Abstract
Background/Aim: We hypothesized that adding bevacizumab to platinum-based neoadjuvant chemotherapy – whose efficacy for patients with recurrent or metastatic cervical cancer has already been proven – could optimize the therapy regimen, leading to improved response rates and survival outcomes. Patients and Methods: Forty patients with histologically confirmed cervical cancer with FIGO stage IB3-IVA who received platinum-based neoadjuvant treatment between March 2008 and January 2019 in the Department of Obstetrics and Gynecology of University Hospital Cologne were analyzed. Twenty patients were treated with additional bevacizumab. The comparative cohort consisted of 18 patients treated with neoadjuvant chemotherapy alone. The response rates and clinical outcomes, including progression-free survival and overall survival, were evaluated. Results: Neoadjuvant chemotherapy combined with bevacizumab significantly improved the response rate (p=0.046). The survival analysis showed that patients treated without bevacizumab had better progression-free survival up to FIGO stage IVA than patients treated with bevacizumab. However, overall survival was similar for both cohorts. For patients with advanced tumor stage, including FIGO IVB, progression-free survival and overall survival improved with the addition of bevacizumab. Pathological complete remission was a statistically significant prognostic factor for progression-free survival (p=0.039) but did not significantly affect overall survival (p=0.098). Conclusion: While bevacizumab did not demonstrate a significant improvement in overall survival rates, it was associated with a notable reduction in tumor size and showed a trend towards improved clinical response rates. These findings suggest that bevacizumab may have potential in optimizing the neoadjuvant treatment approach.
Cervical cancer is the fourth most common cancer in women with a global incidence of 604,000 new cases and 342,000 deaths in 2020 (1). It is most frequently diagnosed in women between the ages of 35 and 44 with the average age at diagnosis of 50. Despite widespread screening programs and the use of vaccines against human papillomavirus, more than half of patients (54%) in the United States are diagnosed with advanced cervical cancer (2), which is even higher in the developing world (3). In Germany, approximately 4,640 women were diagnosed with invasive cervical cancer in 2020. In total, approximately 1,600 women in Germany die of cervical cancer every year (4). The prognosis of advanced cervical cancer is poor with a 5-year survival rate of 17% (3). Moreover, age standardized mortality rate per 100,000 people in Germany has stagnated since early 2000, indicating that rates may be approaching a lower limit (4, 5).
Chemoradiation (CRT) is the standard of care for International Federation of Gynecology and Obstetrics (FIGO) stage IB-IVA since 1999 (6). However, due to its primary role as a local therapy, chemoradiation carries the risk of local failure, which can be accompanied by distant metastasis (7). Moreover, the side effects of chemoradiation, such as vaginal bleeding, fistula, and vaginal atrophy are frequently more severe than those of systemic therapy, especially in young women (8). The limited efficacy of primary chemoradiation for advanced cervical cancer, coupled with a lack of development over the last decade and the poor prognosis associated with advanced stages of the disease, highlights the need for further development and optimization of therapeutic responses.
Neoadjuvant chemotherapy (NACT) followed by radical hysterectomy has been considered an alternative approach to improve disease control and reduce toxicity. Cervical cancer has shown a high response rate to chemotherapy including taxane and platinum-based therapy (9). Additionally, high-dose density NACT appears to be more effective (10). Neoadjuvant chemotherapy could eradicate or alter micrometastasis and reduce systemic failures, in addition to facilitating local control by radical hysterectomy. Although many studies have demonstrated feasible outcomes for NACT and surgery regarding response rates and toxicity, its impact on overall survival remains still unproven (11-13).
Similar to other types of pre-malignant lesions and carcinomas, angiogenesis is associated with high-grade cervical dysplasia and invasive cervical cancer (14). Human Papilloma virus (HPV) is the primary carcinogenic driver of cervical cancer, and the oncogenic capability of this virus is partially due to stimulation of malignant angiogenesis. Vascular endothelial cell growth factor (VEGF) is one of the most important inducers of angiogenesis, which is up-regulated in cervical cancer (15). Bevacizumab is a recombinant human monoclonal VEGF antibody, which inhibits angiogenesis of the tumor through competitive inhibition of VEGF receptors (16, 17). The addition of bevacizumab to chemotherapy has been the standard of care for metastatic and locally advanced cervical cancer following a phase III GOG 240 trial conducted by Tewari et al., in which 452 patients with recurrent or metastatic cervical cancer were assigned to receive standard chemotherapy with or without bevacizumab. Significant improvements in overall survival (OS) and progression-free survival (PFS) were achieved, resulting in approval of bevacizumab (18).
