Abstract
Introduction
For many cancers, features of the metabolic syndrome, such as diabetes and obesity, have been associated with both increased risk of cancer development and poor outcomes.
Materials and Methods
We examined a large retrospective cohort of 342 consecutive patients who underwent liver transplantation for hepatocellular carcinoma between January, 1999 and July, 2010 at our institution. We evaluated the relationship between diabetes, obesity, HCC recurrence, and overall survival.
Results
We found that a body mass index (BMI)>30 was an independent predictor of poor overall survival in a multivariable Cox model, approximately doubling the risk of death after transplant. BMI>30 was also a predictor of recurrent HCC, although this was of borderline statistical significance (HR for recurrence 1.9, 95% CI 0.9–4.1). We also found increased BMI to be an independent predictor of microvascular invasion (MVI) within HCC tumors, lending a possible explanation to these results. Those with diabetes had worsened overall survival compared to those without diabetes in univariate, but not multivariable analysis, possibly related to longer wait times.
Conclusions
Our findings suggest a relationship between higher BMI, tumor vascular invasion, increased recurrence, and worsened overall survival. These findings may help explain why those with high BMI have worse outcomes from their cancers. A better understanding of the role of obesity and diabetes in patients with cancer should help develop better predictors of outcome and improved treatment options for patients with HCC.
Keywords: Liver transplantation, hepatocellular carcinoma, diabetes, obesity, outcomes
Introduction
Hepatocellular carcinoma (HCC) is one of the most common malignancies in the world, with over 500,000 new cases per year.[1] Its age-adjusted incidence rates in the United States have doubled over the last 20 years, with a substantial proportion of this increase attributed to the chronic hepatitis C, and the increasing incidence of metabolic syndrome in the Unites States.[2, 3] The overall prognosis of patients with HCC in the U.S. is poor, especially for HCC patients who do not receive curative therapy.[4, 5]
The purpose of our study was to evaluate the relationship between obesity and diabetes, and HCC recurrence and survival in a large cohort of patients undergoing liver transplantation at our institution. We previously performed a pilot study of 138 patients who underwent liver resection or liver transplant for hepatocellular carcinoma, and found that those with a high body mass index were significantly more likely to have microvascular invasion on their pathology specimens. However, BMI was not an independent predictor of worsened overall survival, perhaps because of small sample sizes. [6] In patients with HCC who have undergone resection, some studies have found that prior diabetes does not impact survival, but does lead to increased cancer recurrence;[7] others found worsened overall survival but no increased risk of recurrence, [8] and still others have found no effect of diabetes on perioperative outcome, long-term survival, or disease-free survival. [9]
In this manuscript, we assess the relationship between BMI, tumor vascular invasion, and survival in a larger and more homogeneous cohort of patients undergoing liver transplantation. We also examine the relationship between BMI and tumor recurrence, which we had not looked at previously. Finally, we evaluate whether a history of diabetes pre-transplant yield similar associations with vascular invasion, increased risk of recurrence, and worsened overall survival.
Results
We analyzed 342 consecutive patients; the median age was 58, and 78% were men. Underlying HCV was noted in 70% of patients, and 16% had underlying HBV. About a third were diabetic (10 were type 1, and the remainder type 2), and 25% had a BMI>30. Other characteristics of our patients are shown in Table 1. Median 1, 3 and 5 year overall survivals were 85%, 68%, and 58% respectively. Recurrence-free survivals were 92%, 86%, and 82% respectively. Diabetics were more likely to be obese than non diabetics, although this relationship just missed statistical significance (p=0.07). We did not define a “NASH” etiology separately because of the difficulties defining NASH pathologically in HCC patients, the majority of whom have developed cirrhosis, which obscures the underlying pathology.
Table 1.
