Prognostic factors of oral squamous cell carcinoma: a five-year retrospective study in Cluj-Napoca, Romania

Abstract

Background and Objective: The survival of patients with oral squamous cell carcinoma (OSCC) is influenced not only by tumor characteristics but also by treatment strategies, follow-up, and patient-related factors. This study aimed to determine the overall five-year survival rate in a Romanian cohort and evaluate the impact of key prognostic factors. Patients, Materials and Methods: Seventy-two patients diagnosed with OSCC and treated surgically at the Department of Oral and Maxillofacial Surgery, Emergency County Clinical Hospital, Cluj-Napoca, Romania, between January 2016 and December 2019 were retrospectively analyzed. Prognostic factors included tumor stage, primary site, histological parameters, lymph node involvement, treatment modalities, and demographic variables. Results: Patients were stratified into four groups: five-year survivors (n=26), disease-specific mortality (n=21), mortality from non-oncological causes (n=5), and lost to follow-up (n=20). Overall, 36.1% of patients achieved five-year recurrence-free survival, corresponding to 50% among those with complete follow-up. Survivors were more often diagnosed at early or intermediate stages, had fewer tongue tumors, better differentiation, clear surgical margins, and consistent adjuvant therapy. Notably, 27.8% of patients were lost to follow-up despite favorable pathological features, suggesting systemic gaps in long-term surveillance. Conclusions: The five-year survival rate of 50% reflects both aggressive tumor biology and systemic barriers to care. Clinical stage remains the strongest prognostic factor, but tumor site and histological features, including perineural and lymphatic invasion, significantly influence outcomes even in early disease. Improved long-term survival requires earlier diagnosis, refined prognostic assessment, and multimodal therapy tailored to tumor behavior, alongside better patient follow-up strategies.

Introduction

Following a peak in cancer-related deaths across Europe during the late 1980s, a decline of approximately 10% was observed in both men and women by the early 2000s [1, 2]. Oral squamous cell carcinoma (OSCC) ranks among the most prevalent cancers globally, with notable regional, gender, and age-related disparities across Europe that significantly affect its incidence and outcomes [3]. In 2020 alone, there were 377 713 new OSCC cases and 117 757 associated deaths recorded worldwide [4, 5]. Projections from the Global Cancer Observatory (GCO) estimate a nearly 40% increase in OSCC incidence by 2040, along with a corresponding rise in mortality [4, 6]. While early-stage OSCC shows a five-year survival rate exceeding 80%, this figure drops less than 30% in advanced stages. Alarmingly, over 60% of OSCC diagnoses occur at late stages, where both morbidity and mortality rates are significantly elevated [5, 7, 8].

Aim

This disease is currently one of the most extensively researched pathologies in the medical field. From in vivo investigations on living subjects to in vitro experiments conducted under controlled laboratory conditions, the clinical management of oral and maxillofacial cancers involves a comprehensive approach encompassing diagnosis, therapeutic intervention, and long-term care. This clinical study focuses on the final stage of this continuum, utilizing critical insights derived from earlier phases to identify prognostic factors relevant to patient outcomes. Conducted as a retrospective, observational, and analytical investigation at the Emergency County Clinical Hospital, Cluj-Napoca, Romania, the study focuses on patients diagnosed with OSCC. Its primary objective is to identify local factors that significantly influence postoperative prognosis. By rigorously examining variables that directly or indirectly affect outcomes, following tumor excision, the study aims to improve both survival rates and quality of life. Furthermore, the findings are intended to inform and refine therapeutic strategies, thereby addressing the elevated mortality associated with oral cancer.

Patients, Materials and Methods

To carry out this clinical investigation on prognostic factors in OSCC, data were retrospectively collected from the AtlasMed information system of the Emergency County Clinical Hospital, Cluj-Napoca, supplemented by a thorough review of patients’ observation charts. Ethical approval for the study was granted by the Ethics Committee of Iuliu Haţieganu University of Medicine and Pharmacy, Cluj-Napoca, and the research was limited to 75 patients.

Inclusion criteria

Patients hospitalized in the Department of Oral and Maxillofacial Surgery, Emergency County Clinical Hospital, Cluj-Napoca, between 01.01.2016–31.12.2019 diagnosed with OSCC, who presented with at least one detectable tumor at the time of initial clinical assessment.

Exclusion criteria

Patients hospitalized in the Department of Oral and Maxillofacial Surgery outside the mentioned period, patients who presented premalignant lesions, cysts or benign tumors of the mouth, rather than histopathological (HP) confirmed OSCC and patients having a tumor outside of the oral cavity, as a metastatic disease or a second primary tumor in other parts of the body. Of the initial cohort of 75 patients included in the study, three individuals were excluded due to preoperative diagnoses other than OSCC, namely: mixed nodular melanoma, cystic adenoid carcinoma, and adenosquamous carcinoma.

