Browse Subject Areas

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Microsatellite instability is inversely associated with type 2 diabetes mellitus in colorectal cancer

  • Yujiro Nakayama,

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Software, Writing – original draft

    Affiliations Department of Surgery, Tokyo Metropolitan Cancer and Infectious diseases Center Komagome Hospital, Tokyo, Japan, Department of Minimally Invasive Surgical and Medical Oncology, Fukushima Medical University, Fukushima, Japan, Department of Surgery, Southern Tohoku General Hospital, Fukushima, Japan

  • Takeru Iijima,

    Roles Investigation

    Affiliation Hereditary Tumor Research Project, Tokyo Metropolitan Cancer and Infectious diseases Center Komagome Hospital, Tokyo, Japan

  • Rika Wakaume,

    Roles Investigation, Resources

    Affiliation Hereditary Tumor Research Project, Tokyo Metropolitan Cancer and Infectious diseases Center Komagome Hospital, Tokyo, Japan

  • Keiichi Takahashi,

    Roles Investigation

    Affiliation Department of Surgery, Tokyo Metropolitan Cancer and Infectious diseases Center Komagome Hospital, Tokyo, Japan

  • Hiroshi Matsumoto,

    Roles Investigation

    Affiliation Department of Surgery, Tokyo Metropolitan Cancer and Infectious diseases Center Komagome Hospital, Tokyo, Japan

  • Daisuke Nakano,

    Roles Investigation

    Affiliation Department of Surgery, Tokyo Metropolitan Cancer and Infectious diseases Center Komagome Hospital, Tokyo, Japan

  • Michiko Miyaki,

    Roles Investigation

    Affiliation Hereditary Tumor Research Project, Tokyo Metropolitan Cancer and Infectious diseases Center Komagome Hospital, Tokyo, Japan

  • Tatsuro Yamaguchi

    Roles Methodology, Resources, Supervision, Validation, Writing – review & editing

    Affiliations Department of Surgery, Tokyo Metropolitan Cancer and Infectious diseases Center Komagome Hospital, Tokyo, Japan, Hereditary Tumor Research Project, Tokyo Metropolitan Cancer and Infectious diseases Center Komagome Hospital, Tokyo, Japan

Microsatellite instability is inversely associated with type 2 diabetes mellitus in colorectal cancer

  • Yujiro Nakayama, 
  • Takeru Iijima, 
  • Rika Wakaume, 
  • Keiichi Takahashi, 
  • Hiroshi Matsumoto, 
  • Daisuke Nakano, 
  • Michiko Miyaki, 
  • Tatsuro Yamaguchi



Microsatellite instability (MSI) is a clonal change in the number of repeated DNA nucleotide units in microsatellites. High-frequency MSI (MSI-H) colorectal cancers (CRCs) are known to have different clinicopathological features compared with microsatellite stable (MSS) CRCs. In addition, previous studies have shown that type2 diabetes mellitus (T2DM) is a risk factor for malignant tumors including CRCs. The aim of this study was to investigate the relationship between T2DM and MSI-H colorectal cancer.


The study design is a single center, cross-sectional study. Data from a series of 936 patients with CRCs were collected and MSI status was assessed.


In total, 29 (3.1%) and 907 (96.9%) tumors were classified as having MSI-H and low-frequency microsatellite instability or being MSS (MSS), respectively. Of the 936 patients, 275 (29.6%) were associated with T2DM. One (3.4%) of the 29 MSI-H patients and 274 (30.2%) of the 907 MSS patients had T2DM. Thus, the incidence of T2DM was significantly less frequent in MSI-H compared with MSS patients (Fisher’s exact test: p = 0.0007).


We conclude that MSS tumors are significantly more common than MSI-H tumors among individuals with T2DM.


Colorectal cancer (CRC) is one of the most common solid tumors, and details associated with its carcinogenesis have been intensively studied. Some colorectal cancers are microsatellite stable (MSS). The typical development of MSS tumors proceeds stepwise by the inactivation of increasing numbers of tumor suppressor genes, including the APC and p53 genes, through mutation as well as loss of heterozygosity (LOH) and the activation of oncogenes such as KRAS [1, 2].