We hypothesized that adding bevacizumab to the platinum-based NACT, which has already demonstrated efficacy in patients with recurrent or metastatic cervical cancer, could optimize the therapy regimen and improve response rates and survival outcomes. We have previously reported on the efficacy of bevacizumab in a NACT setting, demonstrating that combined therapy with bevacizumab led to a better clinical response and improved operability (19). Here, we report on the impact of bevacizumab on OS and PFS.
Patients and Methods
Study design and population. This is a retrospective study approved by the local Ethics committee. In total, 49 patients with histologically confirmed cervical cancer who received platinum-based systemic treatment in the period from March 2008 to January 2019 in the Department of Obstetrics and Gynecology of University Hospital Cologne were screened. Among them, nine were initially diagnosed with FIGO stage IVB. Ultimately, 40 patients with FIGO 2018 stage IB3-IVA were included in the study; 22 patients were treated with NACT with bevacizumab and 18 patients received NACT alone. The deciding factor for adding bevacizumab was the time of treatment. In June 2013, the results of the GOG 240 trial were presented in ASCO. Therefore, patients in the NACT + bevacizumab cohort began treatment in June 2013. Ten patients in the NACT cohort were treated before June 2013. Eight patients were treated without bevacizumab since June 2013: one patient already had a supracervical hysterectomy, four of them had bulky disease with FIGO IBIII, one patient had a history of thrombosis, one patient was treated extra muros and came to our clinic for surgery.
22 patients received NACT with bevacizumab (NACT+B): 12 patients were treated with 12 cycles of paclitaxel 80 mg/m2 weekly, four cycles of cisplatin 75 mg/m2 every three weeks as dose-dense scheme, and four cycles of bevacizumab 15 mg/kg every three weeks. Five patients received initially 18 cycles of paclitaxel 80 mg/m2 weekly, six cycles of cisplatin 50 mg/m2 every three weeks, and six cycles of bevacizumab 15 mg/kg every three weeks, one of which had to change from paclitaxel to topotecan because of an allergic reaction. Five patients were administered a combination of eight cycles of cisplatin 40 mg/m2 weekly as another form of dose dense scheme, eight cycles of docetaxel 35 mg/m2 weekly, and eight cycles of bevacizumab 5 mg/kg weekly. The cumulative dose of cisplatin ranged from 300 mg/m2 to 320 mg/m2, while the cumulative dose of bevacizumab ranged from 40 mg/kg to 90 mg/kg. For the comparative cohort, 18 patients who underwent platinum-based chemotherapy without bevacizumab were included (NACT): 15 patients were treated with eight cycles of cisplatin 40 mg/m2 weekly and eight cycles of docetaxel 35 mg/m2 weekly. Three patients received 12 cycles of paclitaxel 80 mg/m2 weekly and four cycles of cisplatin 75 mg/m2 every three weeks. For this cohort, the cumulative dose of cisplatin ranged also from 300 mg/m2 to 320 mg/m2.
Patient characteristics at the time of initial diagnosis, such as date of histologic confirmation, age, FIGO stage (revised in FIGO 2018 stage), histological type, HPV status, tumor size in mm, NACT regimen, type of radical hysterectomy, pathological/clinical responses, consolidation therapy and adverse events were recorded. Pre-treatment assessment included patient medical history, physical examination, vaginal–pelvic examination, biopsy, blood analysis including tumor markers, chest- abdominal–pelvic computed tomography (CT) scan and magnetic resonance imaging (MRI). CT scans and MRIs were repeated 2-3 weeks after the completion of systemic therapy. The images were assessed radiologically according to Response Evaluation Criteria In Solid Tumors (RECIST). If compared to the pre-treatment imaging the tumor was no longer radiologically detectable, it was defined as clinical complete response (cCR). If tumor size regressed by more than 30% compared to the previous imaging, it was defined in accordance with RECIST criteria as clinical partial response (cPR) (20). When there was no residual tumor on surgical specimen after histopathologic examination, this was recorded as pathologic complete response (pCR). In case of residual tumor with regression in size more than 30% after histologic examination, it was defined as pathologic partial response (pPR). The longest diameter in mm was used to measure tumor size and the change in tumor size was calculated as the mean of the difference before and after therapy. The follow-up was performed on the regular basis with clinical, and if necessary, radiological examinations until patients either died or until May 2022. The median follow-up time was 50 months (range=38-92 months).
