X hits on this document

25 views

0 shares

0 downloads

0 comments

6 / 11

Bec2/BCG Vaccination in SCLC Patients

Median Survival (months)

95% CI

P

18.0 13.9

15.4 to 21.8 11.4 to 16.3

.0084

12.5 18.7

10.7 to 15.4 15.6 to 22.1

.0046

9.7 20.3

8.9 to 12.1 17.1 to 23.0

.0001

17.3 10.9

15.6 to 20.8 8.7 to 15.9

.0025

16.6 15.8

13.6 to 21.8 13.8 to 17.6

.0243

20.7 14.5 13.6

16.0 to 26.5 12.5 to 16.5 9.5 to 29.11

.0624

14.6 17.7

12.4 to 17.0 14.4 to 24.2

.0580

Table 4. Univariate Analysis of Potential Prognostic Factors for Overall Survival

Variable

Response to induction Complete response Partial response Chest radiotherapy Sequential Concomitant PCI No Yes Lactate dehydrogenase Grade 0

  • Grade 0

Platelets

  • 221 109/L

  • 221 109/L

Continent North America Europe Australia/New Zealand Sex Male Female

Abbreviation: PCI, prophylactic cranial irradiation.

  • Global Wald test for United States versus rest of the world: P .0235.

versus sequential chest radiotherapy, PCI versus no PCI; and normal versus abnormal levels of lactate dehydroge- nase and platelets. Sex was borderline nonsignificant. Con- sidering all the factors included in the univariate analysis, a Cox multivariate proportional-hazards model was fitted and stratified for treatment by using a stratified step-down (backward) variable-selection procedure (at the 5% level). The factors listed in Table 5 are those that remained significant in the multivariate analysis model. Adjusting for these factors in a multivariate model, treatment effect on overall survival remained not statistically significant.

Hazard Ratio

95% CI

P .0066

1 0.72

0.57 to 0.91

.0051

1 0.72

0.57 to 0.91

.0001

1 0.52

0.41 to 0.65

.0002

1 1.71

1.30 to 2.26

.0019

Abbreviation: PCI, prophylactic cranial irradiation.

Table 5. Multivariate Analysis for Survival

Variable

Sex Male Female Chest radiotherapy Sequential Concomittant PCI No Yes Lactate dehydrogenase Grade 0 Grade 0 Platelets

  • 221 109/L

  • 221 109/L

1 1.44

1.14 to 1.81

of patients in the two groups of this analysis, more patients received PCI in the responder group (70% v 58%); after stratifying or adjusting the comparison of survival for the presence or absence of PCI, the difference became not sig- nificant (P .1479 and .1970, respectively). When consid- ering only patients who received at least four vaccinations, 124 nonresponders and 65 responders were identified; a survival difference (albeit nonsignificant) was also visible in this comparison (median survival, 22.3 v 14.1 months in responders v nonresponders, respectively; P .0755). After stratification or adjustment for PCI, the signifi- cance was reduced further in this comparison (P .1286 and .1800, respectively).

Humoral Response

Humoral response could not be assessed in 44 vacci- nated cases (17% of all patients enrolled in the vaccination arm), because samples were lost (50%), fewer than three samples were available (32%), no prevaccination sample was available (16%), or no data on humoral response were transferred to the EORTC (2%). Humoral response was negative in 142 cases and positive (responder) in 71 cases; therefore, the responders were one third of all patients for whom humoral response was assessable. The survival of responders was better than that of nonresponders, although this did not reach statistical significance (median survival, 19.2 v 13.9 months for responders v nonresponders; P .0851; see Fig 3). However, looking at the distribution

Fig 3. Kaplan-Meier survival curves of patients with and without humoral response. On the y-axis, the percentage of surviving is reported; on the x-axis, the time from randomization (months) is reported.

www.jco.org

6859

Downloaded from jco.ascopubs.org on February 15, 2015. For personal use only. No other uses without permission. Copyright © 2005 American Society of Clinical Oncology. All rights reserved.

Document info
Document views25
Page views25
Page last viewedSun Dec 04 12:14:03 UTC 2016
Pages11
Paragraphs513
Words8777

Comments