On the other hand, microsatellite instability (MSI) is a phenotype resulting from a defect in mismatch repair genes such as MSH2 [3, 4], MLH1 [4], and MSH6 [5, 6]. No MSI tumor shows LOH at these tumor suppressor genes [7], and target genes for frameshift mutations in CRCs are different from those in MSS tumors. Well-known target genes for MSI tumors include TGFβRII [8], IGFIIR [9], and BAX [10]. Testing colorectal cancers for MSI is an effective method of screening for Lynch syndrome because approximately 90% of Lynch syndrome tumors have high microsatellite instability (MSI-H) [11]. Although MSI is also observed in sporadic CRC, CRCs with MSI-H, irrespective of whether they are hereditary or not, have important therapeutic and diagnostic characteristics. CRCs with MSI-H are generally associated with a better prognosis [12], but their prognosis is less favorable with 5-FU based chemotherapy [13]; in addition, these tumors have less metastasis, and are more likely to be a right sided or metachronous multiple CRCs [12, 14].

Type 2 diabetes mellitus (T2DM) is associated with malignant tumors, including CRC [15]. It has been reported that CRC incidences in diabetic patients were 1.27–1.40 in colon and 1.19–1.36 in rectum [1618]. Jiang Y et al. reported that relative risk of CRC incidence for T2DM were 1.27 (95% CI: 1.21–1.34) by meta-analysis of cohort studies [19]. However, the relationship between microsatellite instability of CRC and T2DM has not been clear yet. In this study, we investigated a relationship between T2DM and MSI in CRC.

Materials and methods


This study is a single center, retrospective cross-sectional study. We selected 936 consecutive colorectal cancer patients who underwent surgical resection at the Tokyo Metropolitan Cancer and Infectious diseases Center Komagome Hospital from January 2008 to January 2014 after obtaining their informed consent. The study was performed after approval of the Tokyo Metropolitan Cancer and Infectious Diseases Center Komagome Hospital Ethical Committee (ID: 612). Patients with inflammatory bowel disease or a known history of familial adenomatous polyposis and Lynch syndrome were excluded. If the patient had two or more colorectal tumors resected, then the tumor that was most advanced was selected for analysis.

We defined patients with T2DM as those who had already been diagnosed with T2DM or those who had a high concentration of glycated hemoglobin (HbA1c > 6.5%) which was measured before the first colorectal surgery. The criteria for T2DM diagnosis depends on Standards of Medical Care in Diabetes 2018 by American Diabetes Association [20].

Body weight (kg) and height (m) were measured immediately before surgery. The body mass index (BMI) was calculated as weight in kilograms divided by the square of the height in meters (kg/m2).

Microsatellite instability analysis and mutation analysis.

Colorectal cancers and corresponding normal tissues were obtained with informed consent and were stored at −80°C immediately after resection. Genomic DNA samples were extracted using the QIAamp DNA mini kit (QIAGEN, Valencia, CA,USA). Methods to determine microsatellite instability and KRAS and BRAF mutations were describe previously. Briefly, polymerase chain reaction (PCR) was performed to amplify at least five repetitive sequence loci from the tumor and normal tissue samples: BAT25, BAT26, D2S123, D5S346, and D17S250. Microsatellite instability status was defined as MSI-H (2–5 of the 5 markers used were unstable) and MSS (none or only 1 of the 5 markers was unstable), as described in the National Cancer Institute guidelines for MSI testing (22). All samples were analyzed to identify any BRAF (V600E) and KRAS (codons 12 and 13) mutations by direct sequencing.

Statistical analysis

The Fisher’s exact test was used to evaluate the relationship between two discrete and dichotomous variables. The analysis of association between categorical variables was performed using logistic regression method. An optimal cut-off value for predicting MSI status was analyzed using receiver operating characteristic (ROC) curves. Areas under the curve (AUC) were also calculated. All statistical tests were 2-sided, and P values of ≤0.05 were considered to indicate statistical significance. All statistical analyses were performed with EZR (Saitama Medical Center, Jichi Medical University, Japan), which is a graphical-user interface for R (The R Foundation for Statistical Computing, Vienna, Austria, version 3.5.1). This interface is a modified version of R commander (version 2.5–1) that includes statistical functions that are frequently used in biostatistics.


