Ovarian cancer is the most lethal gynecological neoplasm in the world, due to its late diagnosis and non-specific symptoms. Within the study, 198 patients with ovarian cancer were evaluated, of which 153 (77.2%) presented advanced stage disease at the time of diagnosis (including epithelial and non-epithelial tumors), which is in line with what is reported in the international literature, where it has been indicated that more than 65% of patients are diagnosed in advanced stages (1). The average age of presentation of the 53 patients included was 57.8 years (SD ± 10.3 years), lower than that reported globally, being 62 years (1, 42). This early presentation has been reported in other neoplastic diseases within the Mexican population, highlighting colorectal and prostate cancer.
The main documented comorbidities were overweight (45.2%) and systemic arterial hypertension (43%). The main symptoms were abdominal and nonspecific, in accordance with the literature (43), highlighting the increase in abdominal girth in 40 patients (75.5%) and diffuse abdominal pain in 35 (66.6%).
The only variable that showed statistically significant difference between both groups was the histological subtype, where patients with endometrioid and clear cells subtypes did not have favorable tomographic response with neoadjuvant treatment. Clear cell and mucinous subtypes in advanced disease are associated with a very poor prognosis and resistance to standard treatment (44, 45), however, this observation is limited in our study due to the small sample size (only one case with clear cell histology) and not including mucinous tumors because of their low production of HE4 biomarker.
Adverse events from chemotherapy occurred more frequently at the hematological and gastrointestinal levels, in accordance with the toxicity profile of the platinum-taxane combination reported in the literature. Some other frequent manifestations such as dysgeusia and headache were not reported during medical visits, in up to 11.8% and 7.1% of cases, respectively.
Regarding the biomarker HE4, it has demonstrated protease activity and participation in cell signalling, acting on processes of adhesion, migration, and promotion of tumour growth (46, 47). In addition to its standard use, approved for the differential diagnosis of malignant adnexal tumors (ROMA algorithm), there are assays that explore the value of HE4 as a prognostic marker (48), a predictor of optimal cytoreduction (49), and a potential tool for early diagnosis of recurrence, even better than CA125 (50, 51).
Most of the reports published to date are retrospective. Studies that relate the role of HE4 to the response to chemotherapy often address the adjuvant setting, following primary cytoreduction. In addition, the authors have used various trials (ARCHITECT by Abott®, ELECSYS by Roche®, or EIA by Fujirebio®) and various statistical methods and approaches for data analysis (percentage decrease, low area to Curve, negativization, etc.), which increase heterogeneity and limit the comparison of results.
The pioneering study, which serves to contrast the results of the work carried out in this investigation, was published by Vallius and cols. (52), where 25 patients with advanced epithelial ovarian cancer, treated with neoadjuvant chemotherapy, were evaluated, quantifying the CA125 and HE4 biomarkers in a basal manner and on a later occasion, after completion of adjuvant chemotherapy, prior to interval cytoreduction. The percentage changes of the biomarkers were compared according to the tomographic response and the surgical outcome, concluding that neither CA125 nor HE4 changes were correlated with the tomographic response, observing a clear reduction in serum levels of both biomarkers in all groups, regardless of the radiological response. In fact, patients with disease progression had an average decrease in CA125 levels of around 83%. This was surprising, since in the pilot study with 11 patients previously conducted, a relationship was observed between changes in serum concentration of biomarkers and the tomographic response.
Despite this finding, the report by Vallius and cols. described a favorable relationship between HE4 decline (> 80%) and prognosis in terms of overall survival, with a median of 3.38 versus 1.60 years (p = 0.01). This put into question the discordance between survival results and radiological therapeutic response. The authors argued that tomographic guidelines may be a limitation in the optimal assessment of the response to chemotherapy, suggesting to study, the possibility of using functional imaging as PET-CT to optimize the assessment. In addition to the small number of patients, a limitation of the study by Vallius and cols. was to have only two measurements for each patient.
Chudecka and cols. (53), published a study involving 90 patients with ovarian cancer, 42 of whom were treated with neoadjuvant chemotherapy, with HE4 and CA125 measurements taken at diagnosis, after chemotherapy, and before interval cytoreductive surgery, although in the final analysis, the authors only considered the third cycle of chemotherapy.
Preoperative HE4 levels were a predictor of platinum sensitivity (p-value = 0.035) and progression-free survival (p-value = 0.0492) when normalized or reduced by 50%, but, unlike the reports of Vallius and cols. (52), in multivariate analysis, normalized HE4 levels after chemotherapy (HR = 0.08, p-value = 0.0003) or with 50% reduction before interval debulking (HR = 0.39, p-value = 0.0496), correlated with improvement in 2 years overall survival.
The present study is original in its design and tries to optimize the limitations of the reported studies previously. It only includes patients with advanced epithelial ovarian cancer treated with neoadjuvant chemotherapy, excluding the group of patients treated with primary cytoreductive therapy and those who have undergone incomplete surgery (oophorectomy, lumpectomy) prior to the start of systemic treatment, due to the alteration that this procedure conditions in the serum level of the biomarkers. In addition, baseline quantifications have been performed and in each cycle of systemic treatment, including all measurements within the final statistical analysis.
Unlike what was reported by Vallius and cols. (52), in this study a statistically significant association was observed between HE4 changes and the tomographic response in patients with epithelial ovarian cancer treated with NACT. These observations highlight the need to increase the sample size (11 and 25 patients, respectively, in the pilot and the trial by Vallius and cols.; 42 patients in the study by Chudecka and cols., standardize the number and timing of the quantifications, as well as the inclusion criteria, homogenizing the type of chemotherapy to be studied (neoadjuvant versus adjuvant) and analyzing both biomarkers to determine the collinearity of the results.
The main findings of our study are that HE4 levels, especially during the second and third cycle, are independently associated with the tomographic response in patients with advanced epithelial ovarian cancer treated with neoadjuvant chemotherapy. Also, a greater reduction between the basal HE4 levels and those of the first cycle of chemotherapy was independently associated with the tomographic response, which is relevant to predict the group of patients who will not respond to treatment.
Despite the main limitation is the number of patients recruited during one year at a single cancer center, this public institution is considered one of the main centers of care for cancer patients nationwide. One the other hand, the main strengths are the careful inclusion of participants and in final analyses that included all quantifications (baseline and in each cycle).