A retrospective study evaluating the effect of trastuzumab addition to carboplatin/paclitaxel on overall survival in patients with advanced-stage HER2/neu-overexpressing uterine serous carcinoma or carcinosarcoma

Abstract

Background

Uterine serous carcinoma and carcinosarcoma are aggressive forms of endometrial cancer with poor survival outcomes. Trastuzumab, a human epidermal growth factor receptor-2 (HER2)-directed monoclonal antibody, has demonstrated tumoricidal efficacy. However, clinical data regarding its efficacy in uterine serous carcinoma are limited, and there are no clinical data available for uterine carcinosarcoma. Therefore, this study aimed to ascertain the efficacy and safety of adding trastuzumab to carboplatin and paclitaxel as a frontline treatment for advanced-stage HER2-overexpressing uterine serous carcinoma and carcinosarcoma.

Methods

This retrospective study used deidentified data from electronic health records from the TriNetX Research Network. Participants were identified using International Classification of Diseases codes, and HER2 positivity was confirmed through immunohistochemistry or fluorescence in situ hybridisation. Propensity score matching was employed to reduce confounders, and survival outcomes and adverse events were assessed.

Results

Following propensity score matching, 280 patients with advanced HER2-positive uterine serous carcinoma or carcinosarcoma were analysed. The group of patients treated with carboplatin/paclitaxel + trastuzumab (CP + T) showed a significantly prolonged median overall survival compared to those treated exclusively with CP (41 months versus 25.2 months, hazard ratio [HR] = 0.51, p = 0.002) in both advanced-stage uterine carcinosarcoma and serous carcinoma. Specifically, patients with uterine carcinosarcoma experienced a prolonged survival benefit (HR = 0.39, p < 0.0001) when trastuzumab was added to their chemotherapy regimen, which surpassed the survival benefit observed in patients with uterine serous carcinoma (HR = 0.56, p = 0.04). However, patients who received trastuzumab experienced increased rates of hypertension, diarrhoea, and left ventricular systolic dysfunction.

Conclusions

The addition of trastuzumab to frontline chemotherapy is effective in treating HER2-overexpressing uterine serous carcinoma and carcinosarcoma, particularly uterine carcinosarcoma. However, careful monitoring of adverse cardiac events is needed.

Background

Endometrial malignancy is the most prevalent gynaecological cancer in developed countries, with its incidence on the rise. By 2040, it is projected to surpass colorectal cancer as the third most common cancer among women [1]. Unlike other solid tumours, its incidence and mortality have continued to increase [2, 3]. Among its subtypes, uterine serous carcinoma is a particularly aggressive, high-grade variant, accounting for 40% of endometrial cancer-related deaths despite representing only 10% of cases. With a 5-year survival rate of 45% (compared to 91% for endometrioid adenocarcinoma) [4, 5], uterine serous carcinoma frequently presents with extrauterine spread (60–70%) [6, 7] and has a poor survival rate (18–30%) in advanced stages [7, 8]. Current chemotherapy agents, such as carboplatin and paclitaxel, exhibit low response rates (20–60%) [9,10,11] in these cases, highlighting an urgent need for new treatments.

Uterine carcinosarcoma, formerly known as malignant mixed Müllerian tumour, is a rare but lethal gynaecologic cancer. Although it accounts for less than 5% of all endometrial cancers, it carries a 1.5–1.6-fold increased mortality risk compared to uterine serous carcinoma [12, 13], with a 5-year survival rate below 30% [14, 15]. Conventional chemotherapy demonstrates limited efficacy, underscoring the necessity for targeted therapies for advanced and recurrent disease [16, 17]. Despite distinct molecular characteristics and differences in metastatic potential, over 70% of the carcinoma component in carcinosarcoma consists of serous carcinoma or high-grade endometrioid carcinoma. [13, 14] Notably, metastatic or recurrent tumours are often dominated by this component, highlighting the role of serous carcinoma in the aggressive behaviour of the tumour [16].

