The Host-index in patients undergoing upfront surgery and free flap reconstruction for head and neck squamous cell carcinoma

Cover figure: Blood parameters used for the calculation of the Host-index and Kaplan-Meier curves of disease-free survival in head and neck squamous cell carcinoma patients stratified ac-cording to it (< 3.6 vs ≥ 3.6; log-rank p = 0.001).

Objective. The Host-index (H-index) is a value obtained using blood laboratory parameters. Elevated H-index is a negative prognostic factor in patients with head and neck squamous cell carcinoma (HNSCC). The purpose of our study was to assess the prognostic impact of the H-Index in patients with locally advanced tumours who underwent reconstructive surgery with free flaps.
Methods. We performed a retrospective study on a cohort of patients referred to our center from January 2013 to October 2018. We assessed the prognostic role of H-index in terms of disease-free (DFS) and overall survival (OS).
Results. A total of 99 patients were studied, with a median age of 66 years. After a median follow-up of 46.5 months (range, 1.4-121.9), 5-year OS was 59.6% (CI 29.3-57.5) and 5-year DFS 47.5% (CI 16.7-49.2). The H-index showed a statistically significant correlation with a shorter DFS (HR 1.2, 95% CI 1.1-1.4, p = 0.006). No correlations were found between surgical complications and the H-index.
Conclusions. This study confirmed that the H-index is an independent prognostic factor for DFS in patients with HNSCC undergoing microvascular reconstructive surgery and should be used to better stratify the risk of mortality and recurrence, with the aim of improving patient management.

Introduction

The immune system and the inflammatory response have a crucial role in cancer promotion and progression. Indeed, avoiding immune destruction and tumour-promoting inflammation have been identified as cancer hallmarks ¹. During carcinogenesis, several genetic alterations and the loss of normal cellular regulatory processes may lead to the formation of neoantigens. These can be recognised by the immune system, triggering its activation ^2,3.

It is also known that tumour microenvironment (TME), made of both tumour cells and stromal cells, directly stimulates the activation of inflammation and the release of mediators (chemokines, cytokines, and prostaglandins) ⁴, which are also involved in migration, invasion, and metastasis ⁵. Understanding the role of the immune system during carcinogenesis and tumour progression paved the way for the introduction of immune checkpoint inhibitors which interfere with the immune escape mechanisms used by cancer cells, resulting in improved survival ⁶.

In recent years, there have been many attempts to correlate blood values and inflammatory markers with the prognosis of patients affected by head and neck squamous cell carcinoma (HNSCC) ⁷, through the development of prognostic models. The Host Index (H-index), described by Valero et al. in 2020 ^9,10, is obtained by using preoperative blood values on haemoglobin, neutrophils, lymphocytes, monocytes, and albumin; its correlation with prognosis of patients with HNSCC has already been demonstrated in several studies ^10-12.

Specifically, high H-index values correlate with a reduction in disease-specific and overall survival (OS). In the literature, the H-index has been studied in HNSCC patients within specific primary sites (e.g., oral cavity). However, it has never been correlated with postoperative complications. Our study aimed to explore the prognostic impact of the H-index in HNSCC patients undergoing reconstructive surgery with free flaps.

Materials and methods

A retrospective observational study was conducted including a series of consecutive patients treated at the Unit of Head and Neck Surgery of the Fondazione IRCCS Istituto Nazionale dei Tumori di Milano (Italy) between January 2013 and October 2018. All patients underwent upfront surgery for Stage II-IVa/b HNSCC, requiring free flap reconstruction. The following data were collected: age, gender, smoking, and alcohol consumption, disease site and subsite, histological grading, surgical margins, number of pathological lymph nodes, and lymph nodes with extranodal extension (ENE), lymphovascular invasion (LVI), and perineural invasion (PNI).

All patients were staged according to the 8th Edition of the AJCC (American Joint Committee on Cancer) classification ¹².

The following pretreatment blood test parameters were collected: haemoglobin, neutrophils, lymphocytes, monocytes, and albumin. The H-index was obtained using the following formula:

(Neutrophils×Monocytes)/(Haemoglobin×Lymphocytes×Albumin) ×100 ¹²

Following international guidelines, patients underwent regular follow-up with endoscopic evaluation and radiological imaging every 1-3 months in the first and second year, and every 4-6 months for subsequent years for at least 5 years. A chest computed tomography (CT) scan was performed annually in previous and current smokers ¹³.

