The incidence of osteosarcoma in this study was highest in the age range of 10–19 years and lowest above 30 years. This pattern aligns with another study that noted most osteosarcoma cases typically occur in children aged 12–16 years [19] and aligns with the notion that most osteosarcoma cases develop between the ages of 14–18 years old, corresponding to the acceleration of pubertal growth. The annual incidence rate is 4.4 cases/1 million population for people aged 0–24 years, with males affected more frequently and the M:F ratio 1.3: 1 [21].

Osteosarcoma is highly malignant and shows strong invasive capabilities, progressive disease development, and a very high mortality rate [22]. Before the discovery of chemotherapy in the 1970s, the prognosis for osteosarcoma patients was deemed very poor, with a survival rate of < 20%. The management involving surgical resection with adequate surgical margins combined with neoadjuvant chemotherapy has increased the survival rate to 60–70% and has reached a plateau. The survival rate for patients with localized osteosarcoma is 65%, while for patients with metastatic osteosarcoma, it is < 20% [23, 24]. In selective cases, patients can be rescued by metastasectomy [25].

Recent approaches using multimodal strategies, including systemic chemotherapy before surgery (neoadjuvant chemotherapy), followed by local surgical procedures and supplemented with postoperative chemotherapy (adjuvant chemotherapy), have improved long-term success rates to 60–70%, enhancing overall survival [26]. However, current therapeutic modalities have various disadvantages, such as the lack of a significant improvement in survival rates despite decades of chemotherapy use, and the side effects associated with high-dose chemotherapy. Given the advancements in scientific knowledge, there is a fundamental need to identify new biological indicators for patient prognosis and chemotherapy response detection. The aim is also to develop innovative and specific therapeutic approaches targeting specific molecular targets (targeted therapy) to improve outcomes for osteosarcoma patients with a poor prognosis [27].

Only four chemotherapy agents have shown objective responses in osteosarcoma: Doxorubicin at 43%, Ifosfamide at 33%, Methotrexate at 32%, and Cisplatin at 26%. Studies utilizing three active agents have demonstrated better outcomes than those using only two regimens [28]. The first-line chemotherapy includes Cisplatin and Doxorubicin (category 1), MAP (high-dose Methotrexate, Cisplatin, and Doxorubicin), and the combination of Doxorubicin, Cisplatin, Ifosfamide, and high-dose Methotrexate (MAPI) [5]. In this study, patients received neoadjuvant chemotherapy following the National Comprehensive Cancer Network (NCCN) guidelines, consisting of Cisplatin and Doxorubicin only. The mean 5-year event-free survival (EFS) and 5-year overall survival (OAS) are 48% and 62% for two-drug regimens, and 58% and 70% respectively for three or more drug regimens. Patients treated with chemotherapy have lower 5y-EFS compared to those who are treated with surgery and chemotherapy, showing that patients with localised high-grade osteosarcoma cannot be treated by chemotherapy alone [29]. Intraarterial cisplatin is more effective than methotrexate. Multidrug regimen without methotrexate achieves similar survival rates to methotrexate-based regimens. The meta-analysis on EFS shows significantly improved survival by administering three drug-regimen compared to a two-drug regimen but shows no significant difference with a 4 drug-regimen [29].

The long-term success of traditional chemotherapy and targeted anticancer agents often relies on immunological effects. Immunogenicity results from a combination of antigenicity and adjuvanticity, and many anticancer drugs activate adaptive stress responses in malignant cells, serving as immunological adjuvants [30]. In some malignancies, both innate and adaptive immune cells play a role in the tumour microenvironment, involving natural killer (NK) cells, antigen-presenting cells (APC) like macrophages and dendritic cells (DC), and lymphocytes, effectively controlling tumours [31].

Cell death plays a pivotal role in cancer, encompassing both programmed cell death and non-programmed cell death. Programmed cell death involves apoptosis, autophagy, and necroptosis (programmed necrosis) [32]. The apoptosis process unfolds through two pathways: extrinsic (receptor-mediated apoptosis) and intrinsic (mitochondria-mediated apoptosis) [33]. The extrinsic pathway is initiated by the binding of death receptors, such as CD95R (and analogous receptors like tumour necrosis factor (TNF) receptor 1 and its counterparts), on the plasma membrane with its extracellular ligand, CD95L [32]. Upon stimulation, CD95L binds to CD95R, forming a death signal complex. Following its formation, the death-inducing signalling complex (DISC) initiates, recruiting various proapoptotic factors, including caspase-8. Subsequent cleavage of caspase-8 and effector caspase-3 leads to proteolytic protein breakdown and apoptosis [34,35,36].

