Ionizing radiation has been used to treat cancers for a century. However, r
adioresistance remains a major problem in the clinic. Recent advances in th
e understanding of the molecular events that occur following ionizing radia
tion leading to DNA damage and repair, apoptosis, and cell cycle arrests su
ggest new ways in which the radiation response might be manipulated. Sevent
y-eight cases of carcinoma of the cervix of the same stage (II A and B) wer
e analyzed retrospectively. All patients were treated with radiotherapy (RT
) with a dose varying from 35 Cy to 50 Gy with 200 cGy per fraction. Subseq
uent to the completion of radiotherapy, all patients underwent surgery 4-6
weeks later. On histological examination of the surgical specimens, 51% of
the cases (40) showed a complete response to therapy with no viable tumor c
ells. 49% of cases (38) had residual tumors ranging from a small focus to l
esions extending; through more than half the thickness of the cervical wall
. p53 (mutant), bcl-2, p21 and bax proteins were studied on the paraffin se
ctions of the biopsies (pretreatment) of those patients who failed to respo
nd to RT and compared to similar studies on biopsies of patients who had a
complete response to RT. In addition, the minichromosome maintenance (MCM)
2 proliferative marker was also done on all cases. Expression of all protei
ns was done using immunohistochemsitry. In the radioresistant cases, 15% (s
ix cases) showed positivity for bcl-2 and p21, respectively, and 34% (13 ca
ses) showed mutant p53. None of the radiosensitive tumors were positive for
the above proteins. 75% of the radiosensitive tumors (30 cases) were posit
ive for the bax antibody, whereas 81% of the radioresistant tumors (31 case
s) were negative for bax. The MCM2 proliferative marker was positive in >80
% of cells in 81.5% of radioresistant tumors (31 cases) as compared to <40%
of cells that were positive in 70% of radiosensitive tumors (28 cases). Th
e P-value for the biological markers was calculated using the chi-squared t
est, and was highly significant (P<0.01) for all the parameters tested. How
ever, there was no statistical significance by univariate analysis when the
dose of radiation was analyzed with respect to the markers and the histolo
gical response. There was also no correlation between the radiation respons
e and timing of surgery. The above data strongly suggest that bax, along wi
th proliferative markers, could play a role in determining which tumors are
likely to respond to radiation therapy. The presence of bcl-2, p21 and p53
could also be related to radioresistance of the tumors.