Advanced

Prognostic instrument for survival outcome in melanoma patients : based on data from the population-based Swedish Melanoma Register

Lyth, J; Mikiver, R; Nielsen, K LU ; Isaksson, K LU and Ingvar, C LU (2016) In European Journal of Cancer 59. p.171-178
Abstract

BACKGROUND: Several major analyses have identified a consistent set of independent risk factors for cutaneous malignant melanoma (CMM). A few prognostic models have been presented but some are based on a limited number of patients and others are based on selected groups of patients referred to major institutions. No nationwide population-based prognostic instrument for survival of CMM has been presented. The Swedish Melanoma Register (SMR) database covers 99% of CMM diagnosed in Sweden and includes today >50,000 cases.

OBJECTIVES: To create a prognostic instrument based on SMR data to give highly reliable risk profiles for patients diagnosed with localised CMM.

METHODS: Clinicopathological data were linked to the cause of... (More)

BACKGROUND: Several major analyses have identified a consistent set of independent risk factors for cutaneous malignant melanoma (CMM). A few prognostic models have been presented but some are based on a limited number of patients and others are based on selected groups of patients referred to major institutions. No nationwide population-based prognostic instrument for survival of CMM has been presented. The Swedish Melanoma Register (SMR) database covers 99% of CMM diagnosed in Sweden and includes today >50,000 cases.

OBJECTIVES: To create a prognostic instrument based on SMR data to give highly reliable risk profiles for patients diagnosed with localised CMM.

METHODS: Clinicopathological data were linked to the cause of death registry for calculation of CMM-specific survival. A generalised gamma method was used to derive 1, 5 and 10year probabilities of death for each combination of patient and tumour data: age, sex, tumour site, tumour thickness, tumour ulceration, Clark's level of invasion and when applicable also outcome of sentinel node biopsy (SNB).

RESULTS: Tumour thickness had the highest prognostic impact, explaining 77% of the model. Women had 30% lower risk of death because of CMM than men. Presence of ulceration nearly doubled the risk. If the patient had a positive SNB status the risk of death due to CMM increased three times versus a negative SNB status.

CONCLUSION: This unique population-based prognostic model for primary CMM shows better survival than the American Joint Commission on Cancer prognostic model widely used. The reason is probably that the referral bias is eliminated in a population-based cohort.

(Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to journal
publication status
published
subject
in
European Journal of Cancer
volume
59
pages
8 pages
publisher
Elsevier
external identifiers
  • scopus:84962629902
  • wos:000375138200019
ISSN
1879-0852
DOI
10.1016/j.ejca.2016.02.029
language
English
LU publication?
yes
id
9a334ba1-8e5e-4199-84d9-25991e4b4d51
date added to LUP
2016-04-28 12:02:16
date last changed
2017-01-01 08:24:03
@article{9a334ba1-8e5e-4199-84d9-25991e4b4d51,
  abstract     = {<p>BACKGROUND: Several major analyses have identified a consistent set of independent risk factors for cutaneous malignant melanoma (CMM). A few prognostic models have been presented but some are based on a limited number of patients and others are based on selected groups of patients referred to major institutions. No nationwide population-based prognostic instrument for survival of CMM has been presented. The Swedish Melanoma Register (SMR) database covers 99% of CMM diagnosed in Sweden and includes today &gt;50,000 cases.</p><p>OBJECTIVES: To create a prognostic instrument based on SMR data to give highly reliable risk profiles for patients diagnosed with localised CMM.</p><p>METHODS: Clinicopathological data were linked to the cause of death registry for calculation of CMM-specific survival. A generalised gamma method was used to derive 1, 5 and 10year probabilities of death for each combination of patient and tumour data: age, sex, tumour site, tumour thickness, tumour ulceration, Clark's level of invasion and when applicable also outcome of sentinel node biopsy (SNB).</p><p>RESULTS: Tumour thickness had the highest prognostic impact, explaining 77% of the model. Women had 30% lower risk of death because of CMM than men. Presence of ulceration nearly doubled the risk. If the patient had a positive SNB status the risk of death due to CMM increased three times versus a negative SNB status.</p><p>CONCLUSION: This unique population-based prognostic model for primary CMM shows better survival than the American Joint Commission on Cancer prognostic model widely used. The reason is probably that the referral bias is eliminated in a population-based cohort.</p>},
  author       = {Lyth, J and Mikiver, R and Nielsen, K and Isaksson, K and Ingvar, C},
  issn         = {1879-0852},
  language     = {eng},
  month        = {04},
  pages        = {171--178},
  publisher    = {Elsevier},
  series       = {European Journal of Cancer},
  title        = {Prognostic instrument for survival outcome in melanoma patients : based on data from the population-based Swedish Melanoma Register},
  url          = {http://dx.doi.org/10.1016/j.ejca.2016.02.029},
  volume       = {59},
  year         = {2016},
}