Efficacy of Recombinant Human Endostatin plus Neoadjuvant Chemotherapy for Osteosarcoma and Its Influence on Serum VEGF and MMP-9 Levels

Objective To investigate the efficacy of recombinant human endostatin (rh-Endo) plus neoadjuvant chemotherapy (NACT) for osteosarcoma (OSA) and its influence on serum vascular endothelial growth factor (VEGF) and matrix metalloproteinase-9 (MMP-9). Methods The case data of 141 OSA patients presented to the North District, Xiangyang Central Hospital Affiliated to Hubei University of Arts and Sciences from January 2018 to June 2019, were analyzed retrospectively. Patients receiving NACT (methotrexate + ifosfamide + adriamycin) were assigned into the control group (CNG; n = 65), while those treated with rh-Endo plus NACT were included in the combination group (CMG; n = 76). The following aspects were compared: clinical efficacy, serum tumor markers, serum VEGF and MMP-9 contents, inflammatory factors, incidence of adverse reactions, limb function scores at 6 months of follow-up, and prognostic quality of life (QOL). Results A statistically higher overall response rate (ORR) was determined in CMG versus CNG (84.2% vs. 64.6%, P < 0.05). The pretreatment serum bone alkaline phosphatase (BALP), insulin-like growth factor (IGF)-1, serum amyloid A (SAA), VEGF, MMP-9, C-reactive protein (CRP), tumor necrosis factor (TNF)-α, and interleukin (IL)-10 levels differed insignificantly between the two cohorts (P > 0.05); while except IL-10 that showed increased expression in both cohorts and was comparatively higher in CMG, the other 8 parameters reduced in both cohorts after 2 weeks of drug withdrawal, and the reduction of each parameter was more significant in CMG (P < 0.05). The total adverse reaction rate was 30.2% in CMG, which was higher than that of 36.9% in CNG, albeit without a statistical difference (P > 0.05). An evidently higher 2-year survival rate was determined in CMG (P < 0.05). Conclusions rh-Endo plus NACT is more effective than NACT alone in the treatment of osteosarcoma, which can validly restore the balance of vascular endothelial cells, reduce inflammation, and is worth promoting in clinic.


Introduction
Osteosarcoma (OSA) is the most commonly seen primary bone malignancy derived from primitive osteoid mesenchymal cells, and its pathogenesis is closely related to puberty hormone changes and bone growth [1,2]. Te onset age of OSA presents a bimodal distribution, most of which occurs in adolescents aged 10-14 and the elderly aged over 60 [3]. In adolescence, the disease often occurs in the long bone metaphysis of the extremities, such as the distal femur and the proximal tibia [4,5]. For the elderly, however, the predilection sites are diverse, which can also occur in the craniofacial region [6]. OSA has a high propensity to metastasize, specifcally to the lungs, with more than 85% of patients developing lung metastasis at the time of initial treatment [7]. At present, surgery combined with adjuvant chemotherapy remains the mainstay treatment option for OSA. Although this strategy has improved OSA patients ' 5-year survival to some extent, the survival is still extremely low, especially for those with metastatic and recurrent OSA, with a survival rate less than 20% [8,9]. In addition to distant metastasis, the prime reasons for an adverse prognosis are poor response to chemotherapy and drug resistance to chemotherapeutics [10]. Terefore, there is a growing demand for new drugs and treatments for OSA.
Precise surgical resection and combined neoadjuvant chemotherapy (NACT) established by methotrexate (MTX), adriamycin (ADM), cisplatin, and ifosfamide (IFO) can improve OSA patients' 5-year survival to 60-70% [11]. Today, most patients receive preoperative NACT and postoperative adjuvant chemotherapy. Unfortunately, the utility of many chemotherapeutics is limited by their toxicity and side efects. For example, MTX can be fatal in excess of a certain dose [12], ADM-induced cardiomyopathy can cause permanent irreversible damage to the heart of juvenile patients [13], and the addition of ifosfamide (IFO) can increase the toxicity of chemotherapy [14].
In recent years, some new treatments for OSA have emerged, with a growing number of drugs developed focusing on the pathogenesis of OSA. Previous literature has revealed that OSA, like other tumors, is dependent on angiogenesis [15]. Terefore, antiangiogenesis has become a new therapeutic approach. Recombinant human endostatin (rh-Endo) is a molecular targeted therapeutic drug, which has the efect of inhibiting tumor growth, invasion, and metastasis with little damage to normal vascular endothelial cells; moreover, it can specifcally inhibit tumor tissues, with less toxic and side efects and strong pertinence [16]. Antiangiogenic therapeutic strategies are divided into two categories: one is the therapeutic strategy that directly produces cytotoxicity on endothelial cells, and the other is the therapeutic strategy that indirectly acts by eliminating key proangiogenic factors such as vascular endothelial growth factor (VEGF) [17]. As an essential target of matrix metalloproteinases (MMPs) and other angiogenesis factors and proteases, VEGF is critical in tumor metastasis [18]. Among MMPs, MMP-9 is the product of VEGF receptor (VEGF-R) activation, and the expression of this downstream factor shows the potential to be a surrogate marker of VEGF-R activity [19]. Studies have shown that the antiangiogenesis of rh-Endo can prevent metastatic OSA from further progressing [20]. Yet there is little information directly correlating serum MMP-9 levels with VEGF activity in OSA. Accordingly, this study's novelty and aim lie in the investigation of the therapeutic efect of rh-Endo plus NACT for OSA and its infuence on serum VEGF and MMP-9 levels.

