Journal of Surgical Oncology 2010;102:338–341

The Clinical Utility of Ki-67 in Assessing Tumor Biology and Aggressiveness in

Patients With Appendiceal Carcinoids

ERIC LIU, MD,1 DANA A. TELEM, MD,1 JOHN HWANG, BA,1 RICHARD R.P. WARNER, MD,2

ANDREW DIKMAN, MD,3 AND CELIA M. DIVINO, MD, FACS1*

1Division of General Surgery, Department of Surgery, Mount Sinai School of Medicine, New York, New York2Division of Gastroenterology, Department of Medicine, Mount Sinai School of Medicine, New York, New York

3Department of Pathology, Mount Sinai School of Medicine, New York, New York

Background/Objective: To elucidate the correlation of Ki-67 with tumor biology and survival in appendiceal carcinoid tumors.

Method: A retrospective chart review conducted on 51 patients with appendiceal carcinoid tumors who underwent surgical intervention from

1991 to 2008. MIB-1, an antibody of Ki-67, was used to determine cell proliferation and correlated with clinical and histological parameters.

MIB-1 index was categorized according to the World Health Organization (WHO) classification.

Result: Of the 51 patients, 32 had tumors <2 cm; 3 >2 cm; and 16 with unspecified tumor size. Increased MIB proliferative index did not

significantly correlate with increasing tumor size (P¼ 0.426). Twelve patients had metastatic disease on presentation: 9 had MIB-1 index <2%,

1 had index 2–15% and 2 with index >15%. No significant correlation between MIB index and metastasis was demonstrated (P¼ 0.68). Median

follow-up was 40 months (range 10–183 months) with a 51% follow-up rate. Seven mortalities and three recurrences presented in 26 patients.

Assessment of survival demonstrated significantly decreased survival by increasing MIB index. Survival rate by MIB index was as follows: <2%

was 97%, 2–15% was 85% and >15% was 67% (P¼ 0.02).

Conclusion: Increased MIB index significantly correlated with decreased survival. No correlation was demonstrated by MIB index and tumor size

or presentation with metastatic disease.

J. Surg. Oncol. 2010;102:338–341. � 2010 Wiley-Liss, Inc.

KEY WORDS: appendiceal; carcinoid; Ki-67; correlation

INTRODUCTION

Carcinoid tumors of the appendix are rare and often discovered

incidentally during appendectomy. These tumors account for up to

60% of all appendiceal tumors, and while many are confined to the

appendix, they can be associated with metastatic disease [1]. Current

treatment guidelines proposed by Moertel et al. in 1987 are based on

tumor size with a proposed increase in the risk of metastasis for tumors

greater than 2.0 cm. The guidelines stipulate treatment of lesions

smaller than 2.0 cm by simple appendectomy and lesions larger than

2.0 cm by right hemicolectomy (RHC) [2–6]. Survival rate of patients

with appendiceal carcinoid tumors are 86% at 5 years and highest

among all GI carcinoid tumors [1]. In a recent study of appendiceal

carcinoid tumors, based on the Surveillance Epidemiology and End

Results Program (SEER) database, tumor size, patient age, presence of

metastasis, and histological cell type were all found to correlate with

survival [1,7,8]. A limitation of this population based study, however,

was that it did not account for tumor biology data, such as cell

proliferative rate which has been shown to correlate with the outcome

in pancreatic and ileocecal carcinoids [8–12].

The Ki-67 protein, a high molecular weight nuclear protein

antigen structurally associated with chromatin [11], is an excellent

marker to measure growth fraction as it is strictly associated with

cell proliferation and not with DNA repair [13]. MIB-1, a highly

specific monoclonal antibody to Ki-67, is commonly used to deter-

mine Ki-67 proliferative index and can be used on formalin-fixed

paraffin-embedded sections [14]. Some studies have suggested that the

prognostic value of Ki-67 depends on tumor type and therefore

conclusions on its utility should not be generalized [15]. The use of

MIB-1 has gained popularity in the workup of carcinoid tumors in

recent years and this study aims to elucidate its correlation of Ki-67

with appendiceal carcinoid tumors in terms of tumor biology,

management, and prognosis.