Statistical analysis. Data were recorded on a predesigned excel spreadsheet (Excel® 2010, Microsoft Corporation, Redmond, WA, USA) from clinical information systems (ORBIS® OpenMED, AGFA HealthCare NV, Cato®) and transferred to SPSS 26.0 software for statistical analysis (SPSS Inc., Chicago, IL, USA). Significant statistical differences between groups were estimated using the independent t-test for quantitative variables and chi-square test for categorical variables. The time between the first cycle of systemic treatment until the detection of progression was defined as PFS. The time between the first cycle of systemic treatment until death or last observation was defined as OS. The median follow-up period was 50 months. Survival curves of PFS and OS were analyzed using Kaplan–Meier method and compared using log-rank test. Using Cox proportional hazard model, the association between survival and impact of bevacizumab was estimated. Given a small sample size, post hoc power calculation was performed.
Results
The clinicopathological characteristics of patients are summarized in Table I. Median age of the patients in the NACT (NACT) cohort was 46.2 years (range=27-70 years) and in the NACT + bevacizumab (NACT+B) cohort 48.2 years (range=29-68 years). Nine patients (50%) in the NACT cohort had Stage IB3 with bulky disease, defined as a tumor of 4 cm or larger. Ten patients (45%) in the NACT+B cohort had Stage IIB with parametrial involvement. The number of patients with FIGO stage IVB who had metastases and were treated with systemic therapy prior to surgery or (chemo)radiation was documented separately for each cohort. The most common histological type was squamous cell carcinoma. Grading was evenly distributed in each cohort. The histological type of two patients in the NACT cohort and the grading of three patients in the NACT+B cohort could not be determined, since the initial diagnosis was confirmed extra muros and these patients reached pCR after NACT, therefore there was no residual tumor to examine. Three patients (17%) in the NACT cohort had positive nodal status, whereas in the NACT+B cohort six patients (27%) had nodal involvement. In case of suspicious positive nodal status, lymphonodectomy was performed prior to hysterectomy, two patients (11%) in the NACT cohort and seven patients (32%) in the NACT+B cohort underwent this procedure. In the NACT cohort, 16 patients (89%) underwent radical hysterectomy, including complete lymphadenectomy if it had not been carried out previously. Of two patients (11%) without radical hysterectomy, one patient had already undergone laparoscopic supracervical hysterectomy before the initial diagnosis. The other patient was pregnant, thus trachelectomy was performed during pregnancy followed by chemotherapy. Post-partum surgery was followed by hysterectomy. Seventeen patients (76%) in the NACT+B cohort underwent radical hysterectomy, and three patients (14%) were treated with chemoradiation directly after NACT+B. One patient wished for a fertility preserving procedure, thus NACT+B followed by trachelectomy was performed. About 1/3 of patients in each cohort underwent consolidation therapy.
Clinicopathological characteristics of patients.
Table II shows the response rate after NACT with or without bevacizumab. In 22 patients (100%) of the NACT+B cohort a significantly improved clinical response rate was observed, whereas in the NACT cohort 15 patients (83%) showed clinical response (p=0.046). The rate of pCR was higher in patients treated with NACT without bevacizumab [44% (n=8/18) in NACT vs. 23% (n=5/22) in NACT+B; p=0.144]. Accordingly, more cCRs were observed in the NACT cohort than in the NACT+B cohort [50% (n=9/18) in NACT vs. 32% (n=7/22) in NACT+B; p=0.243]. In the NACT cohort with pCR, five patients had FIGO IB3, one patient had FIGO IIB, and two patients had FIGO IIA. Additionally, one patient with cCR had FIGO IB3. In the NACT+B cohort with pCR, four patients presented FIGO IIB, and one patient had FIGO IIIC. Two patients with cCR in the NACT+B cohort presented FIGO IIIC and IVA, respectively. Pathological partial response (pPR) was more often observed in the NACT+B cohort than in the NACT cohort [33% (n=6/18) in NACT vs. 55% (n=12/22) in NACT+B; p=0.180]. Accordingly, a higher rate of cPR in the NACT+B cohort was reported, which was statistically significant [33% (n=6/18) in NACT vs. 68% (n=15/22) in NACT+B; p=0.028]. There were three patients with positive nodal status in the NACT cohort; after therapy the status remained unchanged. In the NACT+B cohort, three patients out of six (50%) converted their positive nodal status to negative [0% (n=0/3) in NACT vs. 50% (n=3/6) in NACT+B; p=0.13].