Data collection

Data was collected comprehensively and organized in the form of standardized tables. The relevant information extracted from the information system included: observation chart number, tumor variables {primary tumor site, HP pattern, stage of the disease, postoperative tumor–node–metastasis (TNM) diagnosis, degree of differentiation (G), tumor invasion types [lymphatic (L), vascular (V) and perineural invasion (PNI)], depth of invasion (DOI) and worst pattern of invasion (WPOI)}, treatment modalities (type surgical intervention for the primary tumor and/or cervical lymph node dissection), adjuvant pre and/or post therapies such as radiotherapy and/or chemotherapy, and margin status), disease progression patterns (recurrence and metastasis), demographic variables such as gender and place of residence (rural/urban) and additional relevant information discovered during post-intervention monitoring such as death cause.

Data synthesis and analysis

The primary goal of this study was to establish factors determining a favorable prognosis in selected patients. Patients were stratified into multiple categories based on selected clinical and pathological variables, and the resulting data were compared with findings previously reported in scientific literature.

Results

The analytical results of the study are summarized in Table 1. As the data are presented in aggregate form rather than on an individual patient basis, numerical values reflect overall cohort characteristics. Although it was initially intended to include additional prognostic parameters, such as DOI, WPOI, tumor budding (TB), and extracapsular nodal extension (ENE), these were not consistently available across all cases and were therefore excluded from the final table.

Table 1.

Comparative analytical data

	OSCC overall cohort	Group 1	Group 2	Group 3	Group 4
5-Year survivors	5-Year DSM	5-Year mortality, NOC	Patients LFU
No. of patients	72	26	21	5	20
Gender
▪ M:F ratio	56:16	20:6	15:6	4:1	17:3
Age (average) [years]	59.98	58	60.14	73.8	58.84
▪ <50	9	4	2	0	3
▪ 50–65	40	15	11	0	14
▪ >65	23	7	8	5	3
Residence
▪ Urban	51	18	14	4	15
▪ Rural	21	8	7	1	5
Primary tumor site
▪ Mouth floor	31	12	5	2	12
▪ Tongue	16	4	8	1	3
▪ Alveolar crest	17	6	6	1	4
▪ Buccal mucosa	2	1	0	1	0
▪ Retromolar area	3	2	1	0	0
▪ Intermaxillary commissure	2	1	1	0	0
▪ Other	1	0	0	0	1
Stage
▪ I	18	5	4	1	7
▪ II	10	5	2	0	3
▪ III	12	6	3	1	2
▪ IVa	33	10	12	3	8
Grading
▪ G1	14	7	2	2	3
▪ G2	46	14	17	2	13
▪ G3	12	5	2	1	4
L
▪ L0	58	22	14	4	18
▪ L1	14	4	7	1	2
V
▪ V0	64	23	18	4	19
▪ V1	8	3	3	1	1
PNI
▪ PNI0	42	13	11	2	12
▪ PNI1	30	11	10	3	8
R
▪ R0	63	24	15	4	20
▪ R1	9	2	6	1	0
Radio/chemotherapy	n.a.	22	15	2	n.a.
Recurrences	n.a.	0	8	0	n.a.
Metastasis	n.a.	2	2	0	n.a.

DSM: Disease-specific mortality; F: Female; G: Degree of differentiation; L: Lymphatic invasion; LFU: Lost to follow-up; M: Male; n.a: Not available; NOC: Non-oncological causes; OSCC: Oral squamous cell carcinoma; PNI: Perineural invasion; R: Resection margin; V: Vascular invasion

Overall cohort profile

The study cohort comprised 72 patients diagnosed with OSCC. The mean age of patients in this cohort was 59.98 years, which appears lower than the average reported in comparable international studies [3]. The analysis of gender distribution indicated a clear male dominance, with 56 male and 16 female patients, yielding a male-to-female ratio of 3.5 to 1. Regarding residential background, 70.8% were from urban areas, while 29.2% resided in rural regions. The patients were grouped based on age brackets: less than 50 years, between 50 and 65 years, and over 65 years. The primary tumor site varied among patients. The most common locations were the floor of the mouth and the tongue (65.3%), followed by the alveolar crest (23.6%). Less frequent sites included the retromolar area (n=3), buccal mucosa (n=2), intermaxillary commissure (n=2), and other regions (n=1). The tumor stage, based on the TNM classification, showed that most patients presented with advanced diseases. Specifically, 45.8% were diagnosed at stage IVa, 16.7% in stage III, while 25% and 13.9% were in stages I, and II, respectively. The grading distribution suggests that most tumors (62.5%) were moderately differentiated (G2), highlighting a potential intermediate prognosis for the majority of the cohort. The relatively balanced proportions of G1 (19.4%) and G3 (18.1%) reflect the HP diversity of OSCC within this population. Evaluation of lymphatic invasion (L status) revealed L1 in 19.4%. Vascular invasion (V status) was present in 11.1%. Regarding PNI, 41.7% had confirmed perineural invasion.