OS was defined as the time from surgery to death from any cause (event) or last follow-up (censored), disease-free survival (DFS) as the time from surgery to the first evidence of clinical recurrence (locoregional and/or distant) or death from any cause (event) or last follow-up (censored). Median follow-up was assessed with the reverse Kaplan-Meier method.

Univariate Cox regression models were performed including the following parameters: haemoglobin, lymphocytes, monocytes, neutrophils, neutrophil-to-lymphocyte ratio (NLR), pathological stage, surgical margins, ENE, LVI, PNI and H-Index. The results were reported in terms of hazard ratio (HR) and 95% confidence intervals (CI). All blood test parameters were considered as continuous variables. H-index was analysed as both a continuous variable and dichotomised as ≥ vs < 3.6, according to the study of Valero et al. ⁸.

We defined postoperative complications as adverse events requiring a surgical revision of the flap in the operating room. Mann-Whitney test was used to assess the difference in the H-index distribution between groups.

Variables found to be significantly associated with survival (i.e., OS and DFS) were included in a multivariable model according to the recommendations of Harrell et al. ¹⁴.

Survivals were estimated with the Kaplan-Meier method and compared using the log-rank test. Patients were stratified into three different risk categories: high risk (HR) patients with H-index ≥ 3.6 and with one or more negative prognostic factors (PNI, LVI, R1, ENE); intermediate risk (IR) patients with H-index ≥ 3.6 without negative prognostic factors or H-index < 3.6 and with one or more negative prognostic factors; low risk (LR) with H-index < 3.6 and no negative prognostic factors. Statistical analysis was performed using the Statistical Software for Data Science (STATA). A two-side p value ≤ 0.05 was considered statistically significant. The study was approved by the institutional ethical committee of the Fondazione IRCCS Istituto Nazionale dei Tumori di Milano (local study identifier INT79/23). In accordance with national laws, no informed consent was required from the participants due to the retrospective nature of the study and the absence of identifying information.

Results

A total of 174 patients were identified. Detailed follow-up information were missing for 75 cases, and thus 99 patients were included. Median age was 66 years (range, 30-83 years). Most patients had loco-regionally advanced oral cavity SCC (87%), followed by oropharynx p16- (9%), hypopharynx (1%), and other sites (3%). Stage was II in 12%, III in 32% and IVa/b in 53%. Negative surgical margins were achieved in 70 patients (71%), and lymph node metastases with ENE were observed in 28 cases (28%). Adjuvant treatment (RT or CRT) was administered to 78 patients (79%).

The median follow-up was 46.5 months (range, 1.4-121.9). Five-year OS was 59.6% (CI 29.3-57.5) and 5-year DFS 47.5% (CI 16.7-49.2) (Figs. 1-2).

Forty-two patients died (42%), with the disease being the cause of death in 40 (95%). Twenty patients experienced local recurrence, 13 had local and regional recurrences, 6 patients had regional recurrences, and 9 had distant metastases.

Twenty-six patients (26%) experienced postoperative complications, specifically surgical revision of the flap in the operating room (Tab. I).

During follow-up, a second primary tumour was diagnosed in 5 patients. Median H-index was 1.6 (range, 0.4-9.2).

The H-index distribution did not differ between patients with or without surgical complications (p = 0.07). Higher H-index values were significantly associated with shorter DFS (HR = 1.2, 95% CI 1.1-1.4, p = 0.006), not with OS (HR = 1.2, 95% CI 0.9-1.3, p = 0.055).

H-index was ≥ 3.6 in 10 (10.1%) patients, and < 3.6 in the remaining 88 (88.9%) subjects. In the former group, 5-year DFS was 10% (95% CI 2-29) vs. 51.1% (95% CI 16-52) in the latter (p = 0.01). Five-year OS was 40% (95% CI 9.6-53.7) and 60% (95% CI 29.3-59) in subjects with H-index ≥ 3.6 and < 3.6, respectively (p = 0.5) (Fig. 3).