METHODS

An institutional review board approved retrospective review

identified 51 patients with carcinoid tumor involving the appendix

who underwent operative intervention at The Mount Sinai Medical

Center from 1991 to 2008. Cases were identified through the hospital

electronic medical records and pathology department databases.

Records were reviewed with respect to patient demographics, medical

history, presenting symptoms, type of surgical intervention, outcome,

tumor characteristics with an emphasis on tumor biology and presence

of metastatic disease. Metastatic disease was defined by presence of

nodal metastasis on pathology or by evidence of distant metastasis.

All patients in this study classified as having metastatic disease had

lymph nodes positive for tumor with or without evidence of distant

metastasis. Study exclusion criteria were as follows: age less then 18,

goblet cell pathology and tumors unable to perform MIB immuno-

staining. Follow-ups were obtained through electronic patient records

or telephone interview with patients or immediate family members.

*Correspondence to: Celia M. Divino, MD, FACS, The Stanley Edelman,MD Professor of Surgery, Chief, Division of General Surgery, ProgramDirector, General Surgery Residency, The Mount Sinai Medical Center, E.98th St. Box 1259, 15th floor, New York, NY 10029. Fax: 212-534-2654.E-mail: celia.divino@mountsinai.org

Received 25 February 2010; Accepted 26 April 2010

DOI 10.1002/jso.21634

Published online 6 July 2010 in Wiley Online Library(wileyonlinelibrary.com).

� 2010 Wiley-Liss, Inc.

Tumor proliferation rates were assessed by immunostaining for

the nuclear antigen MIB-1 (Ki-67). MIB-1 immunostaining was

performed on paraffin sections and allowed for retrospective staining of

stored pathological specimens. The MIB-1 Labeling index was defined

as the percentage of positive tumor cell nuclei: MIB-1 labeling

index (%)¼ number of cells with positive nuclear staining/total

number of tumor cells� 100. The MIB-1 indices were then categorized

according to World Health Organization (WHO) classification of GI

endocrine tumors: (1) well-differentiated endocrine tumor (carcinoid)

<2% Ki-67 positive cells, (2) well-differentiated endocrine carcinoma

(malignant carcinoid) 2–15% Ki-67 positive cells, and (3) poorly

differentiated endocrine carcinoma (small cell carcinoma) >15%

Ki-67 positive cells [10,11]. Indices were correlated with clinical and

histological parameters such as age, gender, tumor type excluding

goblet cell carcinoid tumor, presence of metastatic disease, time

elapsed until disease recurrence, and survival. In order to see the

relationship between tumor size and MIB-1 index, we categorized

tumor size according to the surgical treatment guideline proposed

by Moertel et al. [2]. All samples were processed and stained at

The Mount Sinai Hospital and interpreted by a single faculty, board-

certified pathologist who specializes in tumors of the gastrointestinal

tract.

Analysis was conducted by unpaired student t-test with two tail

distribution for quantitative variables and fisher exact test for

categorical values. Kaplan–Meier survival curve was used to analyze

the survival in patients. P values of <0.05 were considered to confer

significance. All statistical analysis were performed using GraphPad

Prism version 4.00 for Windows, GraphPad Software, San Diego

California USA, www.graphpad.com.

RESULTS

Patient Demographics

Of 51 patients, the female to male ratio was 1.8:1, with 69% of

the patients of Caucasian origin (Table I). The median age was 48 years

old with range from 20 to 85 years. The admitting history and

physical examination were available in 26 patients. Seventeen

patients presented with acute abdominal pain with 10 patients having

a preoperative diagnosis of acute appendicitis. Four patients presented

with carcinoid syndrome.

Histopathology

Fifty-five percent of tumors had typical histological presentation of

argentaffin cells and the tip of the appendix was the most common

tumor location (39%). Among 51 patients (Table II), 32 had tumor size

<2 cm (63%), 3 greater than 2 cm (6%) and 16 with unspecified size

(31%). Each group was subdivided according to WHO classification on

MIB-1 index. Overall, 80% (41) of tumors had a proliferative index

<2%. No significant correlation was demonstrated between increasing

tumor size and MIB-1 index (Spearman r¼ 0.027; P¼ 0.85).

