Multi-drug resistance contributes to ovarian cancer relapse following the initial chemotherapy response [
29,
30]. Despite of increased doses of first-line chemotherapy drugs (such as platinum and paclitaxel) or the application of second-line chemotherapy drugs (such as topotecan, gemcitabine, liposomal doxorubicin and docetaxel), patient prognosis is poor and the average duration of survival is less than 1 year [
31,
32]. Aberrant
MDR1 expression has been observed in most chemoresistant tumors. A better understanding of the molecular mechanism linking
MDR1 and
TP53, the most common mutated gene in ovarian cancer, may facilitate accurate prediction of the efficacy of standard chemotherapy and aid in the formulation of a rational, individualized protocol for preventive interventions in selected patients to reduce relapse and improve 5-year survival.
In the current study, we validated the use of immunohistochemical analysis to identify the presence of
TP53 mutation in ovarian cancer, in accordance with previously published data from Yemelyanova [
26].
TP53 mutation closely correlated with immunohistochemical staining patterns for P53 (strong and diffuse expression and complete lack of expression). P53 overexpression, as revealed by immunostaining, was associated with missense or frameshift mutations in
TP53, whereas nonsense mutations resulted in protein truncation and a complete lack of immunostaining. Importantly, in addition to non-mutant
TP53 tissue, some ovarian cancer samples contained a low-abundance frameshift
TP53 mutation that resulted in weak (<10%) positive or positive (10%-50%) signals for P53 staining. Such samples were generally considered as wild-type
TP53. There are two possible explanations for this phenomenon. First, direct sequencing may not be the most accurate gene testing method (massive parallel sequencing increases the accuracy of somatic mutation detection), and thus our detection of the low-abundance mutant tissue was a coincidence. Second, the presence of a low-abundance mutation in a tiny region of tumor tissue still results in adequate P53 protein function without impairment. In previous studies, the overall sensitivity for
TP53 mutation detection based on P53 immunostaining in high-grade serous ovarian cancer was 99%, and for epithelial ovarian cancer, the accuracy of
TP53 mutation diagnosis based on immunohistochemistry was as high as 95% [
26,
27].
A relationship between the presence of
TP53 mutation and the 5-year survival of ovarian cancer patients has been reported in many studies. In a recent analysis, the hazard ratio (HR) of P53 status on survival was only 1.47 (95% CI: 1.33–1.61), and HR differed with various conditions (such as late stage and serous cancer cases). Even in late stage cases (stages III, IV), the HR for
TP53 mutation decreased to 0.91 (95% CI: 0.59–1.39), indicating a possible protective effect, and serum P53 autoantibodies [HR: 1.09 (95% CI: 0.55-2.16)] were not associated with OS [
33,
34]. This result is likely attributable to the constrained P53 function in high-grade tumors and in late-stage serous ovarian cancer, a hypothesis that need to be tested in future studies. In the current study,
TP53 mutation was not an independent factor for the 5-year survival compared with other factors (stage, grade and histotype) according to our multivariate analysis. In contrast, wild-type
TP53 was a crucial factor that correlated with disease progression by preventing chemoresistance in ovarian cancer cells. Despite the proposal that
TP53 mutations impact ovarian cancer recurrence in certain studies, in current study, the in-depth delineation of
TP53 mutation and
MDR1 copy number analysis permitted us to identify a new mechanism by which recurrence of ovarian cancer is accelerated through the evolution of chemoresistance due to
TP53 mutation. Once
TP53 is mutated, cancer cells with DNA damage are not able to activate the apoptosis program. Instead, cells remain in the G1 or G2 phase in anticipation of DNA repair [
35,
36], enabling cancer cells to augment their drug resistance capacity. Notably, cancer cell DNA content and nucleus size were markedly different in the
TP53 mutant and wild-type groups, reflecting remarkable increases in the amounts of cells exhibiting polyploid and aneuploid DNA in the
TP53 mutant group, which resulted in exacerbated genomic instability and a high degree of malignancy. Furthermore, we detected larger numbers of cancer cells with >6
MDR1 copy numbers and >4 copies of chromosome 7 in the
TP53 mutant group. Regardless of the chemotherapeutic choice,
MDR1 copy number in the mutant group was higher than the wild-type group. Based on these results, we predicted that cancer cells with high
MDR1 copy numbers that were confronted with a selective environmental pressure (for example, chemotherapy after surgery) were more likely to survive and seed recurrence. To investigate this hypothesis, we compared
MDR1 copy numbers and MDR1 expressions in original and relapse cancer tissues. The results were as expected: MDR1 protein levels were noticeably increased in relapse tissues, particularly in the
TP53 mutant patients. Upon selective pressure exerted by chemotherapy, the number of
MDR1 copies was remarkably upregulated in the
TP53 mutant group, conceivably caused by the accumulation of chromosome 7. Conversely,
MDR1 copy numbers in relapse patients in the
TP53 wild-type group were not altered, and less chemoresistance was associated with second-round chemotherapy. We attributed recurrence in the wild-type group to other factors, such as late stage or large residual lesions. Our findings confirmed that
TP53 mutation displayed an independent effect on ovarian cancer recurrence after complete remission due to effective chemotherapy. Thus, the detection of
TP53 mutation is equally as important as clinical prognostic indicators (FIGO stage, histotype, grade) for predicting ovarian cancer recurrence.