Subject and tumor characteristics
| Characteristic | Median | Min | Max | N | % |
|---|---|---|---|---|---|
| Age | 57.6 | 27.2 | 74.8 | ||
| Lab MELD | 11.8 | 3.1 | 38.8 | ||
| MAX AFP | 22.5 | 1.7 | 24564.0 | ||
| Days on wait list | 147 | 0 | 3832 | ||
| Gender | |||||
| Male | 268 | 78 | |||
| Female | 74 | 22 | |||
| Race/ethnicity | |||||
| White | 191 | 56 | |||
| Hispanic | 74 | 22 | |||
| Black | 31 | 9 | |||
| Asian | 42 | 12 | |||
| Other | 4 | 1 | |||
| DM | |||||
| Yes | 106 | 31 | |||
| No | 233 | 69 | |||
| BMI>30 | |||||
| Yes | 77 | 25 | |||
| No | 237 | 75 | |||
| Ethanol use | |||||
| Yes | 73 | 25 | |||
| No | 224 | 75 | |||
| Hepatitis C | |||||
| Yes | 239 | 70 | |||
| No | 101 | 30 | |||
| Hepatitis B | |||||
| Yes | 54 | 16 | |||
| No | 285 | 84 | |||
| Within Milan | |||||
| Yes | 225 | 73 | |||
| No | 82 | 27 | |||
| Vascular invasion | |||||
| Yes | 86 | 25 | |||
| No | 256 | 75 | |||
| Grade 4 | |||||
| Yes | 45 | 14 | |||
| No | 269 | 86 | |||
| TNM stage | |||||
| I | 30 | 9 | |||
| II | 141 | 41 | |||
| III | 131 | 38 | |||
| IV | 40 | 12 | |||
| Child-Pugh score | |||||
| A | 141 | 46 | |||
| B | 89 | 29 | |||
| C | 74 | 24 | |||
| Chemoembolization | |||||
| Yes | 242 | 71 | |||
| No | 99 | 29 | |||
| Radiofrequency ablation | |||||
| Yes | 51 | 17 | |||
| No | 246 | 83 | |||
| Resection | |||||
| Yes | 17 | 5 | |||
| No | 325 | 95 | |||
We next evaluated univariate predictors of overall survival. Favorable prognostic features included absence of vascular invasion on tumor explant, absence of a high-grade tumor, having a lower than median AFP, and having a shorter than median wait list time. Having alcohol-related and HBV-related liver disease were of borderline significance predicting better survival (p=0.06, data not shown).
Similar to other groups, we found that being diabetic was a significant univariate predictor of worsened overall survival using Kaplan-Meier analysis (p=0.003). [10, 11] We attempted to determine why this was the case, and assessed the relationship between being diabetic and having a high grade tumor, having vascular invasion, being outside Milan criteria, or having an AFP greater than the median, but none was significantly associated with being diabetic. We did find that those with diabetes waited on average 3 months longer on the wait list than those without diabetes. Those with diabetes did not have worsened overall survival on multivariable analysis, and did not have a significantly higher risk of tumor recurrence.
In a multivariable Cox model for overall survival, we included median AFP, median age, diabetes history, BMI>30, HBV, HCV, ETOH use, high grade disease, being outside Milan criteria, having vascular invasion on explant, and wait list time (Table 2). We found that high AFP, any vascular invasion, and BMI>30 all approximately doubled the risk of death. Those with HBV lived about 4 times as long as those without HBV.
Table 2.
Cox proportional hazards model of predictors of all-cause mortality
| Variable | OR | 95% CI |
|---|---|---|
| AFP | ||
| <22.5 | 1 | Referent |
| ≥22.5 | 2.1 | 1.2–3.7 |
| Age | ||
| <57.6 | 1 | Referent |
| ≥57.6 | 1.3 | 0.8–2.2 |
| DM | ||
| No | 1 | Referent |
| Yes | 1.2 | 0.7–2.0 |
| BMI | ||
| <30 | 1 | Referent |
| ≥30 | 1.8 | 1.1–3.0 |
| Hepatitis B | ||
| No | 1 | Referent |
| Yes | 0.2 | 0.1–0.9 |
| Hepatitis C | ||
| No | 1 | Referent |
| Yes | 0.7 | 0.3–1.4 |
| Ethanol Use | ||
| No | 1 | Referent |
| Yes | 0.8 | 0.4–1.6 |
| Grade 4 | ||
| No | 1 | Referent |
| Yes | 1.1 | 0.6–2.1 |
| Within Milan | ||
| No | 1 | Referent |
| Yes | 0.7 | 0.5–1.2 |
| Vascular invasion | ||
| No | 1 | Referent |
| Yes | 1.8 | 1.1–2.9 |
All variables adjusted for each other
As in our previous pilot study of 138 HCC patients undergoing surgery and resection, here again we found a significant relationship between increased BMI and microvascular invasion seen on explant (p=0.04). Those with a BMI>30 were significantly more likely to have microvascular invasion than those with a lower BMI (31% vs 20%).