Total of 936 CRC patients were enrolled during study period in this study. All patients underwent surgical resection of the primary tumor and the diagnosis of adenocarcinoma was made pathologically. None received neo-adjuvant chemotherapy or radiotherapy. The clinical and pathological characteristics of the patients are shown in Table 1. Of the 936 colorectal cancers, 29 (3.1%) and 907 (96.9%) tumors were classified as MSI-H and MSS, respectively. Significant differences between MSI-H and MSS cancers were observed with respect to sex (p = 0.022), age (p = 0.007), tumor location (p < 0.0001) and histology (p < 0.0001); whereas, there were no significant differences with respect to UICC classification (p = 0.57). One hundred seventy-five (18.7%) patients had T2DM and no patient had type 1 diabetes mellitus. The incidence of T2DM was significantly less frequent in MSI-H than MSS patients (p = 0.0007). However, there was no significant difference in BMI between MSI-H and MSS patients (p = 0.65). Of the 936 CRC patients, 277 (29.6%) had KRAS mutation and 49 (5.2%) had BRAF mutation. None of the 29 MSI-H patients were a KRAS mutation, whereas 277 (30.5%) of the 907 MSS patients had a KRAS mutation, and frequency of KRAS mutation was significantly different between MSI-H and MSS (p < 0.0001). Inversely, frequency of BRAF mutation was significantly high in MSI-H patients than in MSS patients (p < 0.0001). According to logistic regression analysis, age, tumor location, histology and T2DM status were independent factors in MSI-H tumor (Table 2).

Table 3 showed clinicopathological features according to T2DM status. T2DM+ was significantly more frequent in male patients, elderly patients and obesity patients; however, no correlation was seen between T2DM+ and tumor location, histology or UICC classification. BRAF gene mutation was significantly more frequent in T2DM- patients, while KRAS gene mutation was not correlated with T2DM status.


Our study findings indicated that T2DM is significantly less common among MSI-H patients compared with MSS patients. This is the first report to indicate that T2DM may be associated with MSS CRC. There have been no previous reports of a relationship between T2DM and MSI status in CRC, even though it is well known that the incidence of malignant tumors is increased in patients with T2DM [7, 21].

Our study revealed the frequency of MSI-H tumors to be 3.1%, which reported as well as in previous studies in Asian countries is lower than that reported in Western countries [22, 23]. Asaka et al reported on the frequency of MSI in 940 Japanese CRC patients and found that 5.9% were MSI-H and 94.1% were MSS/MSI-L [24]. The incidence of rectal cancer is higher in Japanese individuals (approximately 40% of CRCs) compared with individuals from Western countries (approximately 20% of CRCs), which would reflect a lower rate of MSI-H CRC because rectal cancer is less likely to show MSI-H than colon cancer. In addition, as we reported previously, MSI-H is less frequent even in right colon cancer in Japanese individuals than in Western. [25]. Race may thus affect MSI status.

T2DM has been shown to be a risk factor for malignant tumors including CRC [15]. T2DM and CRC are major causes of morbidity and mortality in the United States, Western countries, and increasingly in Japan [26, 27]. Dietary and lifestyle risk factors for developing insulin resistance and T2DM, such as a Western diet, physical inactivity, and obesity, have also been linked to an increased risk of CRC [2830]. Furthermore, an association between metabolic syndrome and CRC is now supported by a large number of epidemiological studies [16, 17, 3133]. In this study, the frequency of obesity was almost identical between the MSS and MSI-H groups, and thus, we were able to compare the incidence of T2DM as an independent risk factor for colorectal cancer.

Previous reports have shown that in patients who suffer from T2DM, hyperinsulinemia, or factors related to insulin resistance, such as hyperglycemia or hypertriglyceridemia, are associated with colorectal carcinogenesis [16]. IGF-1, which is suggested to stimulate cell proliferation by its activation, is reported to be associated with CRC both epidemiologically and experimentary [34]. Moreover, biologic interactions among insulin, IGF-1, and IGFBPs may increase the risk of CRC through diet and associated factors, including Wnt pathway and PI3K/Akt pathway [3540].

The current study had some limitations as follows: (1) selection bias caused by retrospective nature of the study; (2) a single-center study; (3) our study revealed that T2DM was associated with MSS colorectal cancer; however, we could not provide the mechanism. Nonetheless, considering that there are only a few publications on association with T2DM and MSS colorectal cancer, we believe that our findings will help researchers and physicians clarify the nature of colorectal cancer. And, we also know that further studies are required to overcome these limitations.