Uterine serous carcinoma and carcinosarcoma are molecularly distinct from endometrioid carcinoma. While uterine serous carcinomas are characterised by alterations in TP53 and dysregulation of human epidermal growth factor receptor-2 (HER2/neu), endometrioid cancers are characterised by aberrant PTEN [18]. The prevalence of HER2 expression is approximately 30–35% in uterine serous carcinoma compared to 6–56% in uterine carcinosarcoma, based on the interpretation and scoring criteria [18]. Similar to its role in many cancers, HER2 is a recognised marker of poor prognosis in uterine malignancies and is a potential target for novel therapeutic interventions [13, 19].

Trastuzumab, the first HER2-directed monoclonal antibody approved by the Food and Drug Administration (FDA) in 1998 for the treatment of HER2-overexpression metastatic breast cancer, has demonstrated tumoricidal efficacy through antibody-dependent cellular cytotoxicity (ADCC) and inhibition of HER2-mediated signalling pathways, impeding cell proliferation and angiogenesis [20]. Additionally, the FDA approval for trastuzumab extends to the treatment of HER2-overexpression early-stage breast cancer as well as HER2-overexpression gastric and gastroesophageal junction adenocarcinoma.

In 2018, Fader et al. conducted the only randomised controlled multicentre phase II trial to compare carboplatin + paclitaxel to carboplatin + paclitaxel + trastuzumab to evaluate the clinical effectiveness and safety of trastuzumab for advanced-stage (stage III–IV) or recurrent HER2-overexpressing uterine serous carcinoma [21]. Subsequently, in 2020, an updated survival analysis confirmed the initial progression-free survival (PFS) findings and reported overall survival (OS) differences, which favoured treatment with trastuzumab, particularly for stage III–IV disease [22]. According to the National Comprehensive Cancer Network (NCCN) guidelines, trastuzumab is recommended for the treatment of advanced-stage or recurrent uterine carcinosarcoma with HER2 overexpression, in addition to stage III/IV or recurrent uterine serous carcinoma with HER2 overexpression [23]. However, to date, there have been no clinical trials or retrospective studies on the utility of trastuzumab for treating uterine carcinosarcoma.

Therefore, this retrospective study aimed to examine the efficacy and safety of combining trastuzumab with carboplatin and paclitaxel as a frontline treatment for stage III/IV HER2-overexpressing uterine serous carcinoma and carcinosarcoma, using data from the TriNetX database. HER2 overexpression was identified based on an immunohistochemistry (IHC) staining score of 3 + , or 2 + confirmed using fluorescence in situ hybridisation (FISH).

Methods

Study design and data source

This retrospective, multi-institutional study used deidentified data from the TriNetX Research Network (Cambridge, MA, USA), which provides electronic medical records for approximately 111 million patients across 80 healthcare organisations in the United States and Europe. The TriNetX platform complies with the Health Insurance Portability and Accountability Act (HIPAA) and holds a waiver from the Western Institutional Review Board due to its use of deidentified data.

Data collection and study population

Data for this study were extracted on December 12, 2023. We included patients aged 18–90 years diagnosed with HER2-overexpressing advanced uterine serous carcinoma or carcinosarcoma, identified using specific International Classification of Diseases (ICD) and gene codes (Additional file 1: Table S1). Exclusion criteria included patients diagnosed with breast or ovarian cancer within five years of their endometrial cancer diagnosis to avoid misclassification related to HER2-overexpressed breast cancer or ovarian cancer.

Diagnosis and classification

HER2-overexpression was confirmed through IHC staining scores (3 + , or 2 + with FISH confirmation). Metastasis was identified using ICD codes for lymphatic and distant site metastasis, and disease staging was based on the International Federation of Gynaecology and Obstetrics (FIGO, 2009) criteria for stages III and IV.This classification was confirmed using the TriNetX Curated and North American Association of Central Cancer Registries (NAACCR) and gene codes. Furthermore, corresponding ICD codes were subsequently used to identify metastasis to lymphatic systems and distant anatomical sites. Chemotherapy and targeted therapy regimens were identified using RxNorm and Healthcare Common Procedure Coding System (HCPCS) codes. Surgical procedures and irradiation history were identified using Current Procedural Terminology (CPT) codes. All codes used in this study are listed in Additional file 1: Table S1.