We then evaluated univariate predictors of recurrent hepatocellular carcinoma, using Kaplan-Meier curves. Vascular invasion was strongly related to recurrence as expected (p<0.001), as was being outside Milan (p<0.001), having high grade disease (p<0.001), and having a high AFP (p<0.001). In a multivariable model of recurrence which included AFP, age, HBV, HCV, ETOH-related disease, BMI>30, diabetes history, vascular invasion, grade, Milan criteria and length of time on the wait list, only high median AFP, high grade disease, and vascular invasion were independent predictors of recurrence; BMI>30 approximately doubled risk of recurrence but was of marginal statistical significance. (Table 3)
Table 3.
Cox proportional hazards model of time to recurrence
| Variable | OR | 95% CI |
|---|---|---|
| AFP | ||
| <22.5 | 1 | Referent |
| ≥22.5 | 4.4 | 1.6–12.1 |
| Age | ||
| <57.6 | 1 | Referent |
| ≥57.6 | 1.1 | 0.5–2.3 |
| DM | ||
| No | 1 | Referent |
| Yes | 0.8 | 0.3–1.8 |
| BMI | ||
| <30 | 1 | Referent |
| ≥30 | 1.9 | 0.9–4.1 |
| Hepatitis B | ||
| No | 1 | Referent |
| Yes | 0.3 | <0.1–1.6 |
| Hepatitis C | ||
| No | 1 | Referent |
| Yes | 0.4 | 0.1–1.3 |
| Ethanol Use | ||
| No | 1 | Referent |
| Yes | 1.8 | 0.6–5.0 |
| Grade 4 | ||
| No | 1 | Referent |
| 3.8 | 1.8–8.2 | |
| Within Milan | ||
| No | 1 | Referent |
| Yes | 0.5 | 0.3–1.1 |
| Vascular invasion | ||
| No | 1 | Referent |
| Yes | 4.5 | 1.9–10.6 |
All variables adjusted for each other
Discussion
For many common cancers, features of the metabolic syndrome such as diabetes and obesity have been associated with both increased risk of cancer development and poor outcomes. [12] In our large cohort of patients with HCC undergoing liver transplantation for HCC, we found body mass index (BMI) to be an independent predictor of worsened survival on multivariable analysis. BMI also independently predicted an approximate doubling in risk of recurrence, although the 95% confidence interval spanned 1 (0.9–4.1). As expected, patients with microvascular invasion had poorer survivals than other patients, and we show that those with a BMI>30 were more likely to have microvascular invasion in their tumors. We hypothesize that this worsened survival and increased recurrence risk may be related to increased microvascular invasion (MVI) within HCC tumors in those patients with a higher BMI.
One possible explanation for the relationship between increased microvascular invasion and body mass index is via increased expression of adipokines in patients with higher BMI. Adipose tissue has been shown to induce expression of vascular endothelial growth factor (VEGF) and other cytokines in human and animal models. [13] Vascular invasion in HCC and other cancers is associated with markers of angiogenesis, such as (VEGF) expression. [14, 15] [16, 17] These data support the hypothesis that HCC, in the setting of increased BMI, may be associated with increased angiogenesis, and thus have higher rates of recurrence and worsened overall survival.
It is possible that those who were obese had more post-operative complications than those who were not. We did not have access to post-operative complication rate, but we did examine length of stay in those who were obese versus those who were not obese, and found no significant difference between them (p=0.56). High BMI is also associated with HCV recurrence, but unfortunately we did not have data on HCV recurrence after transplant. It is possible that this may have also contributed to the worse survival outcomes seen.[18]
One of our hypotheses was that diabetes might also affect tumor aggressiveness via increased angiogenesis, but diabetes was not found to be related to vascular invasion in our dataset. It was also not a significant independent prognostic feature for recurrence or survival in our multivariable analyses. Prospective studies with plasma markers related to insulin resistance and better evaluation of performance status and comorbidities may help to clarify this question in the future.
One limitation to our work is its retrospective nature. A related limitation is that we were unable to obtain data on other variables which play a role in predicting outcome, such as performance status. Further, we could not adjust for the presence of ascites, which could confound BMI measurements. We also did not have etiology of death (cancer vs non cancer). However, we were able to validate and expand our pilot data using a much larger sample size.
In summary, we found a significant relationship between increased BMI and the presence of microvascular invasion in 342 patients who underwent liver transplant for HCC. A BMI>30 was an independent predictor of worse overall survival on multivariable analysis, and also was associated an increased risk of HCC recurrence. In addition, those who were diabetic had worsened overall survival in univariate analyses, but not in multivariable analyses. This may be related to increased wait times for diabetic patients.
