Progress Report on cancer research project

Project title
 ： The association of Human Papillomavirus (HPV) infection in occurrence and prognosis of head and neck cancer - a case control and cohort study
Principle investigator
 ： Prof. Josette Sin-yee Chor, Assistant Professor, School of Public Health and Primary Care

Summary of the project

 
Human Papillomavirus (HPV) was suggested in previous overseas studies to be an important risk factor and prognostic factor in head and neck cancer. However, there is no conclusive finding. Since the distribution of HPV genotypes in Hong Kong women is different from that in Caucasians, we need to verify whether the same situation holds in our head and neck cases.
 
This study investigates the distribution of HPV genotypes; and the association of HPV infection with the occurrence and prognosis of head and neck cancer. The study composes of two parts. Part I is a case control study. Patients and control are recruited in surgical clinics. Standardized questionnaires are used to assess any associated epidemiological factors. Tissue and blood sample are taken to assess the HPV infection status of the participants. Prevalence of HPV will be compared in the patients and control group with age, gender and other risk factors taken into account. Pal1 II is a cohort study. Survival data will be traced in these patients. Patients with head and neck cancers will be followed up for 24 months to assess the survival and recurrence rates. Survival and recurrence rates between the HPV positive and HPV negative group will be compared.
 
 
Report of the progress
 
The project has started since Jan 2010. We have recruited 190 patients with head and neck cancer and control with benign lesion which require surgical removal. Questionnaires were administered. Blood and tissue sample were collected and stored in appropriate medium for subsequent analysis. Subjects in the control groups were matched to the case according to age and gender. HPV testing of tissues and blood were initiated.
 
 
Preliminary findings and future work
 
Of the 152 cases with head and neck cancer and control analyzed, the HPV antibody prevalence was 27.7% and 32.6% for cancer group and control group respectively, the HPV prevalence in tissue of cancer group and control group was 12.3% and 11.5%. We will complete the laboratory analysis of the last batch of samples shortly and will continue the analysis of the association of the HPV prevalence with the lifestyle factors. All head and neck cancer patients will be followed up to investigate the significance of the HPV existence in their prognosis.


IB Final Report on the cancer research project

Project title
 ： The association of Human Papillomavirus (HPV) infection in occurrence and prognosis of head and neck cancer - a case control and cohort study
Principle investigator
 ： Prof. Josette Sin-yee Chor, Assistant Professor, School of Public Health and Primary Care

Summary of the project

 
Human Papillomavirus (HPV) was suggested in previous overseas studies to be an important risk factor and prognostic factor in head and neck cancer. However, there is no conclusive finding. Since the distribution of HPV genotypes in cervical samples taken from Hong Kong women is different from that in Caucasians, we need to verify whether the same situation holds in our head and neck cancer cases.
 
This study investigated the distribution of HPV genotypes and the association of HPV infection with the occurrence and prognosis of head and neck cancer. The study composed of two parts. Part I was a case control study. Patients and controls were recruited in surgical clinics (Department of Surgery in Kwong Wah Hospital, North District Hospital, Queen Elizabeth Hospital, United Christian Hospital, and Department of Otorhinolaryngology, Head and Neck Surgery in the Prince of Wales Hospital) from January 2010 to November 2011. All patients recruited were above age of 18. Patients with confirmed diagnosis of head and neck cancer were included in the case group (including esophageal cancer, lip, tongue, floor of mouth, larynx and pharynx cancer). Patients with benign condition in the head and neck region requiring biopsy or surgical removal were recruited as control. Nasopharyngeal cancer, cancer recurrence, past history of head and neck cancer and cancer not of primary origin were excluded. Standardized questionnaires were used to assess any associated epidemiological factors. Tissue and blood samples were taken to assess the HPV infection status of the participants. Prevalence of HPV was compared between the patient and control groups with age, gender and other risk factors taken into account. Chi-square test and multivariable logistic regression were used for the statistical analysis.
 
Part II was a cohort study. Survival data of these patients were traced from medical records. Patients with head and neck cancers were followed up from January 2010 to November 2012 to assess the survival and recurrence rates. Survival and recurrence rates between the HPV-positive and HPV-negative groups were compared. Kaplan-Meier method was adapted for estimating the cumulative survival probability. Statistical analysis was done using SPSS 18.
 
