Personal tools
You are here: Home Research site Dissertations 1997 Henk Struikmans

1997 Henk Struikmans

Prognosis in Brain and Head & Neck Tumours

Clinical Relevance of Cell Proliferation Markers and DNA-index

In this thesis we report about some aspects of the clinical relevance of cell proliferation markers and DNA-index for patients with either a head and neck tumour or a brain tumour. The objective of this study was explained to 500 patients. Ultimately 278 agreed to participate. Cell proliferation markers and DNA-index were obtained by flow cytometry.

In Chapter 2 we analysed the interrelationships between DNA-ploidy, cell proliferation markers, T-stage, and N-stage in 64 single site head and neck tumours, i.e. the larynx. We found that locally advanced (T3-4) laryngeal tumours were characterised by a significantly higher percentage of DNA-aneuploidy. This finding confirms the reported poor prognosis in DNA-aneuploid tumours. A significantly higher proliferative potential, i.e. a higher mean LI and a lower mean Tpot value was found in early stages (T1-2), as opposed to that in locally advanced (T3-4) tumours. This difference in proliferative activity was not related to N-stage. We stated that the tumour stage related decrease of the proliferation rate, might relate to an increased level of hypoxia in the larger tumours. These findings may in part explain the reported poor prognostic relevance of cell proliferation markers in head and neck tumours and imply that the predictive value of (LI and) Tpot should be evaluated only after stratification for T-stage (T1-2 versus T3-4). A higher mean LI value was found in DNA-aneuploid tumours. This emphasises the need for distinction of DNA-diploid tumour cells from DNA-diploid normal cells. We found that Ts tended to be longer in DNA-aneuploid tumours (p=0.07).

In Chapter 3, we reported on the prognostic significance of proliferation markers and DNA-index in 103 patients with head and neck squamous cell carcinomas, treated with curative intent. Treatment regimen consisted of radiotherapy alone, surgery and radiotherapy, or surgery alone. A conventionally fractionated scheme was administered in the case of radiotherapy. We demonstrated that cell proliferation markers (including Tpot) were unrelated to loco-regional recurrence probability. However, we noted that high values of the ratio SPF to LI, were associated with an increased rate of recurrent disease. It was postulated that this ratio might be indicative for the size of the hypoxic fraction. This may, at least in part, explain the absence of prognostic relevance of the cell proliferation markers in laryngeal tumours and imply that the predictive value of (LI and) Tpot should be evaluated only after stratification for T-stage (T1-2 versus T3-4). The reported prognostic significance for DNA-index was confirmed and appeared to be related to the poor prognosis associated with DNA-aneuploidy and specifically DNA-tetraploidy. Furthermore, we found that the probability of loco-regional recurrent disease was significantly decreased in primarily irradiated DNA-diploid tumours. Cell proliferation markers are generally estimated from a single biopsy. Both regional heterogeneity and intraobserver variability may interfere with the reliability of the parameter determination.

In Chapter 4, we determined the intratumoural, intertumoural and intraobserver variability of proliferation markers and DNA-index in head and neck tumours. We found that, i) with regard to regional heterogeneity, DNA-ploidy status and DL showed less variation than did LI, and SPF, ii) the ratio of the intratumoural variability to the intertumoural variability was lowest for Dliii) the regional heterogeneity for Ts and Tpot was moderate, iv) the intraobserver variability was low for DNA-ploidy status, Dl, and LI, and moderate for RM. We thus concluded that, in head and neck tumours, the conditions for Dl as a prognosticator are better than those for SPF, LI, G1PF, Ts, and Tpot.

In Chapter 5, brain tumours were characterised with respect to differences in DNA-index and cell proliferative potential. We noted that malignant brain tumours, as opposed to benign brain tumours, were characterised by malignant FCM features, such as i) a significantly higher proportion of DNA-aneuploid tumours, ii) a significantly higher mean Dl (for DI>1), and iii) a significantly higher proliferative activity (SPF, LI, and Tpot), as noted for diploid tumours only. Ts of benign and malignant tumours appeared to be relatively short, i.e. 4.5 hours, and not to differ between benign and malignant brain tumours. From this, it follows that the differences in Tpot were due mainly to differences in LI. Low grade gliomas were, according to the literature, found to be characterised by low (long-term) survival probability and malignant FCM features. Therefore, low grade gliomas should be considered as a malignant disease.

In Chapter 6, we analysed the prognostic relevance of clinically, histologically, and flow cytometrically derived parameters in 49 glioma patients after long-term follow-up. We noted that LI and Tpot could be adequately obtained in 71% and 63% of the tumours, respectively. The success rate of LI and Tpot determination was higher in head and neck tumours. In glioma however, we found that LI and, hence, Tpot could not be adequately obtained in a number of cases, due to very low labelling indices. Seizures, duration of symptoms, and age appeared to be associated with cell proliferation rate and were found to be associated with prognosis. SPF, LI, and Tpot were also associated with prognosis. However, by regression analysis, no additional prognostic relevance, compared with that of the presence or absence of cells in the mitoses, could be demonstrated. After multivariate analysis of established prognosticators such as necrosis, endothelial proliferation, and mitoses, we found the presence or absence of cells in mitoses to be the strongest single prognosticator. Prognostic significance of LI was noted in the subgroup of low grade gliomas.

In Chapter 7, we evaluated in 27 glioma patients, the prognostic significance of tumour cell proliferation markers obtained by flow cytometry (FCM), i.e. SPF and Llfcm, by immunohistochemistry (IHC), i.e. MIB-1-LI and Llihc and by histologic examination i.e. of the presence or absence of cells in mitoses. Long term follow-up was achieved. After multivariate analysis, we found the presence of mitotic cells, to be of more prognostic significance than Llfcm, or MIB-1-LI. Of the tested cell proliferation markers, Llfcm appeared to be of more prognostic importance in gliomas than MIB-1-LI, SPF, and Llihc. From this, it was concluded that Llfcm may be of prognostic significance, specifically inlow grade gliomas, since cells in mitoses are by definition not found in low grade gliomas.

To contact the author:
MCH Westeinde
t.a.v. Prof.Dr. H. Struikmans, hoofd afdeling radiotherapie
Postbus 432
tel +31 70 3302013 (secretariaat)
fax +31 70 3809459