Materials and Methods
We reviewed 342 patients who underwent liver transplant for hepatocellular carcinoma between January, 1999 and July, 2010 at our institution. Data were collected from available paper charts and an electronic database using an IRB-approved protocol. We assessed age at transplant, gender, self-reported race/ethnicity, health history (ie hepatitis, diabetes, body mass index (BMI) (defined as weight in kilograms divided by height in meters squared), alcohol use (defined as drinking > than 2 drinks/day over a year or more at any time, or having a primary or secondary diagnosis of alcohol-related liver disease). When more than one BMI was available, we used the one closest to transplant. We defined tumors using AASLD criteria, and recorded tumor characteristics (including being within Milan by imaging, above or below median AFP (using the highest value before transplant), grade, and vascular invasion on explant pathology. We also evaluated time on the wait list, defined as time from listing until transplantation.
Pathologic examination
Microvascular invasion (MVI) was defined as identification of hepatocellular carcinoma tumor cells within endothelial-lined spaces on standard hematoxylin and eosin-stained slides. Grade was evaluated according to Edmondson and Steiner. [19] None of the pathologists was aware of the study hypotheses. A subset of about 10% of slides were centrally reviewed by one pathologist (HR).
Statistical analysis
Chi-square or Fisher exact tests were used to evaluate differences between patients with respect to the above variables. We used Kaplan–Meier curves, multivariable regression models, and Cox proportional hazard models to determine risk factors for recurrence and mortality. Survival was determined from the date of HCC diagnosis defined based on patient history, to the date of death. All statistical analyses were performed with SAS software (Cary, NC).
Acknowledgments
Supported by: NIH K12 (KL2 RR024157-03), NIH K23 (CA149084-01A1), and the Steven J. Levinson Medical Research Foundation (to ABS).
Footnotes
Author Contributions:
Abby B. Siegel: Assisted with funding, planned the study, analyzed data, and wrote the paper
Emerson A. Lim: participated in research design and data analysis
Shuang Wang: participated in data analysis and writing of the paper
William Brubaker: participated data analysis and writing of the paper
Rosa D. Hidalgo: participated in data analysis and writing of the paper
Abhishek Goyal: participated in data analysis and writing of the paper
Judith S. Jacobson: participated in research design, data analysis, and writing of the paper
Elizabeth Verna: participated in writing of the paper
Jonah Zaretsky: participated in data analysis and writing of the paper
Karim Halazun: participated in data analysis and writing of the paper
Lorna Dove: participated in writing of the paper
Robert S. Brown Jr. participated in research design and assisted with funding
Alfred I. Neugut: participated in research design and writing of the paper
Tomoaki Kato: assisted with funding for the study, and helped to write the paper
Helen Remotti: reviewed pathology and helped in writing the paper
Yael J. Coppleson: participated in data analysis and writing of the paper
Jean C. Emond: participated in research design and assisted with funding
References
- 1.Parkin DM, Bray F, Ferlay J, Pisani P. Estimating the world cancer burden: Globocan 2000. Int J Cancer. 2001;94:153–6. doi: 10.1002/ijc.1440. [DOI] [PubMed] [Google Scholar]
- 2.El-Serag HB, Davila JA, Petersen NJ, McGlynn KA. The continuing increase in the incidence of hepatocellular carcinoma in the United States: an update. Ann Intern Med. 2003;139:817–23. doi: 10.7326/0003-4819-139-10-200311180-00009. [DOI] [PubMed] [Google Scholar]
- 3.Siegel AB, Zhu AX. Metabolic syndrome and hepatocellular carcinoma: two growing epidemics with a potential link. Cancer. 2009;115:5651–61. doi: 10.1002/cncr.24687. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.El-Serag HB, Siegel AB, Davila JA, et al. Treatment and outcomes of treating of hepatocellular carcinoma among Medicare recipients in the United States: a population-based study. J Hepatol. 2006;44:158–66. doi: 10.1016/j.jhep.2005.10.002. [DOI] [PubMed] [Google Scholar]