We conclude that MSS tumors are significantly more common than MSI-H tumors among individuals with T2DM.

Supporting information

S1 File. NAKAYAMA_PONE-D-18-34914new.

Supplemental data.



  1. 1. Vogelstein B, Fearon ER, Hamilton SR, Kern SE, Preisinger AC, Leppert M, et al. Genetic alterations during colorectal-tumor development. The New England journal of medicine. 1988;319(9):525–32. Epub 1988/09/01. pmid:2841597.
  2. 2. Miyaki M, Seki M, Okamoto M, Yamanaka A, Maeda Y, Tanaka K, et al. Genetic changes and histopathological types in colorectal tumors from patients with familial adenomatous polyposis. Cancer Res. 1990;50(22):7166–73. Epub 1990/11/15. pmid:1977514.
  3. 3. Fishel R, Lescoe MK, Rao MR, Copeland NG, Jenkins NA, Garber J, et al. The human mutator gene homolog MSH2 and its association with hereditary nonpolyposis colon cancer. Cell. 1993;75(5):1027–38. Epub 1993/12/03. pmid:8252616.
  4. 4. Leach FS, Nicolaides NC, Papadopoulos N, Liu B, Jen J, Parsons R, et al. Mutations of a mutS homolog in hereditary nonpolyposis colorectal cancer. Cell. 1993;75(6):1215–25. Epub 1993/12/17. pmid:8261515.
  5. 5. Akiyama Y, Sato H, Yamada T, Nagasaki H, Tsuchiya A, Abe R, et al. Germ-line mutation of the hMSH6/GTBP gene in an atypical hereditary nonpolyposis colorectal cancer kindred. Cancer Res. 1997;57(18):3920–3. Epub 1997/10/27. pmid:9307272.
  6. 6. Miyaki M, Konishi M, Tanaka K, Kikuchi-Yanoshita R, Muraoka M, Yasuno M, et al. Germline mutation of MSH6 as the cause of hereditary nonpolyposis colorectal cancer. Nature genetics. 1997;17(3):271–2. Epub 1997/11/14. pmid:9354786.
  7. 7. Konishi M, Kikuchi-Yanoshita R, Tanaka K, Muraoka M, Onda A, Okumura Y, et al. Molecular nature of colon tumors in hereditary nonpolyposis colon cancer, familial polyposis, and sporadic colon cancer. Gastroenterology. 1996;111(2):307–17. Epub 1996/08/01. pmid:8690195.
  8. 8. Markowitz S, Wang J, Myeroff L, Parsons R, Sun L, Lutterbaugh J, et al. Inactivation of the type II TGF-beta receptor in colon cancer cells with microsatellite instability. Science (New York, NY). 1995;268(5215):1336–8. Epub 1995/06/02. pmid:7761852.
  9. 9. Souza RF, Appel R, Yin J, Wang S, Smolinski KN, Abraham JM, et al. Microsatellite instability in the insulin-like growth factor II receptor gene in gastrointestinal tumours. Nature genetics. 1996;14(3):255–7. Epub 1996/11/01. pmid:8896552.
  10. 10. Rampino N, Yamamoto H, Ionov Y, Li Y, Sawai H, Reed JC, et al. Somatic frameshift mutations in the BAX gene in colon cancers of the microsatellite mutator phenotype. Science (New York, NY). 1997;275(5302):967–9. Epub 1997/02/14. pmid:9020077.
  11. 11. Liu B, Parsons R, Papadopoulos N, Nicolaides NC, Lynch HT, Watson P, et al. Analysis of mismatch repair genes in hereditary non-polyposis colorectal cancer patients. Nature medicine. 1996;2(2):169–74. Epub 1996/02/01. pmid:8574961.
  12. 12. Ward R, Meagher A, Tomlinson I, O'Connor T, Norrie M, Wu R, et al. Microsatellite instability and the clinicopathological features of sporadic colorectal cancer. Gut. 2001;48(6):821–9. Epub 2001/05/19. pmid:11358903; PubMed Central PMCID: PMCPMC1728324.
  13. 13. Ribic CM, Sargent DJ, Moore MJ, Thibodeau SN, French AJ, Goldberg RM, et al. Tumor microsatellite-instability status as a predictor of benefit from fluorouracil-based adjuvant chemotherapy for colon cancer. The New England journal of medicine. 2003;349(3):247–57. Epub 2003/07/18. pmid:12867608; PubMed Central PMCID: PMCPMC3584639.