 
Results
 
Out of the 190 subjects recruited, 150 subjects (75 cases, 75 controls) have completed the questionnaire and with both blood and tissue sample available for HPV testing. Demographic and lifestyle characteristics of subjects are shown in Table 1.
 
Of the 150 subjects analyzed, there was no statistically significant difference in the prevalence of HPV between the cancer group and the control group in both serology and HPV DNA. The overall prevalence of HPV antibody (HPV 16 and HPV 18 combined) was 29.3% (22/75) in both groups. The prevalence of HPV DNA (all types combined) in tissue of cancer group and control group was 9.3% (7/75) and 16% (12/75), respectively. Among the cancer subjects positive for HPV antibody, 68.2% (15/22) were HPV16 antibody positive, 13.6% (3/22) were HPV18 antibody positive, and 18.2% (4/22) were both HPV16 and HPV18 antibody positive. In control group, among subjects with HPV antibody positive, 45.5%(10/22) were HPV16 antibody positive, 31.8% (7/22) were HPV18 antibody positive, and 20.5% (5/22) were both HPV 16 and HPV 18 antibody positive. Altogether four genotypes, HPV 6, HPV 16, HPV 51, HPV 52, were detected in tissue samples of seven cancer patients. Among the four HPV genotypes found, HPV 16 was the most common, and being found in four out of seven (57%) infected subjects. The HPV genotype distribution in the control group was similar to the cancer group, where HPV16 was also the most common genotype, 10 out of 12 (83%) in the infected control group. Other genotypes detected included HPV51 and HPV52 (See Table 2 for HPV prevalence & Table 3 for HPV genotype distribution). The HPV prevalence between gender, age-groups and subjects with different lifestyle factors (smoking and alcohol usage) were compared, no significant differences were observed with respect to the HPV antibody or DNA results (See Table 2).
 
Logistic regression was used to further evaluate the associations between HPV and the occurrence of head and neck cancer. Subjects were divided into 4 groups based on the HPV test results, the HPV antibody negative (-ve) and HPV DNA -ve group, the HPV antibody positive (+ve) and HPV DNA -ve group, the HPV antibody -ve and HPV DNA +ve group, and the HPV antibody +ve and HPV DNA +ve group. Gender, age-group, education level, smoking status, usage of alcohol were adjusted. According to previous studies, HPV were found to be associated with oral-pharyngeal-laryngeal cancers, therefore the multivariable analysis were done in each diagnosis subgroup separately (esophageal group and oral-pharyngeal-laryngeal group). However, there were only 35 subjects in the esophageal group. The limited sample size did not have enough power to describe the associations between HPV and esophageal cancer. In the oral-pharyngeal-laryngeal group， there were no statistically significant evidence to support that HPV antibody positive or HPV DNA positive was a risk factor for head and neck cancer (See Table 4).
 
To evaluate the associations between HPV and prognosis of head and neck cancer， cancer patients were followed up (n=75). The median follow up time was 661 days (25th Quartile-l80 days, 75th Quartile -816 days).
During the follow up period, the overall recurrence rate was 14.5% (9/62). The recurrence rate of the HPV antibody -ve and HPV DNA -ve group and the HPV antibody +ve and HPV DNA -ve group were 14.6% (6/41) and 20%(3/15), respectively. The other two HPV groups had zero recurrence rate, however, their group size were too small for any meaningful comparison and conclusion (See Table 5). There was no significant difference in recurrence rate among groups with different HPV status within the oral-pharyngeal-laryngeal subgroup either (See Table 5).
 
Among the 73 cancer patients with survival record available, 37 patients died during the follow up period (50.3%). Again, the mortality of the patients from esophageal group and oral-pharyngeal-laryngeal group were significantly different (See Table 5), therefore the survival analysis were performed separately. Since smoking was one of the poor prognostic factors of head and neck cancer found in previous studies, subgroup survival analysis was done to evaluate the interactions between smoking and HPV on head and neck cancer mortality. To estimate the cumulative survival probability by the end of the second year of follow up, Kaplan-Meier method was used. Cancer subjects in different HPV groups were compared.
 
In oral-pharyngeal-laryngeal group, there was only one patient in the HPV antibody +ve and HPV DNA +ve group therefore only the other three groups were compared.
 