- 5.Siegel AB, McBride RB, El-Serag HB, et al. Racial disparities in utilization of liver transplantation for hepatocellular carcinoma in the United States, 1998–2002. Am J Gastroenterol. 2008;103:120–7. doi: 10.1111/j.1572-0241.2007.01634.x. [DOI] [PubMed] [Google Scholar]
- 6.Siegel AB, Wang S, Jacobson JS, et al. Obesity and microvascular invasion in hepatocellular carcinoma. Cancer Invest. 28:1063–9. doi: 10.3109/07357907.2010.483500. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Kawamura Y, Ikeda K, Arase Y, et al. Diabetes mellitus worsens the recurrence rate after potentially curative therapy in patients with hepatocellular carcinoma associated with nonviral hepatitis. J Gastroenterol Hepatol. 2008;23:1739–46. doi: 10.1111/j.1440-1746.2008.05436.x. [DOI] [PubMed] [Google Scholar]
- 8.Huo TI, Lui WY, Huang YH, et al. Diabetes mellitus is a risk factor for hepatic decompensation in patients with hepatocellular carcinoma undergoing resection: a longitudinal study. Am J Gastroenterol. 2003;98:2293–8. doi: 10.1111/j.1572-0241.2003.07688.x. [DOI] [PubMed] [Google Scholar]
- 9.Poon RT, Fan ST, Wong J. Does diabetes mellitus influence the perioperative outcome or long term prognosis after resection of hepatocellular carcinoma? Am J Gastroenterol. 2002;97:1480–8. doi: 10.1111/j.1572-0241.2002.05792.x. [DOI] [PubMed] [Google Scholar]
- 10.Watt KD, Pedersen RA, Kremers WK, Heimbach JK, Charlton MR. Evolution of causes and risk factors for mortality post-liver transplant: results of the NIDDK long-term follow-up study. Am J Transplant. 10:1420–7. doi: 10.1111/j.1600-6143.2010.03126.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Samuelson AL, Lee M, Kamal A, Keeffe EB, Ahmed A. Diabetes mellitus increases the risk of mortality following liver transplantation independent of MELD score. Dig Dis Sci. 55:2089–94. doi: 10.1007/s10620-010-1267-5. [DOI] [PubMed] [Google Scholar]
- 12.Calle EE, Rodriguez C, Walker-Thurmond K, Thun MJ. Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults. N Engl J Med. 2003;348:1625–38. doi: 10.1056/NEJMoa021423. [DOI] [PubMed] [Google Scholar]
- 13.Rega G, Kaun C, Demyanets S, et al. Vascular Endothelial Growth Factor Is Induced by the Inflammatory Cytokines Interleukin-6 and Oncostatin M in Human Adipose Tissue In Vitro and in Murine Adipose Tissue In Vivo. Arterioscler Thromb Vasc Biol. 2007 doi: 10.1161/ATVBAHA.107.143081. [DOI] [PubMed] [Google Scholar]
- 14.Kanda M, Nomoto S, Nishikawa Y, et al. Correlations of the expression of vascular endothelial growth factor B and its isoforms in hepatocellular carcinoma with clinico-pathological parameters. J Surg Oncol. 2008;98:190–6. doi: 10.1002/jso.21095. [DOI] [PubMed] [Google Scholar]
- 15.Amaoka N, Osada S, Kanematsu M, et al. Clinicopathological features of hepatocellular carcinoma evaluated by vascular endothelial growth factor expression. J Gastroenterol Hepatol. 2007;22:2202–7. doi: 10.1111/j.1440-1746.2006.04790.x. [DOI] [PubMed] [Google Scholar]
- 16.Wang YD, Wu P, Mao JD, Huang H, Zhang F. Relationship between vascular invasion and microvessel density and micrometastasis. World J Gastroenterol. 2007;13:6269–73. doi: 10.3748/wjg.v13.i46.6269. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Li SG, Ye ZY, Zhao ZS, Tao HQ, Wang YY, Niu CY. Correlation of integrin beta3 mRNA and vascular endothelial growth factor protein expression profiles with the clinicopathological features and prognosis of gastric carcinoma. World J Gastroenterol. 2008;14:421–7. doi: 10.3748/wjg.14.421. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Bressler BL, Guindi M, Tomlinson G, Heathcote J. High body mass index is an independent risk factor for nonresponse to antiviral treatment in chronic hepatitis C. Hepatology. 2003;38:639–44. doi: 10.1053/jhep.2003.50350. [DOI] [PubMed] [Google Scholar]
- 19.Edmondson HA, Steiner PE. Primary carcinoma of the liver: a study of 100 cases among 48,900 necropsies. Cancer. 1954;7:462–503. doi: 10.1002/1097-0142(195405)7:3<462::aid-cncr2820070308>3.0.co;2-e. [DOI] [PubMed] [Google Scholar]