Cigarette smoking and malignant melanoma: a case-control study

1
Cigarette smoking and malignant melanoma a
case-control study

• American Federation for Medical Research Eastern
  Regional Meeting, Washington D.C.
• April 15, 2009

Presented by Maria Christina Kessides M.S.
2
Disclosures

• This work is funded by the Doris Duke Charitable
  Foundation

3
Background

• Smoking is well-known to be associated with
  cancers of several internal organs, including the
  lung, bladder, pancreas, larynx and cervix
• Knowledge of numerous carcinogens in cigarette
  smoke and their link to cancers of internal
  organs eventually led to studies on smoking and
  skin cancer (NMSC and melanoma)

4
Previous studies Ever smoker and melanoma risk
OR or RR
5
Previous studies Former smoker and melanoma risk
OR or RR
6
Previous studies Current smoker and melanoma
risk
OR or RR
7
Distribution of adjusted variables in previous
studies
8
Background

• Inverse association is somewhat puzzling,
  especially in light of well-known risks of
  smoking
• High probability of a confounder
• Inconsistent controlling for skin type and
  history of UV exposure, or history of blistering
  sunburns among previous studies

9
Hypothesis

• When carefully adjusting for sun exposure, skin
  type, and history of blistering sunburns,
  cigarette smoking will either
• Show no association with risk for melanoma
• OR
• 2. Confer an increased risk for melanoma
• OR
• 3. Show a persistent inverse association

10
Materials/Methods the CLUE database

• CLUE II est. in 1989 by the Johns Hopkins School
  of Public Health
• 32,000 volunteers gave blood samples, social and
  medical histories
• Follow-up questionnaires sent in 1996, 1998,
  2000, 2003, 2007
• 2007 questionnaire collected information on hours
  spent outdoors, tanning salon use, history of
  sunburning, and sun protection practices

Source www.jhsph.edu/comstockcenter
11
Materials/Methods study design

• Case-control study
• Over 8,000 subjects responded to the 2007
  questionnaire
• Identified 103 confirmed cases of malignant
  melanoma who fulfilled our inclusion and
  exclusion criteria

• Matching criteria
• Each case matched to two controls along the
  following variables
• Race
• Age ( 5 years)
• Sex
• Skin type

• Statistics Chi-square and logistic regression
  analysis

12
Study Design

• Inclusion criteria
• Cases confirmed history of melanoma (Maryland
  tumor registry)
• Controls no history of melanoma

• Exclusion criteria
• Cases with a history of another cancer diagnosed
  before melanoma
• Controls with history of melanoma or confirmed
  history of any other cancer
• Anyone missing matching data

13
Methods skin typedetermination
If you went into the sun at noon for 1 hour in
the summer
ANSWERS Fitzpatrick Skin Type Skin Type description
a. Blistering sunburn I Always burn, no tan
b. Sunburn w/o blisters I/II Always burn, no tan/Always burn, sometimes tan
c. Mild sunburn w/o blisters III Sometimes burn/always tan
d. Tan or darken w/o burn IV Never burn/always tan
e. No change in color V/VI Never burn/tan profusely
14
Methods determiningnumber of blistering
sunburns

• How many blistering sunburns have you had in a
  lifetime
• None
• 1 or 2
• 3 or 4
• 5-9
• 10

15
Methods determining UV exposure history
On a typical weekday/weekend in the summerhow
many hours (w/o sun protection) in the midday
sun?
Teens 20s 30s Last 10 yrs
LOW 0 0 0 0
LOW 1 1 1 1
MEDIUM 2 2 2 2
MEDIUM 3 3 3 3
HIGH 4 4 4 4
HIGH 5 5 5 5
HIGH 6 6 6 6
16
ResultsUV exposure history-WEEKENDS
Case Control p value
Exposure to midday sun - teens N 87 N 186 0.97
Low 14.9 16.1
Medium 32.2 32.3
High 52.9 51.6

Exposure to midday sun - 20s N 88 N 189 0.89
Low 15.9 18.0
Medium 45.5 45.5
High 38.6 36.5

Exposure to midday sun - 30s N 89 N 188 0.19
Low 21.4 30.9
Medium 51.7 40.4
High 21.4 28.7

Exposure to midday sun - past decade N 90 N 193 0.56
Low 54.4 47.7
Medium 30.0 33.7
High 15.6 18.7
Excluding subjects for which there was missing
data
17
ResultsUV exposure history-WEEKDAYS
Case s Control p value
Exposure to midday sun - teens N 91 N 187 0.89
Low 15.4 15.5
Medium 35.2 38.0
High 49.5 46.5