Depth of tumor invasion at diagnosis was available in 31 of

51 patients and their relationship with MIB-1 index was examined

(Table III). Four tumors invaded to the serosa, 11 to the periap-

pendiceal fat and 3 to mesoappendix by the time of diagnosis. Only

one tumor showed lymphovascular invasion. While of the majority of

tumors (74%) stained MIB-1 index <2% including the one with

lymphovascular invasion, there was no significant relationship between

increasing MIB-1 index and tumor invasion at diagnosis (P¼ ns). Of

the 12 patients were diagnosed with metastatic disease on presentation

(Table IV). Nine of these tumors (75%) presented with MIB-1 index

<2%. Twelve patients did not show any correlation in MIB-1 index

expression.

Management and Follow-Up

Table V summarizes patient operative management. Sixteen

patients underwent sole appendectomy, 15 right hemicolectomy,

14 ileocolic resections, 4 subtotal colectomies, 1 total proctocolect-

omy, and 1 unspecified surgery. Of the 51 patients, 41 (80%) had MIB-

1 index less than 2%. Eleven patients (22%) received more extensive

resections due to the co-existing neoplasm in their medical history and

one patient died in few hours after subtotal colectomy from multiple

co-morbidities and sepsis. MIB-1 index did not correlate with the

extent of resection patient received.

Follow up was available in 26 patients. Median follow up time

was 40 months with range of 10–183 months. Recurrence was

diagnosed in three patients and all their MIB-1 indices were less

than 2%. These patients were diagnosed with liver, bone, and lung

metastasis with relapse time of 1, 20, and 183 months. Seven

mortalities were reported. Among the seven deaths, four patients had

MIB-1 <2%; one with MIB-1 between 2% and 15% and two with

MIB-1 >15% (50% and 75% respectively). A significant decrease

in survival (Fig. 1) was demonstrated by increased MIB-1 index

(P¼ 0.02).

Journal of Surgical Oncology

TABLE I. Demographic Data for 51 Patients With Appendiceal Carcinoid

Tumors

No. (%)

Age

Range 20–85

Mean 48.5

Sex

Male 18 (35.3)

Female 33 (64.7)

Race/ethnicity

White 35 (68.6)

Black 2 (3.9)

Hispanic 4 (7.8)

Unspecified 10 (19.6)

Tumor cell types

Typical 28 (54.9)

Tubular 4 (7.8)

Unspecified 19 (37.3)

Location

Tip 20 (39.2)

Base 1 (2.0)

Wall 12 (23.5)

Unspecified 18 (35.3)

TABLE II. Tumor Size and MIB-1 Index

MIB <2% 2–15% >15%

Tumor <2 cm 27 (84%) 4 (13%) 1 (3%)

>2 cm 3 (100%) 0 0

Unspecified size 11 (69%) 2 (13%) 3 (18%)

TABLE III. Depth of Tumor Invasion and MIB-1 Index

MIB-1 <2% 2–15% >15%

Localized 5 (56%) 3 (33%) 1 (11%)

Transmural 3 (100%) 0 0

Serosal 3 (75%) 0 1 (25%)

Periappendiceal fat 10 (91%) 0 1 (9%)

Mesoappendix 1 (33%) 1 (33%) 1 (33%)

Lymphovascular 1 (100%) 0 0

Assess Appendiceal Carcinoid With Ki-67 339

DISCUSSION

Ki-67, a high molecular weight protein antigen, is located in the

nucleolus and closely associated with chromatin. Its expression is

highly regulated and essential during cell proliferation [11,13,15,17].

Although its exact function was not well understood, Ki-67 had been

widely utilized to measure the mitotic activities of cell proliferation

and as a prognostic factor for various types of cancer [15,17].

Numerous studies have tried to correlate Ki-67 with known prognostic

factors, for example, tumor size, presence of metastasis and patient

survival. The proliferative index shows positive correlation with tumor

size and metastatic disease presentation as well as negative correlation

with survival in pancreatic endocrine tumor and gastrointestinal

neuroendocrine tumors [11,13,15,16,18]. To the authors’ knowledge,

however, this is the first study that examined the relationship between

Ki-67 and appendiceal carcinoid tumors.