  14. 14. Shitoh K, Konishi F, Miyakura Y, Togashi K, Okamoto T, Nagai H. Microsatellite instability as a marker in predicting metachronous multiple colorectal carcinomas after surgery: a cohort-like study. Diseases of the colon and rectum. 2002;45(3):329–33. Epub 2002/06/18. pmid:12068189.
  15. 15. Larsson SC, Orsini N, Wolk A. Diabetes mellitus and risk of colorectal cancer: a meta-analysis. Journal of the National Cancer Institute. 2005;97(22):1679–87. Epub 2005/11/17. pmid:16288121.
  16. 16. Jarvandi S, Davidson NO, Schootman M. Increased risk of colorectal cancer in type 2 diabetes is independent of diet quality. PloS one. 2013;8(9):e74616. Epub 2013/09/27. pmid:24069323; PubMed Central PMCID: PMCPMC3771921.
  17. 17. Wang M, Hu RY, Wu HB, Pan J, Gong WW, Guo LH, et al. Cancer risk among patients with type 2 diabetes mellitus: a population-based prospective study in China. Scientific reports. 2015;5:11503. Epub 2015/06/18. pmid:26082067; PubMed Central PMCID: PMCPMC4469976.
  18. 18. Liu X, Hemminki K, Forsti A, Sundquist K, Sundquist J, Ji J. Cancer risk in patients with type 2 diabetes mellitus and their relatives. International journal of cancer. 2015;137(4):903–10. Epub 2015/01/22. pmid:25604005.
  19. 19. Jiang Y, Ben Q, Shen H, Lu W, Zhang Y, Zhu J. Diabetes mellitus and incidence and mortality of colorectal cancer: a systematic review and meta-analysis of cohort studies. European journal of epidemiology. 2011;26(11):863–76. Epub 2011/09/23. pmid:21938478.
  20. 20. 2. Classification and Diagnosis of Diabetes: Standards of Medical Care in Diabetes-2018. Diabetes care. 2018;41(Suppl 1):S13–s27. Epub 2017/12/10. pmid:29222373.
  21. 21. Kuriki K, Tokudome S, Tajima K. Association between type II diabetes and colon cancer among Japanese with reference to changes in food intake. Asian Pacific journal of cancer prevention: APJCP. 2004;5(1):28–35. Epub 2004/04/13. pmid:15075001.
  22. 22. Nosho K, Kure S, Irahara N, Shima K, Baba Y, Spiegelman D, et al. A prospective cohort study shows unique epigenetic, genetic, and prognostic features of synchronous colorectal cancers. Gastroenterology. 2009;137(5):1609–20.e1-3. Epub 2009/08/19. pmid:19686742; PubMed Central PMCID: PMCPMC2859181.
  23. 23. Vilar E, Gruber SB. Microsatellite instability in colorectal cancer-the stable evidence. Nature reviews Clinical oncology. 2010;7(3):153–62. Epub 2010/02/10. pmid:20142816; PubMed Central PMCID: PMCPMC3427139.
  24. 24. Asaka S, Arai Y, Nishimura Y, Yamaguchi K, Ishikubo T, Yatsuoka T, et al. Microsatellite instability-low colorectal cancer acquires a KRAS mutation during the progression from Dukes' A to Dukes' B. Carcinogenesis. 2009;30(3):494–9. Epub 2009/01/17. pmid:19147861.
  25. 25. Natsume S, Yamaguchi T, Takao M, Iijima T, Wakaume R, Takahashi K, et al. Clinicopathological and molecular differences between right-sided and left-sided colorectal cancer in Japanese patients. Jpn J Clin Oncol. 2018;48(7):609–18. Epub 2018/05/17. pmid:29767751.
  26. 26. Kim H, Jen J, Vogelstein B, Hamilton SR. Clinical and pathological characteristics of sporadic colorectal carcinomas with DNA replication errors in microsatellite sequences. The American journal of pathology. 1994;145(1):148–56. Epub 1994/07/01. pmid:8030745; PubMed Central PMCID: PMCPMC1887287.