Treatment and outcomes

Patients underwent primary or interval staging/debulking surgery followed by carboplatin and paclitaxel, with optional trastuzumab. The index event for the carboplatin and paclitaxel (CP) group was defined as the initiation date of carboplatin and paclitaxel administration, which served as the primary first-line treatment. In contrast, for the trastuzumab (T) group, the index event was defined as the date when trastuzumab was initially administered within a 3-month timeframe following the commencement of CP treatment as the initial therapeutic approach. The primary outcome was the median overall survival (OS) measured from the index event date to the date of death. The secondary outcome was the incidence of adverse events (haematological, hepatic, and systemic) in both groups, excluding patients with pre-existing events within six months before the index event.

Statistical analysis

Baseline characteristics, including comorbidities (assessed via the Charlson Comorbidity Index(CCI), Additional file 1: Table S2), were compared using chi-square and t-tests. We performed 1:1 propensity score matching (PSM) to reduce potential confounders (age, race, irradiation, and CCI score) (Additional file 2: Fig. S1). Matched group comparisons were conducted using the McNemar and paired t-tests. Survival outcomes were assessed using stratified log-rank tests, with hazard ratios calculated via conditional Cox regression. Statistical significance was set at p < 0.05.

Results

We identified 603 patients aged 18–90 years who were newly diagnosed with advanced HER2-overexpression uterine serous carcinoma or carcinosarcoma across 23 healthcare organisations between January 2012 and December 2023. According to the inclusion and exclusion criteria, 267(57%) patients received frontline CP, and 198(43%) patients received CP + T. Following 1:1 PSM, the two cohorts were well-balanced and included 140 patients each (Additional file 2: Fig. S1). The details of the CCI scores are provided in Additional file 1: Table S2. Compared to those of the CP group, the CP + T group had a higher proportion of African American patients (19% in the CP + T group vs. 14% in the CP group, p = 0.04) and a greater percentage of patients diagnosed with carcinosarcoma prior to the index date (63% in the CP + T group vs. 58% in the CP group, p = 0.03). In contrast, the CP group had higher rates of radiotherapy (36% in the CP group vs. 27% in the CP + T group, p < 0.001), pelvic lymph node metastasis (66% in the CP group vs. 61% in the CP + T group, p = 0.02), and colorectal metastasis (17% in the CP group vs. 12% in the CP + T group, p = 0.04) than the CP + T group. A comparison of the baseline characteristics for the CP and CP + T groups is presented in Table 1.

Table 1 Baseline characteristics of patients in the CP and CP + T groups before and after PSM

In the study population, 31 and 93 deaths were noted in the CP + T and CP groups, respectively. At the time of analysis, participants in the CP arm received a median of 6 cycles of carboplatin and paclitaxel (range: 1–10 cycles), while those in the trastuzumab arm received a median of 6 cycles of carboplatin and paclitaxel (range: 2–9 cycles) along with a median of 14 cycles of trastuzumab (range: 4–73 cycles). The median OS times were significantly different between the CP (25.2 months) and CP + T groups (41 months) (hazard ratio [HR]: 0.51, 95% confidence interval [CI] 0.345–0.784, p = 0.002) (Fig. 1). In the subgroup of uterine carcinosarcoma, 63 patients in the CP group and 21 patients in the CP + T group died due to the disease. This resulted in a notable contrast in the median OS time between the CP (16.8 months) and the CP + T groups (41.7 months) for uterine carcinosarcoma, with an HR of 0.39 (95% CI 0.221–0.560, p < 0.0001) (Fig. 2). Additionally, within the uterine serous carcinoma subgroup, 29 and 13 patients in the CP and CP + T groups respectively succumbed to the disease. Within the uterine serous carcinoma subgroup, the median OS time significantly differed between the CP (27.1 months) and CP + T (38.7 months) groups, with an HR of 0.56 (95% CI 0.294–0.869, p = 0.04) (Fig. 3).