For analytical purposes, the total cohort of 72 patients was stratified into four distinct groups based on five-year outcomes and follow-up status, using survival data and medical record continuity:

• Group 1: five-year survivors (patients who remained alive and disease-free for at least five years post-treatment, with no documented recurrence or secondary malignancy);

• Group 2: five-year disease-specific mortality (patients who died within five years due to progressive or recurrent OSCC, as verified by medical records and/or cause of death documentation);

• Group 3: five-year mortality – non-oncological causes [patients who died within five years of treatment due to causes unrelated to OSCC (e.g., cardiovascular events, infections, or comorbid conditions)];

• Group 4: patients lost to follow-up (patients for whom no outcome data were available beyond initial treatment (e.g., unreachable, non-compliant, or lacking follow-up documentation).

This stratified approach enables a clearer, more precise interpretation of survival outcomes by distinguishing cause-specific mortality, isolating follow-up losses to reduce attrition bias, and allowing direct comparison of survivors and deceased based on clinical and pathological factors. It mirrors established methodologies in OSCC research [9, 10], thereby strengthening the validity and clinical relevance of the study’s conclusions.

Group 1 profile (five-year survivors; n=26)

Among the 72 patients treated for OSCC, 26 (36.1%) individuals survived at least five years post-treatment without recurrence or secondary malignancy. Based on the cohort of patients with available follow-up data, the observed survival rate was 50% (26 of 52 patients). Survivors had a slightly lower average age (58.0±9.2 years), and the majority were between 50 and 65 years old, similar to the overall cohort distribution. The long-term survivors were predominantly male (77%), continuing the well-documented gender disparity in OSCC incidence. The majority of survivors (69%) lived in urban areas, potentially benefiting from earlier diagnosis, proximity to specialized care, or better treatment compliance. The floor of the mouth was the most common tumor site among survivors (46%). Tumors of the tongue, typically considered more aggressive due to rich lymphatic drainage, accounted for only 15%. Survivors were evenly distributed across the clinical spectrum: 19% were diagnosed at stage I, 19% at stage I, 23% at stage III, and notably, 38% were diagnosed at stage IVa.

This staging profile challenges the assumption that only early-stage OSCC ensures long-term survival. Instead, it illustrates that even advanced-stage patients can achieve five-year survival, likely due to aggressive treatment strategies, clear surgical margins, and effective adjuvant therapy. Histologically, more than half of the survivors had moderately differentiated (G2) tumors (n=14), while well-differentiated tumors (G1) accounted for seven cases, and poorly differentiated tumors (G3) for five cases. This distribution highlights that histological grade, though important, is not singularly predictive of prognosis. It also underscores the fact that G3 tumors are not universally fatal when managed with a comprehensive treatment plan. Despite being long-term survivors, many patients presented traditionally high-risk features: perineural invasion (PNI1) was observed in 42%, lymphatic invasion (L1) in 15%, vascular invasion (V1) in 12% of cases. These findings demonstrate that the presence of adverse histological features does not preclude long-term survival, particularly when addressed with timely and appropriate adjuvant treatment. Perhaps the most consistent finding among survivors was complete tumor resection. Twenty-four of the 26 (92%) patients had R0 margins, confirming no microscopic residual tumor after surgery. Only two patients had R1 resections, yet still achieved five-year survival, likely due to effective adjuvant radiotherapy. A striking 22 (85%) patients received radiotherapy and/or chemotherapy, either in a definitive or adjuvant setting. This emphasizes the role of multimodal therapy in managing OSCC and improving survival, especially in patients with high-grade tumors, lymphovascular invasion, or stage III/IVa disease.