In the Ever-smoker subgroup, the cumulative survival probability was the highest in the HPV antibody -ve and DNA -ve group (0.714), followed by the HPV antibody +ve and HPV DNA -ve group (0.60) and then the HPV antibody -ve and HPV DNA +ve group (0.25). The cumulative proportion of survivals of the three groups were statistical significantly different (p-value of Log-Rank test = 0.019) (See Table 6 and Figure 1a).
 
In the Non-smoker group, the cumulative survival probability was again the highest in the HPV antibody -ve and DNA -ve group (0.844), followed by the HPV antibody +ve and HPV DNA -ve group (0.617) and then the HPV antibody -ve and HPV DNA +ve group (0.5). However, the cumulative survival probability between different HPV groups were not significant different in the non-smoker subgroup (See Table 6 and Figure 1b).
 
In esophageal group, there were no subject in the HPV antibody and HPV DNA +ve group and the HPV antibody -ve and HPV DNA +ve group. Therefore, only two groups, HPV antibody +ve and HPV DNA -ve group and the HPV antibody and DNA negative group, were compared.
 
In the ever-smoker subgroup, the cumulative survival probabilities of the two groups were comparable, which were 0.25 and 0.33 for the HPV antibody +ve and HPV DNA -ve group and the HPV antibody and DNA -ve group respectively (See Table 6 and Figure 2a). In non-smoker subgroup, although the cumulative survival probability of both HPV group were zero, there were only five patients in this subgroup, therefore the cumulative probability of this subgroup was not conclusive (See Table 6 and Figure 2b).
 
 
Summary
 
No significant differences were found in the prevalence of HPV antibody and HPV DNA between cancer patients and controls. This study did not reveal evidence to support HPV infection is a risk factor of head and neck cancer in Hong Kong. However, oral-pharyngeal-laryngeal cancer patients, who were ever-smoker, with HPV antibody negative and HPV DNA positive in tissue were having lower probability of surviving 2 years after diagnosis, compared to patients with both HPV antibody negative and HPV DNA negative in tissue. Further study with a larger sample size is worthwhile to confirm the findings.

Table I. Demographic characteristics and lifestyle characteristics

 

Number of subjects (%)

Case

Control

Total

 

75

75

150

Demographic characteristics

Gender

Male

52(66.3)

38(50.7)

90(60)

 

Female

23(30.7)

37(49.3)

60(40)

 

Age

Mean (SD)

62.4(13.6)

58.1(14.3)

 

 

Age group

18-49

13(17.3)

25(33.3)

38(25.3)

 

50-69

35(46.7)

29(38.7)

64(42.7)

 

70 or above

27(36)

21 (28)

48(32)

 

Education level

Primary or below

38(50.7)

30(40)

68(45.3)

 

Secondary

28(37.3)

41(54.7)

69(46)

 

Tertiary or above

9(12)

4(5.3)

13(8.7)

 

Household
monthly income

$5,000 or below

32(45.7)

29(40.3)

61(43)

 

$5,001-10,000

14(20)

13(18.1)

27( 19)

 

$10,001-20,000

16(22.9)

16(22.2)

32(22.5)

 

$20,001-30,000

6(8.6)

8(11.1)

14(9.9)

 

$30,001 or above

2(2.9)

6(8.3)

8(5.6)

 

Lifestyle characteristics

Smoking status

Non-smoker

35(46.7)

54(73)

89(59.7)

 

Ex-smoker

33(44)

16(21.6)

49(32.9)

 

Current Smoker

7(9.3)

4(5.4)

II (7.4)

Chi-Square test for independence P<0.01

Alcohol user

No

33(44.6)

49(66.2)

82(55.4)

 

Yes

41 (55.4)

25(33.8)

66(44.6)

Chi-Square test for independence P<0.01

Diagnosis

Group

Esophageal

22(29.3)

11(14.7)

33(22)

 

Oral-pharyngeal-laryngeal

53(70.7)

64(85.3)

117(78)


Table 2. Prevalence of HPV

 

HPV Prevalence (no. of subject)

HPVantibody

Male

Female

Overall

Case (n=52)

Control
(n=38)

Case
(n=23)

Control
(n=37)

Case (n=75)

Control
(n=75)

Overall prevalence

15 (13)

34.2 (13)

39.1(9)