Fig. 1

Overall survival (OS) of all HER2-overexpression uterine cancer patients who were treated with CP or CP + T. CP, carboplatin plus paclitaxel; T: trastuzumab

Fig. 2

Overall survival (OS) of patients with HER2-overexpression uterine carcinosarcoma who were treated with CP or CP + T. CP, carboplatin plus paclitaxel; T: trastuzumab

Fig. 3

Overall survival (OS) of patients with HER2-overexpression uterine serous carcinoma who were treated with CP or CP + T. CP, carboplatin plus paclitaxel; T: trastuzumab

Regarding adverse events, the CP + T group had a higher incidence of hypertension (8% vs. 25%, p = 0.002), diarrhoea (3% vs. 9%, p = 0.007), and left ventricular systolic dysfunction (2% vs. 9%, p < 0.001) than those of the CP group (Table 2). However, the rates of haematological toxicities, electrolyte imbalances, fatigue, peripheral sensory neuropathy, gastrointestinal symptoms, and renal dysfunction were similar between the two groups.

Table 2 Adverse events in the CP and CP + T groups after PSM

Discussion

Our study represents a ground-breaking investigation of real-world data on the addition of trastuzumab to frontline chemotherapy for HER2-overexpressing uterine carcinosarcoma and uterine serous carcinoma, offering promising insights into its potential benefits. Although adding trastuzumab significantly improved OS, it is crucial to emphasise the importance of vigilant monitoring of cardiac health and the early recognition of cardiac-related symptoms in patients receiving trastuzumab.

Based on the multicentre, randomised phase II trial by Fader et al. [21, 22], in which HER2 overexpression was identified based on an IHC score of 3 + , or 2 + and ERBB2 amplification was confirmed by FISH testing, the NCCN has issued guidelines recommending trastuzumab for HER2/neu-overexpressing advanced or recurrent uterine serous carcinoma and carcinosarcoma. These promising phase II findings demonstrated that patients in the trastuzumab cohort, particularly those with stage III–IV disease, had a longer median survival time (not reached) than those in the control group (24.4 months) (HR 0.49, 90% CI, 0.25–0.97, p = 0.041). Despite these findings, there have been no further clinical trials or investigations of real-world data on the efficacy and safety of trastuzumab for uterine carcinosarcoma.

In our study, participants in the CP arm received a median of 6 cycles of carboplatin and paclitaxel (range: 1–10 cycles), while those in the trastuzumab arm received a median of 6 cycles of carboplatin and paclitaxel (range: 2–9 cycles) in addition to a median of 14 cycles of trastuzumab (range: 4–73 cycles). These findings align with the results of the phase II randomised controlled trial conducted by Fader et al. in 2020 [22]. In that trial, 28 participants in the control arm completed a total of 156 cycles of CP (range: 1–8 cycles), whereas 30 participants in the trastuzumab arm completed 178 cycles of CP (range: 4–9 cycles). Additionally, participants in the trastuzumab arm received a median of 16 cycles of trastuzumab (range: 5–86 cycles). In the present study, the OS was 41 months in the CPT arm and 25.2 months in the CP arm (HR: 0.52, 95% CI 0.345–0.784, p = 0.002), consistent with findings from previous studies [21, 22]. Moreover, 164 patients in this study (59%) were diagnosed with uterine carcinosarcoma. Subgroup analysis based on histological type revealed that the addition of trastuzumab significantly extended OS in the carcinosarcoma group (from 16.8 months to 41.7 months, HR: 0.39), outperforming the extension observed in the serous carcinoma group (from 27.1 months to 38.7 months, HR: 0.56). Notably, an ongoing phase II/III trial (NCT05256225) is investigating whether adding trastuzumab, alone or in combination with pertuzumab (another HER2-targeted antibody), to CP improves outcomes for patients with newly diagnosed HER2-overexpressing uterine serous carcinoma or carcinosarcoma. The findings from this trial, which will conclude in October 2027, may provide insights into HER2-targeted treatment strategies.