Group 2 profile (five-year disease-specific mortality; n=21)

The profile of patients who died within five years (29.2% of the total cohort and 40.38% of the patients with available follow-up data, respectively) reflects a clinically and pathologically complex disease course. The majority were male (71.4%) and predominantly from urban areas (66.7%), indicating that gender and residence did not mitigate risk. A distinct pattern in tumor localization emerged, with the tongue being the most frequently affected site (38.1%). This region, known for early lymphatic spread and difficult surgical access, likely contributed to worse outcomes. A significant prognostic determinant was disease stage at diagnosis: 57.1% of patients were diagnosed at stage IVa, while only 19% were stage I, 9.5% stage II, and 14.3% stage III. This distribution highlights the impact of delayed presentation or inherently aggressive tumor biology. Most tumors (81%) were moderately differentiated (G2), which, while typically suggesting intermediate behavior, were frequently associated with high-risk features in this subgroup.

Markers of tumor invasiveness were prevalent: L1 occurred in 33.3%, V1 in 14.3%, and PNI1 in 47.6% of patients. The high frequency of PNI strongly correlates with local recurrence and adverse outcomes. Despite theoretical access to specialized care, 28.6% of patients had positive resection margins (R1), indicating incomplete tumor excision. Even among those with clear margins (R0: 71.4%), survival was not guaranteed in the presence of other high-risk features. Notably, only 71% of these patients received radio/chemotherapy, suggesting potential underutilization of adjuvant treatments in cases with aggressive pathology. This subgroup highlights the crucial role of early diagnosis, meticulous surgical management, and comprehensive adjuvant therapy in improving OSCC survival outcomes.

In summary, while both groups 1 and 2 shared high-risk tumor features, survivors were more likely to benefit from early to intermediate-stage diagnosis, less frequent tongue involvement, better tumor differentiation, clear surgical margins, and consistent use of adjuvant therapy.

The deceased group, despite similar access to care, had more aggressive tumor sites, a higher burden of stage IVa, positive surgical margins, and lymphatic spread, all of which strongly influenced outcomes.

This comparison supports the notion that long-term survival in OSCC is achievable even in advanced cases, provided that complete tumor resection (R0) and effective adjuvant therapy are delivered.

Group 3 profile (five-year mortality – non-oncological causes; n=5)

This subgroup demonstrates that successful treatment of OSCC is achievable even in advanced-stage cases, as evidenced by the absence of recurrence or metastasis. Deaths in this group likely resulted from non-cancer-related comorbidities, such as cardiovascular or metabolic conditions – common in this patient demographic. Patients who died from non-oncological causes had the highest mean age (73.8±2.4 years), suggesting that advanced age may have contributed to mortality unrelated to cancer. Despite the presence of PNI in 60% and stage IVa disease in 60%, all patients showed stable oncological outcomes, reaffirming the value of effective initial treatment and surgical margin clearance (R0 in 80%). The lower rate of adjuvant therapy (40%) further suggests that oncological control was achieved predominantly through surgery. These cases emphasize the importance of differentiating OSCC-related mortality from unrelated causes in outcome analyses, as this subgroup may otherwise skew survival statistics and understate treatment efficacy.

Group 4 profile (patients lost to follow-up; n=20)

Out of the 72 patients included in the study, 20 (27.8%) were lost to follow-up, meaning no survival or recurrence data were available beyond their initial treatment. This subgroup displayed a demographic and clinical profile largely consistent with the overall cohort, yet their absence from long-term analysis introduces important considerations.

Notably, 50% were early-stage (I–II), potentially indicating missed opportunities for successful monitoring and long-term outcome documentation. All 20 (100%) patients had R0 resections, indicating that complete tumor excision was achieved during surgery, and initial treatment was likely curative in intent. Invasion markers were relatively infrequent overall, but the presence of PNI in 40% of cases suggests that some patients might have been at elevated risk for recurrence or progression, further emphasizing the clinical impact of follow-up attrition. The loss of 27.8% of patients to follow-up, despite their largely urban background and favorable pathological characteristics (low L1, V1, and complete resections), may reflect systemic gaps in post-treatment surveillance or patient disengagement rather than clinical severity or access limitations. This represents a potential source of selection bias in outcome analysis and underscores the importance of strengthening survivorship care planning, patient tracking systems, and follow-up adherence – especially in high-risk cancers like OSCC where recurrence may occur even after initial curative treatment. Further investigation into socioeconomic, psychological, or institutional factors contributing to follow-up attrition is warranted to improve long-term data quality and patient care continuity.

Discussions

Despite advances in treatment, OSCC persists as a highly aggressive and globally prevalent cancer, with an estimated five-year survival of around 50%, falling less than 30% in later stages of the disease [11, 12]. Recent studies report survival rates ranging between 56% and 71% in certain institutions, although most centers continue to observe rates less than 60% [13, 14, 15, 16]. In our study, the observed five-year survival rate was 50%, reflecting both the biological aggressiveness of the disease and region-specific challenges in Romania, such as delayed presentation, rural healthcare disparities, and high prevalence of modifiable risk factors.