Exposure to midday sun - 20s N 92 N 188 0.54
Low 29.4 29.8
Medium 35.9 41.5
High 34.8 28.7

Exposure to midday sun - 30s N 91 N 188 0.31
Low 42.9 46.8
Medium 44.0 35.1
High 13.2 18.1

Exposure to midday sun - past decade N 92 N 192 0.92
Low 59.8 59.4
Medium 29.4 28.1
High 10.9 12.5
Excluding subjects for which there was missing
data
18
Results Sunburn history andsun protective
practices
Cases Controls P value
Total blistering sunburns in lifetime N 99 N 195 0.06
Never 29.3 33.9
1-4 43.4 51.8
5-9 19.2 9.2
10 8.1 5.1

Sunscreen Use N 103 N 206 0.13
Regular 29.1 21.4
Non-regular 70.9 78.4

Sun-protective clothing use N 100 N 203 0.07
Regular 35.0 25.1
Non-regular 65.0 74.9
Excluding subjects for which there was missing
data
19
Results Odds ratios and 95CIs for smoking and
melanoma
Logistics regression analysis with matched
variables and Adjusted for history of sunburns
and sun exposure history
Smoking history Cases (n 103) Controls (n206) Odds Ratio (95 Confidence interval)

Never 59.2 55.8 1.00 (reference)
Former 27.2 33.5 0.75 (0.39-1.45)
Current 13.6 10.7 1.29 (0.47-3.58)
Smoking frequency Odds Ratio (95 Confidence interval)

None 59.2 55.8 1.00 (reference)
lt20 / day 14.6 18.0 0.68 (0.29-1.56)
20 /day 26.2 26.2 1.00 (0.48-2.11)
20
CONCLUSIONS

• After carefully controlling for age, sex, race,
  skin type, UV exposure history, and history of
  blistering sunburns we did not find evidence that
  cigarette smoking is associated with either an
  increased or decreased risk of cutaneous
  malignant melanoma.

21
CONCLUSIONS

• Limitations

• Strengths

• Recall bias with any survey
• Low number of subjects
• 80 power to detect an OR 2.0 or 0.5
• No information on number of overall nevi

• More precise adjustment of UV exposure history,
  skin type, history of sunburns
• Average age of 69.2 at 2007 follow-up for cases
  and controls allows for adequate time for
  observation

22
ACKNOWLEDGEMENTS

• Dr. Rhoda M. Alani
• Dr. Anthony J. Alberg
• Mr. Lee Wheless
• Ms. Sandy Clipp
• Ms. Judy Hoffman
• Doris Duke Charitable Foundation
• American Federation of Medical Research

23
THANK YOU!
Can I take any question/comments?
24
References

• Source www.jhsph.edu/comstockcenter
• Coups EJ, Manne SL, Heckman CJ. Multiple skin
  cancer risk behaviors in the U.S. population. Am
  J Prev Med. 200834(2)87-93. 10.1016/j.amepre.200
  7.09.032.
• Freedman DM, Sigurdson A, Doody MM, Rao RS, Linet
  MS. Risk of melanoma in relation to smoking,
  alcohol intake, and other factors in a large
  occupational cohort. Cancer Causes Control.
  200314(9)847-857.
• Osterlind A, Tucker MA, Stone BJ, Jensen OM. The
  danish case-control study of cutaneous malignant
  melanoma. IV. no association with nutritional
  factors, alcohol, smoking or hair dyes. Int J
  Cancer. 198842(6)825-828.

25
References

• Odenbro A, Gillgren P, Bellocco R, Boffetta P,
  Hakansson N, Adami J. The risk for cutaneous
  malignant melanoma, melanoma in situ and
  intraocular malignant melanoma in relation to
  tobacco use and body mass index. Br J Dermatol.
  2007156(1)99-105. 10.1111/j.1365-2133.2006.07537
  .x.
• Shors AR, Solomon C, McTiernan A, White E.
  Melanoma risk in relation to height, weight, and
  exercise (united states). Cancer Causes Control.
  200112(7)599-606.
• Grant WB. A meta-analysis of second cancers after
  a diagnosis of nonmelanoma skin cancer
  Additional evidence that solar ultraviolet-B
  irradiance reduces the risk of internal cancers.
  J Steroid Biochem Mol Biol. 2007103(3-5)668-674.
  10.1016/j.jsbmb.2006.12.030.

26
References

• Shaw HM, Milton GW. Smoking and the development
  of metastases from malignant melanoma. Int J
  Cancer. 198128(2)153-156.
• Shaw HM, Milton GW, McCarthy WH, Farago GA,
  Dilworth P. Effect of smoking on the recurrence
  of malignant melanoma. Med J Aust.
  19791(6)208-209.