  27. 27. Jemal A, Siegel R, Xu J, Ward E. Cancer statistics, 2010. CA: a cancer journal for clinicians. 2010;60(5):277–300. Epub 2010/07/09. pmid:20610543.
  28. 28. Schulze MB, Hu FB. Primary prevention of diabetes: what can be done and how much can be prevented? Annual review of public health. 2005;26:445–67. Epub 2005/03/12. pmid:15760297.
  29. 29. Fung TT, Schulze M, Manson JE, Willett WC, Hu FB. Dietary patterns, meat intake, and the risk of type 2 diabetes in women. Archives of internal medicine. 2004;164(20):2235–40. Epub 2004/11/10. pmid:15534160.
  30. 30. Giovannucci E. Modifiable risk factors for colon cancer. Gastroenterology clinics of North America. 2002;31(4):925–43. Epub 2002/12/20. pmid:12489270.
  31. 31. Khaw KT, Wareham N, Bingham S, Luben R, Welch A, Day N. Preliminary communication: glycated hemoglobin, diabetes, and incident colorectal cancer in men and women: a prospective analysis from the European prospective investigation into cancer-Norfolk study. Cancer epidemiology, biomarkers & prevention: a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology. 2004;13(6):915–9. Epub 2004/06/09. pmid:15184246.
  32. 32. Colangelo LA, Gapstur SM, Gann PH, Dyer AR, Liu K. Colorectal cancer mortality and factors related to the insulin resistance syndrome. Cancer epidemiology, biomarkers & prevention: a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology. 2002;11(4):385–91. Epub 2002/04/03. pmid:11927499.
  33. 33. Pais R, Silaghi H, Silaghi AC, Rusu ML, Dumitrascu DL. Metabolic syndrome and risk of subsequent colorectal cancer. World journal of gastroenterology. 2009;15(41):5141–8. Epub 2009/11/06. pmid:19891012; PubMed Central PMCID: PMCPMC2773892.
  34. 34. Ma J, Pollak MN, Giovannucci E, Chan JM, Tao Y, Hennekens CH, et al. Prospective study of colorectal cancer risk in men and plasma levels of insulin-like growth factor (IGF)-I and IGF-binding protein-3. Journal of the National Cancer Institute. 1999;91(7):620–5. Epub 1999/04/15. pmid:10203281.
  35. 35. Sandhu MS, Dunger DB, Giovannucci EL. Insulin, insulin-like growth factor-I (IGF-I), IGF binding proteins, their biologic interactions, and colorectal cancer. Journal of the National Cancer Institute. 2002;94(13):972–80. Epub 2002/07/04. pmid:12096082.
  36. 36. Desbois-Mouthon C, Cadoret A, Blivet-Van Eggelpoel MJ, Bertrand F, Cherqui G, Perret C, et al. Insulin and IGF-1 stimulate the beta-catenin pathway through two signalling cascades involving GSK-3beta inhibition and Ras activation. Oncogene. 2001;20(2):252–9. Epub 2001/04/21. pmid:11313952.
  37. 37. Liang J, Slingerland JM. Multiple roles of the PI3K/PKB (Akt) pathway in cell cycle progression. Cell cycle (Georgetown, Tex). 2003;2(4):339–45. Epub 2003/07/10. pmid:12851486.
  38. 38. Hennessy BT, Smith DL, Ram PT, Lu Y, Mills GB. Exploiting the PI3K/AKT pathway for cancer drug discovery. Nature reviews Drug discovery. 2005;4(12):988–1004. Epub 2005/12/13. pmid:16341064.
  39. 39. Polakis P. The many ways of Wnt in cancer. Current opinion in genetics & development. 2007;17(1):45–51. Epub 2007/01/09. pmid:17208432.
  40. 40. van de Wetering M, Sancho E, Verweij C, de Lau W, Oving I, Hurlstone A, et al. The beta-catenin/TCF-4 complex imposes a crypt progenitor phenotype on colorectal cancer cells. Cell. 2002;111(2):241–50. Epub 2002/11/01. pmid:12408868.