The most common tumor site among survivors was the floor of the mouth (46%), while tongue tumors predominated in the deceased group (38%). Tongue carcinomas are recognized for their aggressive clinical course due to rich lymphatic drainage and early nodal spread [17]. Conversely, tumors in the floor of the mouth may be more amenable to early surgical resection, possibly contributing to improved outcomes. Despite this, survival even in advanced tongue cancers was observed, emphasizing the importance of comprehensive treatment rather than tumor site alone.

As expected, survival declined with increasing disease stage. In our cohort, stage IVa disease was more prevalent among deceased patients (57%) compared to survivors (38%). Nonetheless, a substantial number of IVa stage patients reached the five-year survival mark, demonstrating the impact of effective surgical clearance (R0) and adjuvant radio/chemotherapy. This supports findings from Ajmani et al., who reported improved outcomes even in advanced OSCC when multimodal treatment is applied early and aggressively [18].

Histological grades were distributed similarly between groups, although well-differentiated (G1) tumors were more prevalent among survivors. While moderately differentiated (G2) tumors accounted for the majority in both groups, this grade is increasingly recognized as a prognostically heterogeneous category [12]. Importantly, five long-term survivors also had G3 tumors, reaffirming that histological grade should be interpreted alongside other indicators such as PNI, margin status, and DOI [19, 20].

Among survivors, 92% had R0 resections, compared to only 71% among deceased patients. Margin status remains one of the strongest predictors of survival, as incomplete tumor resection leads to recurrence and reduced long-term control [21]. Additionally, L1 and PNI1 were more prevalent in the deceased group. However, 42% of survivors also presented with PNI1, suggesting that adjuvant therapy can mitigate the negative effects of certain high-risk histological features [18, 22, 23].

The impact of tumor invasion patterns was further illustrated through detailed comparisons of stage I patients with identical histological and treatment parameters but differing lymphatic invasion status. One patient with L1 progressed rapidly to metastasis, dying within two years, while the other (L0) experienced a delayed course, ultimately succumbing five years later. This underscores how invasion patterns, particularly L1, can dramatically alter prognosis, even in early-stage disease.

Unlike Southeast Asian regions where betel quid is a major carcinogen, in Romania, tobacco smoking and alcohol consumption remain the dominant etiological factors. These disproportionately affect male patients, reflected in the 78% male predominance in our cohort, consistent with global epidemiological patterns [24, 25, 26]. Additional contributors to poor prognosis include comorbidities (e.g., diabetes, cardiovascular disease, chronic pulmonary disorders) and low socioeconomic status [27].

One of the most striking observations is that 67.3% of patients are presented with stage III or IV disease, a statistic reflective of limited public awareness, absence of national screening programs, and disparities in healthcare access, especially in rural areas. Although 71% of our patients reside in urban areas, this is likely to be due to centralization of care in referral centers. Delays in presentation from rural regions continue to drive advanced-stage diagnoses. Public health strategies aimed at early detection, routine oral examinations, and cessation of tobacco and alcohol use are urgently needed [28, 29]. The stage at which the patient is presented to the clinic remains one of the most critical predictors of survival [30]. Early-stage OSCC (I and II) is more amenable to successful surgical resection with negative margins and lower risk of recurrence [5, 10]. Advanced stage (III and IV) frequently involves lymphatic spread, vascular invasion or distant metastases, complicating treatment and reducing survival [5, 31]. The likelihood of a successful intervention and long-term survival is greatly decreased for patients who put off seeking medical help because they frequently arrive with advanced diseases.

Conclusions

This study highlights the complex interplay between tumor characteristics, patient-specific risk factors, and healthcare infrastructure in shaping survival outcomes for OSCC in Romania. The five-year survival rate of 50% is reflective of both aggressive tumor biology and systemic barriers to early diagnosis and care. While clinical stage at diagnosis remains the strongest prognostic indicator, our findings show that histological factors such as PNI and L invasion significantly influence outcomes, even in early-stage disease. To improve long-term survival, there must be a dual focus on enhancing early detection through public health education and screening and on tailoring treatment to tumor behavior, including aggressive multimodal therapy when indicated. Efforts to bridge urban-rural disparities, increase awareness, and improve access to expert surgical and adjuvant care will be critical in shifting the prognosis for OSCC in Romania and similar regions.

Conflict of interests

The authors declare no conflict of interests.

Source of funding

This research received no external funding.