Response rates depending on neoadjuvant chemotherapy (NACT) with or without bevacizumab.
The pre- and post-treatment tumor size could be assessed objectively by imaging (MRI, CT) according to RECIST criteria (Figure 1). With bevacizumab, 85% reduction in tumor size was achieved, whereas in the NACT cohort without bevacizumab, 71% reduction was observed (p=0.149).
The waterfall plot shows the percentage change in tumor size before and after treatment, comparing patients who received bevacizumab with those who did not.
Figure 2 shows the 3 year-PFS and OS with the advanced FIGO stage in patients treated with platinum-based NACT with or without bevacizumab. For patients up to FIGO Stage IIIC, the 3-year PFS was 88% in the NACT cohort and 78% in the NACT+B cohort (p=0.54), while the 3-year OS was the same for both cohorts at 86% (p=0.89) (Figure 2A1 and 2B1). After including patients with stage IVA, the 3-year PFS was 84% in the NACT cohort and 78% in the NACT+B cohort (p=0.91), and the 3-year OS was still 86% for both cohorts (p=0.90) (Figure 2A2 and 2B2). When all stages of cervical cancer including stage IVB with metastasis are considered, the 3-year PFS was 66% in the NACT cohort and 72% in the NACT+B cohort [p=0.48, hazard ratio=1.434 (95%CI=0.519-3.957)]. The 3-year OS showed a trend favoring the NACT+B cohort [71% in NACT cohort vs. 76% in NACT+B cohort, p=0.617, hazard ratio=1.352 (95%CI=0.412-4.432)] (Figure 2A3 and 2B3). This trend suggests that, with advanced tumor stages, PFS and OS improve with the addition of bevacizumab. Additionally, we evaluated whether pCR plays a role as a prognostic factor. As shown in Figure 2C, pCR is a statistically significant prognostic factor for PFS (p=0.039), while a trend toward significance was observed for OS (p=0.098).
Progression-free survival (PFS) and overall survival (OS) percentages according to FIGO stage in patients treated with platinum-based neoadjuvant chemotherapy with/without bevacizumab. A1) PFS up to FIGO IIIC, A2) PFS up to FIGO IVA, A3) PFS for all stages including IVB. B1) OS up to FIGO IIIC, B2) OS up to FIGO IVA, B3) OS for all stages including IVB. C) PFS (C1) and OS (C2) for the patients with pathological complete remission (pCR) and non-pCR.
As pCR was confirmed as a prognostic factor, characteristics that could be associated with pCR were examined. These included grading, histological subtype, dose dense regimen, age of patients (with a cut off at 50 years), and cumulative doses of cisplatin and bevacizumab. No significant associations were found with these variables (Table III).
Relationship between pathological complete response (pCR) and tumor characteristics.
Regarding adverse events (AEs), there were no treatment-limiting side effects in both cohorts, based on the evaluation possible in this retrospective study. No fistulas were reported in both cohorts. Of note, in the NACT+B cohort, there were no thromboembolic events or AEs of Common Terminology Criteria for AEs (CTCAE) grade ≥3, including hypertension or proteinuria (21). However, post-surgical complications, such as lymphocele, ischemic ureter lesion, and dehiscence were reported in the NACT+B cohort. In the NACT cohort, one patient experienced pancytopenia Grade III and another patient had kidney failure Grade III. There were no reported surgical complications in this cohort.
