18,483

50.6

22,396

46.9

Education

        

 Elementary school

865

33.3

2,218

25.6

14,229

38.9

17,312

36.2

 Secondary school

953

36.7

3,955

45.7

11,969

32.8

16,877

35.3

 University

404

15.6

2,112

24.4

4,189

11.5

6,705

14.0

 Missing

373

14.4

364

4.2

6,145

16.8

6,882

14.4

Age at birth of first child

        

 No children

585

22.5

2,267

26.2

7,053

19.3

9,905

20.7

 <30 years

687

26.5

3,370

39.0

7,448

20.4

11,505

24.1

 ≥30 years

155

6.0

864

10.0

1,805

4.9

2,824

5.9

 Missing

1,168

45.0

2,148

24.8

20,226

55.4

23,542

49.3

Oestrogen use

        

 Yes

375

14.5

1,222

14.1

5,093

13.9

6,690

14.0

 No

2,220

85.5

7,427

85.9

31,439

86.1

41,086

86.0

Metformin use

        

 Yes

1,278

49.2

1,237

14.3

14,214

38.9

16,729

35.0

 No

1,317

50.8

7,412

85.7

22,318

61.1

31,047

65.0

aDefined as a person who reported not smoking in 2003, 2004 and 2005

bDefined as a person who reported smoking in 2003 or 2004, but reported not smoking in 2005

cDefined as a person who reported smoking in 2005

Table 5

Incidence rate ratio for breast cancer among women using insulin glargine alone or using insulin glargine in combination with other types of insulin, compared with women using types of insulin other than insulin glargine

Model

Incidence rate ratioa (95% CI)

Insulin glargine alone

Insulin glargine in combination with other types of insulin

Unadjusted

1.91 (1.25–2.89)

0.92 (0.66–1.29)

Adjusted for

  

Age

1.99 (1.31–3.03)

1.10 (0.77–1.56)

  Age and metformin

1.98 (1.30–3.01)

1.11 (0.78–1.59)

  Age and oestrogen

1.99 (1.31–3.02)

1.08 (0.76–1.54)

  Age and BMI

1.97 (1.31–3.00)

1.16 (0.81–1.65)

  Age and smoking

1.97 (1.31–3.00)

1.16 (0.81–1.65)

  Age and age at onset of diabetes

1.96 (1.29–2.97)

1.16 (0.81–1.65)

  Age and cardiovascular disease

2.00 (1.31–3.04)

1.10 (0.77–1.56)

  Age and age at birth of first child

2.00 (1.31–3.04)

1.11 (0.77–1.56)

  Multiple variablesb

1.97 (1.29–3.00)

1.15 (0.80–1.65)

Restricted to age 35–79 years and adjusted for

  

  Multiple variablesb

1.98 (1.25–3.13)

1.14 (0.78–1.67)

  Multiple variablesb and educational level

1.98 (1.25–3.13)

1.14 (0.78–1.67)

aIncidence rate ratio is referred to as the RR in the text

bAge, age at birth of first child, age at onset of diabetes, BMI, cardiovascular disease, oestrogen use and smoking

Taking an extra latency period of 1 year into account, thereby starting follow-up on 1 January 2007, yielded a somewhat higher RR for users of insulin glargine alone (2.22, 95% CI 1.24–3.99). After extending the exposure period to 1 year, from 1 July 2005 to 30 June 2006, and starting follow-up on 1 July 2006, we found that, among users of insulin glargine alone, the RR of breast cancer was 2.14 (95% 1.24–3.71) for women with three or more prescriptions dispensed for insulin glargine and 1.53 (0.49–4.79) for women with one to two prescriptions dispensed, compared with those with no prescriptions dispensed for insulin glargine. We saw no statistically significant increase in incidence rate with increasing number of daily defined doses of insulin glargine (data not shown).

When restricting the analysis to those with type 1 diabetes or those with type 2 diabetes, a statistically significant difference in breast cancer occurrence between users of insulin glargine alone and users of types of insulin other than insulin glargine was seen for both types of diabetes (data not shown). Further analysis of the RR for each type of diabetes was not possible because of the small number of cases. The incidence rate of breast cancer was low before age 55; it was therefore not possible to study pre- and postmenopausal breast cancer. To obtain a measure of breast cancer occurrence after having used insulin glargine, irrespective of whether alone or together with other types of insulin, we combined the two groups of insulin glargine users and adjusted the RR for age, age at birth of first child, age at onset of diabetes, BMI, having had cardiovascular disease, oestrogen use and smoking. This gave an RR of 1.40 (1.04–1.89).

When we studied acute myocardial infarction as an outcome and adjusted for age, age at onset of diabetes, BMI and smoking, we obtained an RR of 0.77 (0.59–1.00) for female users of insulin glargine alone, and 0.88 (0.74–1.05) for female users of insulin glargine together with other types of insulin, compared with female users of types of insulin other than insulin glargine. When investigating mortality, we adjusted for age, age at onset of diabetes, BMI, smoking and having had cardiovascular disease. Female users of insulin glargine alone had a mortality ratio of 0.83 (95% CI 0.71–0.96), and female users of insulin glargine together with other types of insulin had a mortality ratio of 0.87 (0.77–0.97), compared with female users of types of insulin other than insulin glargine.

Discussion

We found that the incidence rates for gastrointestinal cancer, prostate cancer, and any type of malignancy among users of insulin glargine were similar to the incidence rates among users of other types of insulin. In Sweden, during 2006 and 2007, users of insulin glargine alone had a higher incidence rate of breast cancer than users of types of insulin other than insulin glargine.