In this study, a significant difference was found between poor and good responses, with the highest CD95R expression in poor responses, suggesting that higher CD95R expression corresponds to increased chemotherapy resistance. This contradicts previous studies indicating that low CD95R expression is associated with a worse prognosis in osteosarcoma patients [37]. CD95R expression is inversely related to tumour metastasis potential, with increased CD95R expression correlated with decreased tumour cell metastasis potential [38]. This is supported by studies showing prolonged survival trends in patients with CD95R expression [39].

CD95R, expressed on the tumour cell surface, contributes to immune cell-mediated cytotoxicity. Patients with a poor chemotherapy response have low CD95R levels, making it easier for tumour cells to metastasise. Those with high CD95R levels have the potential for longer survival and a lower probability of death [39]. Another study investigated the effects of Doxorubicin and Edelfosine lipid nanoparticles on osteosarcoma. The study revealed that this chemotherapy combination increased CD95R expression on tumour cell surfaces, triggering apoptosis [40].

CD95R is associated with the TNF receptor family and mediates apoptosis when bound to its natural ligand, CD95L (CD178/TNFSF6), or stimulated with agonistic antibodies [35]. CD95L is predominantly expressed in activated T lymphocytes and NK cells, the primary subset of innate immune cells responsible for antiviral and antitumor responses [41, 42]. NK cells, defined as CD56+/CD3.− lymphocytes, exhibit cytotoxicity, and cytokine production, particularly in response to exogenous cytokines like IL-2, IL-12, IL-15, IL-18, and IL-21 [43,44,45,46]. Among the most prominent cytokines produced by NK cells are TNF-α and IFN-γ [47].

IFN-γ, a crucial cytokine, enhances tumour cell death mediated by CD95R, playing a vital role in immunomodulation and antitumor activity [48]. IFN-γ is an immunomodulatory cytokine that promotes apoptosis directly or through induced lipid peroxidation and ferroptosis, contributing to an effective antitumor immune response [49]. This is supported by a study that evaluated the influence of IFN-γ on ligand expression in various NK cells on cancer cells, with none of these ligands being downregulated by IFN-γ. However, CD274/PD-L1, CD54/ICAM-1, HLA-DR, MHC class I, CD95/FasR, and CD270/HVEM are regulated in various tumour types [34]. More specifically, in osteosarcoma, a study mentioned that IFN-γ sensitizes osteosarcoma cells to CD95R, inducing apoptosis through the upregulation of CD95R and caspase-8 [50, 51].

This study revealed significant differences in IFN-γ expression among chemotherapy response groups. This suggests that IFN-γ might play a key role in the effectiveness of chemotherapy, with lower levels associated with increased resistance. The results are in line with a previous study that suggested a correlation between elevated PSMD14 expression and a higher-risk category (younger age group, ≤ 20.83 years of age), metastasis within five years, and a higher tumour grade in osteosarcoma patients [37]. PSMD14 is a component of the 26S proteasome. The group with increased PSMD14 expression exhibited decreased immune responses, including reduced levels of IFN-γ [52].

A previous study emphasized the significant contribution of IFN-γ to antitumor and antimetastatic effects, triggering FAS ligand formation and apoptosis induction in cancer cells [50]. Interestingly, previous studies highlighted the positive impact of chemotherapy on IFN-γ levels, demonstrating its correlation with good treatment response in osteosarcoma [53]. Additionally, a previous study underscores the potential preventive role of NK cells in osteosarcoma development, as evidenced by higher circulating NK cell counts in normal controls compared to osteosarcoma patients [54, 55]. Natural Killer (NK) cells release IFN-γ as one of the most potent effector cytokines [47].

Our study suggests that in a good response, the immune system remains resilient against chemotherapy, marked by an increase in IFN-γ. The phenomenon of decreasing CD95R expression from poor to good responses, accompanied by an increase in IFN-γ, suggests a declining number of viable osteosarcoma cells as the Huvos grading advances.

Excessive reactive oxygen species (ROS) leading to oxidative stress plays a pivotal role in carcinogenesis [56]. Cells maintain intracellular homeostasis by developing an extensive antioxidant system, including catalase, superoxide dismutase (SOD), and glutathione peroxidase (GPx). Catalase facilitates the decomposition of hydrogen peroxide into water and oxygen (2H2O2 → 2H2O + O2), crucial for degrading H2O2 and preserving cells against oxidative damage. The expression of catalase varies in tumour cells [57]. Low catalase expression correlates with high H2O2 production, influencing signalling pathway activation to induce proliferation, migration, and invasion in cancer cells [58,59,60]. Changes in catalase expression after short-term H2O2 exposure are influenced by factors such as exposure time, H2O2 concentration, basal antioxidant enzyme capacity, and the cellular model used [57]. High catalase expression has been observed in certain human cancer cell lines, including gastric cancer exposed to cisplatin chemotherapy [61].