Discussion
Bevacizumab combined with first line chemotherapy has been shown to improve OS in patients with metastatic and recurrent cervical cancer. Additionally, the RTOG 0417 trial, which treated 49 patients with FIGO stage IB-IIIB with bevacizumab concurrently with chemoradiation for curative intent, demonstrated the efficacy of this therapeutic approach. The toxicities observed in this trial were manageable (22). Here, we hypothesized that adding bevacizumab to NACT could optimize the treatment regimen, leading to better response rates and outcomes compared to conventional therapy.
The rationale for utilizing bevacizumab is based on clinical data from metastatic cervical cancer and other tumor types, suggesting that angiogenesis is a pathologic process that promotes neoplastic progression. Surrogates for angiogenesis such as micro-vessel density (MVD) and VEGF expression have been shown to be up-regulated in cervical cancer and are associated with poor prognosis (15, 23). Moreover, HPV infection, which is the primary carcinogenic factor of cervical cancer, and the subsequent incorporation of the pathogenic E6 oncoprotein, leads to the degradation of the p53 tumor suppressor. Inactivation of p53 has numerous deleterious effects on various pathways, including hypoxia inducible factor 1 (HIF-1), a transcription factor stimulating neo-angiogenesis through VEGF and other gene products (24). The involvement of HPV proteins in the angiogenic switch has shown to significantly alter endothelial cell behavior (25). Co-expression of the VEGF promoter-Luc (luciferase) reporter gene with E6 showed that E6 oncoprotein up-regulates VEGF promoter activity in a p53-independent manner, suggesting that the HPV oncoprotein E6 may contribute to tumor angiogenesis by direct stimulation of the VEGF gene (14). In our study, we observed that NACT with bevacizumab can significantly improve response rates (p=0.046). While a higher rate of pCR, both clinically and pathologically, was observed in the NACT cohort without bevacizumab, every patient treated with additional bevacizumab showed a significantly better response to the therapy.
Regarding PFS and OS, it should be noted that the post hoc power calculation for survival analysis was low, ranging between 0.066 and 0.12, due to the small number of patients in the present work. Therefore, no precise conclusions can be drawn about survival outcomes. Even though it shows that patients treated with NACT without bevacizumab showed better PFS up to FIGO stage IVA compared to patients treated with bevacizumab, OS was same for both cohorts. More patients with advanced disease from FIGO IIIA were treated with bevacizumab [3 patients (17%) in NACT cohort vs. 9 patients (41%) in NACT+B cohort]. When all stages of cervical cancer including metastatic disease are considered, the 3-year PFS was 66% in the NACT cohort and 72% in the NACT+B cohort, respectively (p=0.48, hazard ratio=1.434, 95%CI=0.519-3.957). The 3-year OS also changes in favor of the NACT+B cohort [71% in NACT cohort vs. 76% in NACT+B cohort, p=0.617, hazard ratio=1.352, 95%CI=0.412-4.432]. These results are in concordance with the results from the GOG 240 trial for recurrent, persistent, or metastatic cervical cancer (18) and suggest that anti-angiogenesis does not play a crucial role in curative de novo situations as it does in metastatic settings. Microscopic disease does not seem to have an established vasculature and thus may be resistant to anti-angiogenesis therapy. Tumors establish a vasculature in different ways with varying sensitivity to VEGF inhibition (26). The vascularization and, consequently, sensitivity to angiogenesis inhibition may differ between micro-metastatic and macro-metastatic disease (27).
The hypothesis was that adding bevacizumab to NACT would increase the rate of pCR. This objective was based on the fact, that optimal pathological response represents a strong prognostic factor for OS (28). In the SNAP01 study, which compared the combination of paclitaxel, ifosfamide, and cisplatin versus ifosfamide and cisplatin before radical surgery, an optimal pathological response was an independent predictive factor of OS with an HR of 5.88 (95%CI=2.50-13.84) (29). A further Italian multicenter retrospective CTF study, in which 333 patients with FIGO stage IB2-IIB cervical cancer treated with different platinum-based regimens followed by radical surgery were recruited, confirmed that the pathological response to NACT was an independent prognostic factor for both PFS and OS (30). Non-responders had a 2.757-fold higher risk of recurrence and a 5.413-fold higher risk of death than those who obtained an optimal response. Although the results did not meet our goal, our study reaffirmed that pathological complete remission is a statistically significant prognostic factor for PFS (p=0.039), while it did not reach statistical significance for OS (p=0.098).