This study demonstrated a decrease in survival by increased MIB-1

index. This is similar to data from gastrointestinal neuroendocrine

tumors and pancreatic endocrine tumors as well. High MIB-1 index

indicates increased mitotic activities and attributes to increased tumor

aggressiveness. In the indolent course of appendiceal carcinoid tumor,

the tumor aggressiveness can result in decrease in long term patient

survival. With limited number of patients with MIB-1 index >2% in

our study, the proliferative index correlate with decreased survival in

appendiceal carcinoid patients.

In contrast to available studies [18,19], our study did not

demonstrate an association with tumor size, depth of tumor invasion

or presence of metastatic disease at time of diagnosis. The discrepancy

might arise from the differences in power of study with 51 appendiceal

carcinoid tumors in our study comparing to 2 appendiceal lesions in

another study [18]. We further postulated that the diversity might be

due to the heterogeneity of GI carcinoid tumors presenting different

degrees of tumor aggressiveness. The heterogeneity is demonstrated in

different probability of metastasis with increasing probability from

duodenum to ileum in the small intestine and highest probability

in colonic carcinoid. Different tumor aggressiveness of carcinoid

tumors at various sites also attributes to different patient survival with

lowest survival rate of 33% at sigmoid colon comparing to 86% at

appendix [20,21]. In addition, the presence of carcinoid syndrome,

which is prevalent in midgut carcinoid with liver metastasis, is

associated with higher risk of mortality and it is not normally seen in

appendiceal carcinoid [22]. The different tumor characteristics

in carcinoid tumors result in the discrepancy between our study

and previous studies on gastrointestinal neuroendocrine tumors and

pancreatic endocrine tumors.

Limitations of this study include the retrospective study design and

sample size. Appendiceal carcinoid tumor is rare and indolent diseases

with majority of cases are an incidental finding. In addition, MIB-1

proliferative index has only recently been used in pathologic tumor

assessment. Another limitation is that the majority of patients had

MIB index <2%. Future studies directed at increasing sample size

particularly in tumors with MIB-1 index >2% should be performed.

CONCLUSION

This study demonstrated Ki-67 (MIB-1) to be significantly

associated with decreased survival. Ki-67, however, did not correlate

with tumor size, depth of invasion or presentation with metastatic

disease. The discrepancy definitely warrants further study of this

cellular marker.

ACKNOWLEDGMENTS

Eric Liu, Dana Telem and John Hwang attest to full access of all

data in the study and take responsibility for the integrity of the data and

the accuracy of the data analysis.

REFERENCES

1. Sandor A, Modlin IM: A retrospective analysis of 1570appendiceal carcinoids. Am J Gastroenterol 1998;93:422–428.

2. Moertel CG, Weiland LH, Nagorney DM, et al.: Carcinoid tumorof the appendix: Treatment and prognosis. N Engl J Med 1987;317:1699–1701.

3. Roggo A, Wood WC, Ottinger LW: Carcinoid tumors of theappendix. Ann Surg 1993;217:385–390.

Journal of Surgical Oncology

TABLE IV. Metastatic Disease at Time of Diagnosis and MIB-1 Index

Mets (%) No mets (%)

MIB-1< 2% 9 (22) 32 (78)

2–15% 1 (17) 5 (83)

>15% 2 (50) 2 (50)

TABLE V. Appendiceal Carcinoid Tumors Management

MIB-1< 2% 2–15% >15%

Appendectomy (16)a

<2 cm 8 2 0

>2 cm 0 0 0

Unspecified 4 2 0

Right hemicolectomy (15)b

<2 cm 7 0 0

>2 cm 2 0 0

Unspecified 6 0 0

Ileocolic resection (14)c

<2 cm 10 1 1

>2 cm 1 0 0

Unspecified 0 0 1

Subtotal colectomy (4)d

<2 cm 0 1 0

>2 cm 0 0 0

Unspecified 3 0 0

aThree patients with co-existing ovarian adenocarcinoma also received total

hysterectomy/bilateral salphingoophorectomy (MIB index are all 0%).bSix patients had co-existing malignant neoplasm of right colon.cTen patients had history of Crohn’s disease.dOne patient died in few hours postoperatively from multiple co-morbidities and

sepsis.eOne total proctocolectomy was performed on patient with MIB-1 index 90%

and history of rectal cancer and ulcertative colitis.fOne patient with MIB-1 index 75% had unspecified surgery.