We cannot rule out that the increased incidence rate for breast cancer in Sweden during 2006 and 2007 among users of insulin glargine alone, compared with users of other types of insulin, was the result of random fluctuation; statistical significance does not exclude this possibility. When examining problems with validity, we are limited to the information available in the registers; we cannot rule out validity problems for which we have no data. Specifically, we lack data on important possible confounding factors and, for some, we depend on data from the National Diabetes Register, which only covers about half of the individuals we study. All risk factors for breast cancer may confound the results we obtained; risk factors explaining a large proportion of the variation in the incidence rate for breast cancer in Sweden are more likely to confound the association than risk factors explaining only a small percentage.

Adjusting for available variables changed the RR to such a small degree that we can exclude significant confounding [5] by risk factors for breast cancer such as age at birth of first child [15], BMI [16], educational level [17], metformin intake [18] and oestrogen intake [19]. We do not know to what extent available variables reflect alcohol intake [20, 21], food intake [22] or health consciousness, and we do not have any information on genetic predisposition to breast cancer (e.g. presence of a mutation in BRCA1 [23, 24, 25] or BRCA2 [26, 27]).

To prevent the development of cancer having any influence on selection for observation of cancer occurrence (‘reversed causality’), we performed a fixed-cohort analysis, classifying individuals according to drug prescriptions dispensed in 2005. Some individuals registered as having a drug prescription dispensed for other types of insulin may certainly have used insulin glargine before 2005 or during the observation period during 2006 and 2007. This error often causes a non-differential misclassification of exposure, which shifts the RR towards 1.0 [28]. In other words, the uncertainty surrounding the extent to which a registered prescription dispensed for insulin glargine reflects real-life use of insulin glargine limits our ability to detect the true effect on the occurrence of malignancies. This limitation does not, however, as a rule, produce a spurious result indicating an effect. Thus, if we had been able to eliminate this inaccuracy, allowing us to only study the true users of insulin glargine with respect to the development of malignancies over the relevant time period, we would probably have obtained an age-adjusted RR higher than the 1.99 we obtained for breast cancer for users of insulin glargine alone. The available information indicates that loss to follow-up was small and did not differ between the different treatment groups. We did not find any indication of reversed causality, that is, of the presence of cancer having influenced the likelihood of using, or not using, insulin glargine. The mortality rate and the incidence rate of acute myocardial infarction were lower for users of insulin glargine alone than for users of types of insulin other than insulin glargine; we have no indication that users of insulin glargine alone had a higher rate of morbidity than the comparison group.

Duration from the start of exposure of industrial agents to an excess risk of cancer is usually 10–25 years. Among users of the drug chlornaphazine, an agent used to treat polycytemia verae, an increased risk of bladder cancer was observed as early as 3–5 years after treatment initiation [29, 30]. Chlornaphazine resembles the extremely strong and remarkably species-unspecific bladder cancer carcinogen β-naphthylamine. We are not aware of any documentation of an increased incidence of cancer the year after initiation of treatment with a specific drug. Insulin glargine appeared on the Swedish market in 2003, and we do not know the percentage of users of insulin glargine who started using the drug in that year. Nevertheless, the short duration from the start of insulin glargine use to the increased incidence rate for breast cancer suggests that our results could be due to random fluctuation.

We have no evidence of whether the difference in incidence rate for breast cancer among users of insulin glargine alone, compared with users of insulin glargine together with other types of insulin, is caused by random fluctuations, interaction between insulin glargine and another insulin, or the presence of an as-yet-unidentified effect-modifying factor in the insulin glargine alone group. Any suggestion of an explanation would be pure speculation. The group of users of insulin glargine in combination with other types of insulin had a lower mean age and had more often type 1 diabetes than the other two groups.

Weinstein and co-workers demonstrated that insulin glargine stimulates the proliferation of cultured colorectal, prostate and breast cancer cells [2]. Epidemiological studies link circulating IGF-I concentrations to the incidence of breast, prostate and colorectal cancer [31, 32, 33]. Insulin glargine may have a similar mechanism of action to IGF-I. Thus, before the study, we expected that an increased incidence rate of breast cancer, if present, would occur in parallel with an increased incidence rate of gastrointestinal and prostate cancer. We found no statistically significant results for prostate or gastrointestinal cancer, which, again, strengthens the interpretation that the breast cancer results were due to random fluctuation.

Data from other settings are needed before any true effects of the incidence of malignancies related to insulin glargine can be evaluated with a high level of confidence. When more evidence becomes available, the absolute increase in incidence, if any, must be weighed against the beneficial effects, as well as other adverse effects, of using insulin glargine compared with other types of insulin.

Copyright information

© Springer-Verlag 2009

Authors and Affiliations

  • J. M. Jonasson
    • 1
    • 2
  • R. Ljung
    • 3
    • 4
    • 5
  • M. Talbäck
    • 3
  • B. Haglund
    • 3
  • S. Gudbjörnsdòttir
    • 6
    • 7
  • G. Steineck
    • 1
    • 2
  1. 1.Division of Clinical Cancer Epidemiology, Department of Oncology, Onkologiskt CentrumSahlgrenska AcademyGothenburgSweden
  2. 2.Division of Clinical Cancer Epidemiology, Department of Oncology-PathologyKarolinska InstitutetStockholmSweden
  3. 3.The National Board of Health and WelfareStockholmSweden
  4. 4.Equity and Health Policy, Department of Public Health SciencesKarolinska InstitutetStockholmSweden
  5. 5.Upper Gastrointestinal Research, Department of Molecular Medicine and SurgeryKarolinska InstitutetStockholmSweden
  6. 6.Diabetes CentreSahlgrenska University HospitalGothenburgSweden
  7. 7.Nordic School of Public HealthGothenburgSweden

Personalised recommendations