24.3(9)

29.3(22)

29.3(22)

Prevalence in Subgroup

 

Age group

18-49

27.3(3/11)

28.6(4/14)

0(0/2)

36.4(4/11)

23.1 (3/ 13)

32(8/25)

 

50-69

25(6/24)

47.1 (8/17)

63.6(7/11)

25(3/12)

37. 1 ( 13/35)

37.9(11/29)

 

70 or above

23.5(4/17)

14.3(117)

20(2/10)

14.3(2/14)

22.2(6/27)

14.3(3/21)

 

 

 

 

Smoking
status

Non-smoker

25(3/12)

33.3(7/21)

39.1 (9/23)

18.2(6/33)

34.3(12/35)

24.1(13/54)

 

Ever-smoker

25(10/40)

37.5(6/16)

NA

75(3/4)

25(10/40)

45(9/20)

 

 

 

 

Alcohol
usage

No

23.1 (3/ 13)

26.7(4/15)

35(7/20)

23.5(8/34)

30.3(10/33)

24.5(12/49)

 

Yes

25.6(10/39)

40.9(9/22)

50(1/2)

33.3(1/3)

26.8(11/41)

40(10/25)

 

 

 

 

Diagnosis

Esophageal

22.2(4/18)

33.3(1/3)

25(1/4)

37.5(3/8)

22.7(5/22)

36(4/11)

 

Oral-pharyngeal-laryngeal

26.5(9/34)

34.3(12/35)

42.1 (8/19)

20.7(6/29)

32(17/53)

28.1(18/64)

 

 

 

 

HPV DNA in tissue

Male

Female

Overall

Overall

Case (n=52)

Control
(n=38)

Case
(n=23)

Control
(n=37)

Case (n=75)

Control
(n=75)

11.5(6)

18.4(7)

4.3(1)

13.5(5)

9.3(7)

16(12)

Prevalence in Subgroup

 

 

 

Age group

18-49

27.3(3/11)

28.6(4/14)

0(0/2)

0.27(3/11)

23.1 (3/13)

28(7/25)

 

50-69

4.2(1/24)

11.8(2/17)

9.1(1/11)

0(0/12)

5.7(2/35)

6.9(2/29)

 

70 or above

11.8(2/17)

14.3(1/7)

0(0/10)

14.3(2/14)

7.4(2/27)

14.3(3/21)

 

 

 

 

Smoking
status

Non-smoker

16.7(2/12)

23.8(5/21)

4.3(1/23)

15.2(5/33)

8.6(3/35)

18.5(10/54)

 

Ever-smoker

10(4/40)

12.5 (2/16)

NA

0(0/4)

10 (4/40)

10(2/20)

 

 

 

 

Alcohol
usage

No

7.7 (1/13)

6.7(1/15)

5(1/20)

11.8 (4/34)

6.1 (2/33)

10.2 (5/49)

 

Yes

12.8 (5/39)

27.3 (6/22)

0(0/2)

33.3 (1/3)

12.2(5/41)

28(7/25)

 

 

 

 

 

 

 

 

Diagnosis

Esophageal

0(0/18)

0/(0/3)

0/(0/4)

0(0/8)

0(0/22)

0(0/11 )

 

Oral-pharyngeal-laryngeal

17.6(6/34)

20(7/35)

5.3(1/19)

17.2(5/29)

13.2(7/53)

18.8( 12/64)


Table 3. HPV genotype distribution

 

Number of subject (%)

HPVantibody

HPV Ab genotype

Case (n=22)

Control (n=22)

Total (n=44)

 

HPV16

15(68.2)

10(45.5)

25(56.8)

 

HPV18

3(13.6)

7(31.8)

10(22.7)

 

HPV16 & HPV18

4(18.2)

5(22.7)

9(20.5)

 

 

 

 

 
HPV DNA in tissue

HPV DNA genotype

Case (n=7)

Control (n= 1)

Total (n=19)

 

HPV6

1(15.3)

0

1(5.3)

 

HPVI6

4(57)

10(83)

14(73.7)

 

HPV51

1(15.3)

0

1(5.3)

 

HPV52

0

1(8.5)

1(5.3)

 

HPV16, HPV51

0

1(8.5)

1(5.3)

 

HPV51, HPV52

1(15.3)

0

1(5.3)