Fig. 1. Patient survival with different MIB-1 index.

340 Liu et al.

4. Moertel CG: Treatment of the carcinoid tumor and the malignantcarcinoid syndrome. J Clin Oncol 1983;1:727–740.

5. Moertel CG, Dockerty MB, Judd ES: Carcinoid tumors of thevermiform appendix. Cancer 1968;21:270–278.

6. Bamboat ZM, Berger DL: Is right hemicolectomy for 2.0-cmappendiceal carcinoids justified? Arch Surg 2006;141:349–352.

7. van Gompel JJ, Sippel RS, Warner TF, et al.: Gastrointestinalcarcinoid tumors: Factors that predict outcome. World J Surg2004;28:387–392.

8. Landry CS, Woodall C, Scoggins CR, et al.: Analysis of 900appendiceal carcinoid tumors for a proposed predictive stagingsystem. Arch Surg 2008;143:664–670, discussion 670.

9. Bettini R, Boninsegna L, Mantovani W, et al.: Prognostic factorsat diagnosis and value of WHO classification in a mono-institutional series of 180 non-functioning pancreatic endocrinetumors. Ann Oncol 2008;19:903–908.

10. La Rosa S, Klersy C, Uccella S, et al.: Improved histologic andclinicopathologic criteria for prognostic evaluation of pancreaticendocrine tumors. Hum Pathol 2009;40:30–40.

11. Rorstad O: Prognostic indicators for carcinoid neuroendocrinetumors of the gastrointestinal tract. J Surg Oncol 2005;89:151–160.

12. Cunningham JL, Grimelius L, Sundin A, et al.: Malignantileocaecal serotonin-producing carcinoid tumours: The presenceof a solid growth pattern and/or Ki67 index above 1% identifiespatients with a poorer prognosis. Acta Oncol 2007;46:747–756.

13. Vilar E, Salazar R, Perez-Garcıa J, et al.: Chemotherapy and roleof the proliferation marker Ki-67 in digestive neuroendocrinetumors. Endocr Relat Cancer 2007;14:221–232.

14. Cattoretti G, Becker MH, Key G, et al.: Monoclonal antibodiesagainst recombinant parts of the Ki-67 antigen (MIB 1 and MIB3) detect proliferating cells in microwave-processed formalin-fixed paraffin sections. J Pathol 1992;168:357–363.

15. Brown DC, Gatter KC: Ki67 protein: The immaculate deception?Histopathology 2002;40:2–11.

16. Kloppel G, Perren A, Heitz P: The gastroenteropancreaticneuroendocrine cell system and its tumors: The WHO classi-fication. Ann NY Acad Sci 2004;1014:13–27.

17. Gerdes J, Lemke H, Baisch H, et al.: Cell cycle analysis of acell proliferation-associated Human nuclear antigen definedby the monoclonal antibody Ki-67. J Immunol 1984;133:1710–1715.

18. Kawahara M, Kammori M, Kanauchi H, et al.: Immunohisto-chemical prognostic indicators of gastrointestinal carcinoidtumors. Eur J Surg Oncol 2002;28:140–146.

19. Sokmensuer C, Gedikoglu G, Uzunalimoglu B: Importance ofproliferation markers in gastrointestinal carcinoid tumors: Aclinicopathologic study. Hepatogastroenterology 2001;48:720–723.

20. Godwin JD II: Carcinoid tumors: An analysis of 2837 cases.Cancer 1975;36:560–569.

21. Hof KH, van der Wal HC, Kazemier G, et al.: Carincoid tumor ofthe appendix: An analysis of 1485 consecutive emergencyappendectomies. J Gastrointest Surg 2008;12:1436–1438.

22. Janson ET, Holmberg L, Stridsberg M, et al.: Carcinoid tumors:Analysis of prognostic factors and survival in 301 patients from areferral center. Ann Oncol 1997;8:685–690.

Journal of Surgical Oncology

Assess Appendiceal Carcinoid With Ki-67 341