Research Participants.
Te clinical records of 141 patients with a confrmed diagnosis of OSA by symptoms, signs, imaging, and pathology and treated in the North District at Xiangyang Central Hospital Afliated to Hubei University of Arts and Sciences from January 2018 to June 2019 were analyzed retrospectively ( Figure 1). Patients receiving NACT (MTX + IFO + ADM) were assigned into control group (CNG; n � 65), while those treated with rh-Endo plus NACT were included in the combination group (CMG; n � 76). Inclusion criteria were included as follows: Patients with no distant metastasis found by pathological examination, an estimated survival time greater than 1 year, voluntary limb salvage awareness, and no other neoplastic diseases, hematopoietic abnormalities, or chemotherapy contraindications. Te exclusion criteria included OSA with distant metastasis, severe acute infection, pyogenic infection, and chronic infection, protracted course of wound healing or severe bleeding tendency, vital organ dysfunction, severe heart disease, pregnant and lactating patients, severe mental illness, and poor compliance with clinical treatment. Te two cohorts had similar general clinical data, indicating comparability (Table 1). Tis Ethics Committee of Xiangyang Central Hospital Afliated to Hubei University of Arts and Sciences has given its approval for conducting this study.

Treatment
Regimens. Both groups received the MMIT regime for NACT: MTX (Haizheng Pfzer Pharmaceutical Co., Ltd., 5 mg × 5 injections) was injected intravenously at 10 g/m 2 in the frst and second weeks; in the third week, IFO (Guangdong Lingnan Pharmaceutical Co., Ltd., 1 g for injection) was given by intravenous drip at 2 g/m 2 for 5 days, and in the third week, intravenous ADM (30 mg/m 2 Shanxi Pude Pharmaceutical Co., Ltd., 10 mg × 10 injections) was administered for 3 days. With 4 weeks as a cycle, the chemotherapy was performed for two cycles. On this basis, patients in CMG were treated with rh-Endo (Shangdong Simcere Pharmaceutical Group Ltd., 15 mg: 3 mL), 15 mg per time intravenously, for 2 weeks. Te granisetron, dexamethasone, and metoclopramide were routinely used for antiemetic treatment during chemotherapy, and a subcutaneous injection of colony-stimulating factor was used prophylactically after chemotherapy to protect the bone marrow. Both groups were treated for 2 cycles, and surgery was performed 2 weeks after drug withdrawal. [21]. Complete resolution of the lesion was considered a complete response (CR); ≥50% decrease of the maximum bi-diameter product of the lesion that lasted for more than 4 weeks was deemed a partial response (PR); stable disease (SD) is indicated if the maximum bi-diameter product of the lesion was reduced by < 50% with no new lesions for over 4 weeks;: ≥25% increase of the maximum bi-diameter product of the lesion or occurrence of new lesions was deemed progressive disease (PD). Te efective rate � (CR + PR) cases/total cases × 100%.

Detection of OSA-Related Tumor
Markers. From all patients, fasting peripheral venous blood was collected before treatment and two weeks after drug withdrawal, for anticoagulation treatment and serum separation. Serum levels of OSA-associated tumor markers (bone alkaline phosphatase (BALP), insulin-like growth factor (IGF)-1, and serum amyloid A (SAA)) were detected by enzyme-linked     Te Kolmogorov-Smirnov Z normal distribution test and Levene variance homogeneity test shall be passed before the test. For intragroup before-after comparisons of continuous variables represented by mean ± standard deviation, the paired sample t-test was used; for between-group comparisons, the independent sample t-test was adopted; for multigroup comparisons, a one-way ANOVA was used followed by the Bonferroni post-hoc test (α � 0.05). Categorical variables, described as n (%), were analyzed using the chi-square test. Te signifcance level was P < 0.05.

Serum VEGF and MMP-9 Levels.
Te pretreatment serum VEGF and MMP-9 levels difered insignifcantly between the two cohorts (P > 0.05). Two weeks after treatment discontinuation, the two parameters decreased in both cohorts, and the decreases were more obvious in CMG (P < 0.05). Figure 3.

Serum Infammatory Factors.
No signifcant intergroup diferences were observed regarding the pretreatment serum CRP, TNF-α, and IL-10 levels (P > 0.05). Two weeks after treatment discontinuation, CRP and TNF-α levels decreased in both cohorts while IL-10 increased, and the changes were more obvious in CMG (P < 0.05). Figure 4.

