Development of a score that predicts survival among patients with bone metastasis revealing solid tumor

Support Care Cancer. 2008 Sep;16(9):1089-93. doi: 10.1007/s00520-008-0455-8. Epub 2008 May 14.

Abstract

Background: Bone metastases revealing solid tumors require a multidisciplinary treatment that may include aggressive surgery. Nevertheless, decision making is very difficult because of lack of reliable guidance for life expectancy. Therefore, we had conducted a retrospective study to analyze prognostic factors in this vulnerable population.

Materials and methods: We reviewed the medical charts of 119 consecutive patients with solid tumors treated between August 1999 and December 2007. Survival was estimated using Kaplan-Meier method. Prognostic factors were assessed with Log-rank test and Cox model. There were 77 men and 42 women.

Results and discussion: The median age was 57 (range, 29-84). The most frequent primaries were lung (51 cases), unknown primary (42 cases), and breast (six cases). The median overall survival was 118 days (1-2,815). Six independent prognostic factors were identified by multivariate analysis: Karnosky index < 80% (HR = 1.92), albuminemia < 38 g/l (HR = 2.60), natremia < 135 mEq/l (HR = 2.78), platelet count > 500,000/mm(3) (HR = 2.71), cutaneous metastasis (HR = 4.6), and pleural metastasis (HR = 4.76). For example, patients with < or =2 of these prognostic factors experienced a median overall survival of 196 days (1-2,875) and their risk of death within the 90 days was 15/77. On the contrary, the patients with more than two poor prognostic factors experienced a median overall survival of about 80 days (1-834) and their risk of early death was 22/32. We had proposed an easily obtained at bedside score that needs further validation on independent cohort.

MeSH terms

  • Adult
  • Aged
  • Aged, 80 and over
  • Bone Neoplasms / mortality*
  • Bone Neoplasms / pathology*
  • Decision Making
  • Female
  • Humans
  • Kaplan-Meier Estimate
  • Karnofsky Performance Status
  • Life Expectancy
  • Male
  • Medical Records
  • Middle Aged
  • Predictive Value of Tests
  • Prognosis
  • Proportional Hazards Models
  • Retrospective Studies
  • Risk Factors