Next, we aimed to identify predictive factors associated with pCR, such as tumor grading, histological subtype, dose-dense chemotherapy regimen, age of patients, and cumulative doses of cisplatin and/or bevacizumab. Due to the limitations of a retrospective study, factors like HPV status, p16 status, and lymphatic and vascular invasion could not be fully evaluated. Since neuroendocrine tumors and adenocarcinomas display increased MVD and VEGF expression, therapy with bevacizumab may result in a better response (15, 31). Furthermore, the assessment of dose-dense or weekly regimens, such as weekly paclitaxel and carboplatin has shown promising results with favorable toxicity profiles compared with standard platinum-based chemotherapy in patients with recurrent or metastatic cervical cancer (32). Dose-dense regimens can increase the proportion of cancer cells exposed to the vulnerable phases of the cell cycle while cytotoxic concentrations are still present. Additionally, shorter infusion times and lower fractional doses can reduce myelosuppression and other toxicities associated with the standard 3-weekly schedule (33). However, none of these factors showed association with pCR, which may be possibly attributable to the small patient cohort.
The adverse effect profile of bevacizumab was not beyond the known range and did not lead to discontinuation of the treatment. Noteworthy, more post-surgical complications such as ischemic ureter lesion and dehiscence were reported in the NACT+B cohort. Therefore, more caution is needed in case of surgery, ensuring a sufficient interval after the last administration of bevacizumab (34).
The results of the present study did not prove our hypothesis. The small number of patients and non-homogeneous patient characteristics were weak points in this study. All patients received platinum-based chemotherapy with or without bevacizumab, but the therapy regimen was not consistent. Nevertheless, despite the advances in treatment and the practice of definitive chemoradiation, the 5-year survival rate remains low (35), highlighting a clinical unmet need for patients with advanced cervical cancer. NACT before definitive further therapy still represents an attractive approach. The possible advantages include a reduction of tumor size and control of circulating tumor cells and micro-metastasis. In the INTERLACE trial, 500 patients with advanced cervical cancer were treated with either chemoradiation (CRT) alone or induction chemotherapy (IC) with six cycles of carboplatin AUC2 and paclitaxel 80 mg/m2 followed by the same chemoradiation. This simple concept resulted in significantly improved 5-year PFS rate with 73% with IC, versus 64% with CRT alone (36). In the KEYNOTE-A18 trial, patients with high-risk and locally advanced cervical cancer were randomized 1:1 to receive either pembrolizumab or placebo concurrent with CRT, followed by maintenance therapy. The PFS hazard ratio was 0.59 (95%CI=0.43-0.82) (37), resulting in FDA approval of pembrolizumab with chemoradiotherapy for FIGO 2014 Stage III-IVA cervical cancer in January 2024 (38). All the evidence underscores the importance of systemic therapy, in addition to cisplatin as a sensitizer, to optimize the response to chemoradiation and extend local control. Especially for young patients wishing to preserve fertility, NACT followed by fertility preserving surgery in selected patients could be a viable alternative, offering a treatment option that does not compromise oncologic or obstetric outcomes (39, 40). Although we failed to demonstrate the efficacy of bevacizumab as a practice changing neoadjuvant treatment, we observed a significant reduction in tumor size and a trend towards improved clinical response rates with its addition. This suggests that bevacizumab has the potential to optimize neoadjuvant therapy. Currently, the results of carboplatin-paclitaxel-pembrolizumab as neoadjuvant treatment for locally advance cervical cancer (MITO CERV 3 trial) are anticipated (41) and further investigation is needed to optimize response rates and outcomes.
Acknowledgements
The Authors would like to thank all the patients that participated in the study.
Footnotes
Authors’ Contributions
SB wrote the draft of the manuscript and performed data collection and statistical analysis. DR contributed to the study concept. CD and MR contributed to the data analysis and interpretation. BS performed additional histological analysis. KN contributed to further data collection. YZ was involved in conceptualizing the study. PM approved the final version. HP contributed to writing the final version of the manuscript and supervised the data analysis.
Conflicts of Interest
All Authors declare that they have no conflicts of interest in relation to this study.
Funding
There was no external funding for the study.
- Received July 19, 2024.
- Revision received August 17, 2024.
- Accepted August 20, 2024.
- Copyright © 2024 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.