HER2-targeted antibody‒drug conjugates (ADCs) like trastuzumab deruxtecan (T-DXd) represent an innovative therapy option. T-DXd combines a HER2-specific antibody with a cytotoxic drug payload, allowing it to deliver chemotherapy directly to HER2-positive cells and potentially affecting nearby cells regardless of HER2 status [24]. In the Japanese STATICE trial, T-DXd showed a 54.5% response rate in HER2-high and a 70% response rate in HER2-low uterine carcinosarcoma patients, with response durations of 6.9 and 8.1 months, respectively [25]. Similarly, the DESTINY-PanTumor02 trial demonstrated that in heavily pretreated patients with HER2-expressing solid tumours, endometrial cancer patients achieved an overall response rate of 84.6% and a median PFS of 11.1 months [26].

Despite the promise of trastuzumab as targeted therapy for HER2- HER2-overexpression endometrial cancer, there are significant adverse effects associated with the addition of trastuzumab to CP for the treatment of late-stage HER2-overexpressing uterine cancer. In this study, adverse events included left ventricular systolic dysfunction, hypertension, and diarrhoea. Two percent of patients in the CP group and 9% of patients in the CP + T group experienced a decrease in the left ventricular ejection fraction (LVEF). Although the overall rate of hypertension was higher in patients who received trastuzumab than in those who did not, there was no significant difference in the incidence of Grade 3 hypertension between the two groups. Furthermore, diarrhoea rates were higher in the trastuzumab group than in patients who did not receive the group; however, this did not result in a significant increase in the rate of electrolyte imbalance (Table 2). This outcome aligns with that of the multicentre phase II trial conducted by Fader et al., which indicated that 3% of patients in the CP + T group and no patients in the CP group experienced a decrease in LVEF [21]. Hence, these findings underscore the importance of monitoring cardiac function and related symptoms in patients receiving trastuzumab combined with CP. Early detection and intervention may help to mitigate the risk of severe cardiac complications.

The mechanisms behind trastuzumab-induced cardiotoxicity are not fully understood; however, impaired HER2 signalling may play a role. Trastuzumab disrupts HER signalling, leading to an autophagy-inhibitory cascade (Erk/mTOR/Ulk1) that affects the ability of cardiomyocytes to clear toxic substances, resulting in cardiotoxicity [27]. A meta-analysis of seven trials demonstrated cardiac dysfunction rates of 3–7% with trastuzumab alone, which increased to 13% with trastuzumab plus paclitaxel, and 27% with trastuzumab combined with anthracyclines, compared to under 7% with anthracyclines alone [28, 29]. This highlights a greater risk of cardiac toxicity when trastuzumab is used alongside other chemotherapy drugs.

Although our study has the advantage of including the largest cohort of patients with late-stage HER2-overexpressing uterine serous carcinoma and carcinosarcoma, it is crucial to recognise certain limitations. Firstly, the study did not have a randomised controlled design, and there was no detailed information on dosages for each treatment, nor on the timing or classification of second or third-line treatments. Nevertheless, efforts were made to minimise selection bias through PSM of baseline characteristics and prior treatment. Second, our reliance on coding systems, such as ICD, RxNorm, HCPCS, and CPT, to identify patient diagnoses, adverse events, lines of chemotherapy, and treatment regimens may have introduced misclassifications or reporting delays that potentially affected the analysis outcomes. However, the TriNetX data originated from developed countries in Europe and America, where the personnel responsible for data entry are typically well-trained, which reduces the likelihood of errors. Nonetheless, further clinical study is required to corroborate and validate our findings, and a deeper understanding of trastuzumab-induced cardiotoxicity is crucial for optimising treatment strategies and ensuring the best possible patient outcomes within the evolving landscape of targeted therapies for HER2-overexpressing cancers.

Conclusions

The addition of trastuzumab to frontline treatment with CP for patients with stage III/IV HER2-overexpressing uterine serous carcinoma or carcinosarcoma significantly improved OS. Although this finding highlights the potential benefits of incorporating trastuzumab into treatment regimens for this patient population, it is crucial to emphasise the importance of routine, careful monitoring of cardiac function and signs of congestive heart failure when administering trastuzumab. The increased risk of adverse cardiac events associated with trastuzumab underscores the need for early detection and intervention to manage these potential complications effectively.