In this study, we demonstrated that the HRD-positive group had a lower recurrence rate than the HRD-negative group, although the correlation between HRD score and recurrence was not statistically significant. We also found a strong correlation between HRD score and Hb level. Interestingly, multifactorial COX regression analyses revealed that Hb levels and NACT demonstrated significant independent influences on the rate of 1-year recurrence in OC patients. In addition, recurrence curves showed that NACT was associated with the 1-year recurrence rate in OC. However, our study did not find a significant association between HRD scores and platinum-resistant recurrence. Moreover, there was no statistical evidence that the HRD score was able to predict the 1-year recurrence rate in patients undergoing non-maintenance therapy.
Multiple clinical studies have shown that patients with OC and a high HRD score had a better prognosis than those with a lower HRD score [18–21]. Similar results were observed in our study. The reason for this result may be due to the defective HR repair pathways in HRD-positive tumors. Cells with homologous recombination defects are more susceptible to intrachain and interchain cross-linking effects of platinum agents, resulting in increased platinum sensitivity in HRD-positive tumors [25, 26]. However, when assessing the deaths at follow-up, there were more deaths in the HRD-positive group (7.32%) than in the HRD-negative group (3.57%), which seems to contradict the above findings. This is due to the differences in follow-up time between the two groups, or due to unavoidable individual differences. Further studies may obtain more valuable statistical differences between HRD score and recurrence, by extending the follow-up time and increasing the sample size.
Surprisingly, a low Hb level (Hb < 107g/L) before treatment was found to be associated with a higher 1-year risk of relapse, similar to what has been observed in other studies [27–29]. The oxygenation status of the tumor is related to the local blood supply to the tumor and the systemic oxygen carrier Hb. Thus, lower Hb can lead to a hypoxic state in tumor cells [30]. Tumor cell hypoxia first activates hypoxia-inducible factor 1, which binds the hypoxia response element (HRE). HRE activation can induce the transcription of target genes necessary for angiogenesis, glucose metabolism, cell proliferation/survival, and invasion/metastasis [31]. Therefore, hypoxia is considered a driver of tumor progression and a negative prognostic factor. The physiological selection pressure provided by hypoxia creates a more malignant and aggressive tumor cell phenotype that is resistant to conventional chemotherapy and radiotherapy, thus contributing to poor patient prognosis [31–33]. In addition, we found that the HRD score was closely related to Hblevels, which may be one of the reasons for the differences in rates of recurrence in the HRD-positive and HRD-negative groups.
Our study demonstrated a significantly higher risk of 1-year recurrence in patients undergoing NACT compared to those without NACT. It is unknown whether NACT affects the prognosis in OC. Several studies have found that the prognosis of patients with OC initially treated with NACT was not inferior to those treated with primary debulking surgery (PDS) [34, 35]. Conversely, other studies have found no survival benefit or, in fact, an even worse prognosis in patients treated with NACT compared to PDS [36, 37]. According to the primary treatment guidelines for OC developed by the Society of Gynecologic Oncology and the American Society of Clinical Oncologists, women with a high perioperative risk or a low likelihood of cytoreduction to residual lesions < 1 cm (ideally without significant lesions) should receive NACT [38]. The choice of NACT treatment at our hospital is made based on the patients Suidan score, a predictive score that better predicts sarcoid residual disease when responding to tumor load [39]. Patients with low Suidan scores underwent direct initial tumor cytoreductive surgery, while patients with high scores were treated with NACT first. Therefore, our results may indirectly indicate that patients with OC who also present with a high Suidan scores may have a poorer prognosis.
The test samples used for the HRD genetic test were formaldehyde-fixed paraffin-embedded tissues. The test must be done within 2 years. Given that the quality of the specimens decreased year by year, the HRD scores became progressively inaccurate in the longer the life of the samples. Moreover, since the application of PARP inhibitors in the Chinese population has become quite widespread, the HRD test is more valuable and better reflects the HRD score application in the OC population without PARP inhibitors. Therefore, this study is non-reproducible to some extent. As with all retrospective studies, our study was not completely immune to selection bias. We attempted to reduce interference by excluding patients with active infection, chronic inflammatory disease, or from whom follow-up data was not available. Our shortcomings also include the fact that data from only one company AmoyDx rather than the two FDA-approved companies, Myriad myChoice® CDx and FoundationFocus™ CDx BRCA LOH, were utilized. In addition, the number of patients enrolled in this study was low. Thus, in the future, we hope to expand the database by collaborating with other hospitals.