No significant difference in catalase expression was observed among the chemotherapy response groups, implying that catalase may not play a decisive role in chemotherapy outcomes. Similarly, a previous study conducted on male patients under 20 years old with osteosarcoma did not find a significant association between catalase level and chemotherapy response [62]. However, its consistently elevated expression in poor responders raises the need for further investigation. Catalase expression did not exhibit the capacity to protect against ROS attacks induced by chemotherapy.

The role of Hsp in maintaining cellular homeostasis and protection is vital, with their heightened production occurring under stressful conditions [63]. These proteins play a crucial role in safeguarding cells from stress-related injuries and are often overexpressed in various malignancies [64]. Among them, Hsp70, an ATP-dependent molecular chaperone, regulates protein conformation, stability, and interactions, including essential proteins for cell survival, growth, and immune responses [65, 66]. Excessive Hsp70 expression in cancer cells is implicated in various oncogenic events, such as anti-apoptotic responses, antitumor immune responses, tumour growth, and cell migration [67].

The poor response group exhibited the highest Hsp70 expression. Previous studies on Hsp in osteosarcoma revealed prognostic associations, with Hsp27, Hsp60, and Hsp70 linked to poor prognosis [68]. Additionally, Hsp27 and Hsp70 were identified as potential markers to distinguish conventional and low-grade central osteosarcoma [64]. Lower Hsp70 expression correlated with higher tumour cell necrosis rate (TCNR), indicating its utility in predicting and evaluating the neoadjuvant chemotherapy's effectiveness in inhibiting osteosarcoma cell proliferation [69]. These findings contribute valuable insights into the complex interplay of Hsp70 in osteosarcoma and its response to chemotherapy, providing avenues for further study and potential therapeutic interventions.

Angiogenesis is a crucial aspect of tumorigenesis, influencing tumour growth and metastatic potential [70]. Vascular endothelial growth factor (VEGF), a key player in angiogenesis and vasculogenesis, primarily acts on various cell types, particularly endothelial cells [71]. VEGF is considered a major mediator of angiogenesis and has been implicated in carcinogenesis and metastasis [72]. Tumour malignancy requires a persistent supply of new blood vessels to sustain growth and facilitate metastasis [73]. In osteosarcoma, VEGF-A expression, as demonstrated by immunohistochemistry, has been associated with a higher risk of lung metastasis and poorer survival [74].

In our study, VEGF expression exhibited distinct patterns, notably higher in good response cases compared to both moderate and poor responses. This variability may be attributed to significant cell death in osteosarcoma, hindering the formation of new blood vessels. Despite numerous studies exploring the correlation between VEGF overexpression and clinical outcomes in osteosarcoma patients, the results remain inconclusive. Meta-analyses assessing the connection between VEGF expression and overall survival in osteosarcoma patients have consistently indicated that elevated VEGF expression is associated with poorer overall survival, with no significant heterogeneity among studies [72]. The positive expression of VEGF in surviving tumour cells following neoadjuvant chemotherapy emerges as a prognostic factor in osteosarcoma, offering valuable insights for identifying potential angiogenic targets in the pursuit of personalized future therapies [75].

We studied the expression on resected post-chemo specimens. Osteosarcoma is an extremely heterogeneous tumour and diagnostic (core-needle) biopsies are only small specimens, which are taken for diagnosis and do not necessarily reflect the expression of molecules throughout the whole tumour. Our study focuses on the response to chemotherapy correlated with immunological response. Including paired biopsy specimens would not result in reliable data given the tumour heterogeneity, so the results reflect the expression profile of the surviving cells post-chemotherapy. Pre-chemo expression in the tumours and changes over time during chemotherapy treatment are not assessed in this study.

The novelty of this study lies in the conventional chemotherapy's mode of action, typically involving DNA synthesis inhibition through folate metabolism disruption (methyl tetrahydrofolate). However, it appears that chemotherapy can also influence the immune response, as evidenced by the findings of increased IFN-γ expression and decreased CD95R expression in a good response to chemotherapy.

Considering our findings, suggestions for further study and clinical implications emerge. There is a need for further investigations to delve into the specific functions of lymphocytes in the pathogenesis of osteosarcoma malignancy. Understanding the intricate interplay between the immune system and osteosarcoma progression can pave the way for targeted therapeutic interventions and personalized treatment strategies. Expanding our knowledge in these areas will undoubtedly contribute to refining prognostic assessments and advancing the development of more effective therapeutic approaches for osteosarcoma patients.