QOL.
Patients' QOL was compared based on scores of physical functioning, cognitive functioning, role functioning, social functioning, and emotional functioning after one year of follow-up. Te results revealed no marked diference in pretreatment QOL between groups (P > 0.05). After one year of follow-up, the scores of the various dimensions in CMG were obviously higher, suggesting a better QOL in patients receiving rh-Endo plus NACT (P < 0.05).

Discussion
Te specifc etiology of OSA has not been clarifed. But relevant research reports have revealed its correlation with immune system disorders, radiation, virus infections, and genetic factors [24,25]. After the episode, there will be intermittent pain in the afected part, which will be aggravated with the progression of the disease. Finally, painkillers will not help, and the tumor may develop and metastasize to the lungs to harm the patient's respiratory function, leading to high disability and mortality [26]. Chemotherapy is currently one of the most efective methods in addition to surgical treatment, but chemotherapeutics generally lack targeting and have extensive cytotoxicity. When chemotherapeutics enter the body, normal cells, especially those with vigorous metabolism, can be damaged to varying degrees. MMIT is a commonly used NACT regimen for OSA in the clinic to kill tumor cells, in which MTX can inhibit DNA synthesis mainly by mediating the S phase of the cell cycle, IFO can cross-link with DNA to exert cytotoxicity, and ADM can inhibit the synthesis of RNA and DNA [27,28]. rh-Endo injection is a therapeutic drug with molecular targeting function, which can specifcally block the formation of tumor blood vessels, thus inhibiting tumor cell growth, infltration and metastasis. Previous studies have shown that [29], the combination of rh-Endo in NACT can signifcantly prolong progression-free time, inhibit disease progression, and reduce recurrence. Te present research aims to explore the efcacy of rh-Endo plus NACT for OSA and its infuence on serum VEGF and MMP-9 levels. Te results revealed a statistically higher ORR in CMG than in CNG (84.2% vs. 64.6%). Moreover, serum tumor markers BALP, IGF-1, and SAA were reduced in both cohorts after two weeks of chemotherapy drug withdrawal, with lower levels in CMG. Te detection of tumor markers is a routine means to refect the tumor load of malignant lesions. BALP is a marker for OSA cell growth, which is upregulated in more than 40% of OSA patients [30]. IGF-1 participates in OSA cell division and multiplication, and its high expression can promote the distant metastasis of OSA cells [31]. As to SAA, it is one of the markers that distinguish benign from malignant OSA, and is upregulated in OSA tissues and released into the blood [32]. After NACT plus rh-Endo intervention, these indexes were signifcantly reduced, with a reduction degree better than that of NACT alone. In addition, serum VEGF and MMP-9 were notably  8 Journal of Oncology lower in CMG than in CNG two weeks after stopping chemotherapy. VEGF is a key factor in tumor angiogenesis, which mainly stimulates the proliferation and diferentiation of vascular endothelial cells by autocrine or paracrine, and participates in every link of tumor cells in the form of abnormally high expression in many organ tumors of the human body [33]. MMP-9 can increase tumor cell invasion and metastasis and promote vascular endothelial cell migration to form new blood vessels through the degradation of the extracellular matrix and basement membrane [34]. Te results of this study, together with many previous studies, prove that VEGF and MMP-9 are highly expressed in OSA patients, and rh-Endo can efectively reduce their expression. Furthermore, we found that two weeks after drug withdrawal, serum CRP and TNF-α levels decreased and IL-10 levels increased, and the relief of infammation was more signifcant in CMG. CRP and TNF-α are both vital pro-infammatory factors in the human body, and their elevated expression is not only related to OSA-induced infammatory responses but also to the invasive proliferation, infltration, and injury induced by tumor cells [35,36], while IL-10 is an essential anti-infammatory factor. Under normal conditions, the levels of anti-infammatory and pro-infammatory factors will form a dynamic balance [37]. However, IL-10 will be inhibited when there is a malignancy, and will be gradually increased after treatment. By comparing the incidence of adverse reactions produced by the two treatment regimens, we found no signifcant difference between the two cohorts. Moreover, after 6 months and one year of follow-up, patients' limb function and QOL were better in CMG, suggesting that rh-Endo can improve the limb function and QOL of patients. However, this research still shows some shortcomings. First, due to the nature of being a single-center retrospective analysis, a multicenter prospective clinical trial is needed for confrmation. Second, a randomized double-blind clinical trial is required to validate the fndings as the chemotherapy regimen in this study is not randomly selected. Tird, the sample size is relatively small, so it is necessary to further assess the efectiveness and safety of rh-Endo plus NACT in OSA. We will increase patient samples covering diferent stages or pathological types, and extend the follow-up time to further evaluate the efectiveness of this combination therapy for OSA.

Conclusion
To sum up, rh-Endo plus NACT is efective for OSA, which can help restore the balance of infammatory factor expression in patients, inhibit tumor angiogenesis to reduce tumour invasion and metastasis, and improve limb function and QOL of patients, providing a feasible treatment plan for follow-up treatment.

Data Availability
Te labeled dataset used to support the fndings of this study are available from the corresponding author upon request.

Conflicts of Interest
Te authors declare that they have no conficts of interest. Journal of Oncology 9