At univariate analysis, the only blood test parameter correlated with a significant increase of the HR for 5-year DFS was serum level of monocytes. None revealed a significant association with OS.

Disease parameters significantly associated with DFS at univariate analysis were stage, ENE, PNI, the presence of positive margins, and H-index. Factors associated with OS at univariate analysis were stage, ENE, and PNI (Tab. II).

Parameters included in the multivariable models were ENE, PNI, stage, and H-index.

At multivariable analysis for DFS, ENE, PNI and H-index were independently associated with DFS, while only PNI maintained a statistical significance with OS (Tab. III). Six HR patients (H-index ≥ 3.6 and +more than one negative prognostic factors) experienced disease recurrence. Among the 73 IR patients, 38 (38.4%) had recurrence, while only 3 of the 20 patients belonging to the LR group (14.2%) had disease relapse.

Discussion

Our results showed that the H-index is a prognostic factor for HNSCC patients treated with reconstructive surgery. A high H-index (≥ 3.6) was significantly associated with shorter DFS, but it did not predict OS.

At univariate analysis, pathological tumour stage, ENE, and PNI were significantly associated with worse OS and DFS, confirming their well-known negative prognostic value ^15-17. Among the various blood parameters taken into account, only monocytes revealed a significant association with outcomes, but only at univariate analysis. This finding is in line with the role of monocyte derivatives (i.e., macrophages) in tumour progression, promoting the proliferation and survival of tumour cells, and lymphangiogenesis ¹⁸, consistent with what was highlighted in the study by Boscolo-Rizzo et al. ¹¹.

H-index maintained an independent prognostic significance at multivariable analysis for DFS only, together with ENE and PNI. The only independent prognostic factor for OS that maintained significance at multivariable analysis was PNI.

In a study by Košec et al., several systemic inflammatory markers were identified as possible predictors of postoperative complications and poor survival in HNSCC patients undergoing reconstructive surgery with microvascular flaps ¹⁹. However, in our study, no association was found between the H-index and risk of postoperative complications.

We tried to stratify patients into different risk groups (HR, IR, LR) and observed that all patients with an H-index greater than 3.6 and with more than one negative prognostic factor experienced disease relapse. Though the results are not validated in terms of DFS, due to the small sample size, they are. The results are, however, very interesting and pave the way for new future studies.

Pre-treatment inflammatory and nutritional index values confirm their prognostic importance ²⁰. Albumin is a surrogate marker of nutrition and an acute-phase protein. It has also been reported in the literature to be associated with increased postoperative morbidity and mortality, as well as more frequent surgical site infection or postoperative complications ^21,22. For this reason, many studies recently underlined the importance of a proper preoperative boosting of the nutritional status of patients to be treated by major surgery, including those planned for microvascular reconstruction ²³. Low levels of haemoglobin in peripheral blood in patients with HNSCC was also associated with tumour hypoxia, which induces radioresistance, compromises the effectiveness of treatments, and negatively influences prognosis with a substantial increase in the relative risk of mortality ^24,25.

Drawbacks of this study include the limited sample size and the retrospective nature of analysis. Acknowledging these limitations, a high H-index (≥ 3.6) was associated with a higher risk of disease relapse, although further evaluation and confirmation on larger samples are advisable.

Conclusions

H-index is an easily retrievable tool composed of parameters routinely collected in the preoperative setting. Our study showed that H-index is a prognostic factor for DFS in HNSCC patients undergoing microvascular reconstructive surgery. Larger case series are needed to validate our findings.

Conflict of interest statement

All authors declare no conflicts of interest, except for Stefano Cavalieri. He declares occasional fees for participation as a speaker at conferences/congresses from AccMed; support for attending meetings and/or travel from AccMed, MultiMed Engineers srl, Care Insight sas.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Author contributions

AD: conceptualization, supervision; AD, VC: methodology; VC, LG, SC, AD: writing – original draft preparation; AD, SC, LG, AA, FF, FI, MP: writing – review and editing; VC, LG: visualization. All authors have read and agreed to the published version of the manuscript.

Ethical consideration

This study was approved by the Institutional Ethics Committee (Fondazione IRCCS Istituto Nazionale dei Tumori of Milan) (approval number/protocol number INT79/23).

The research was conducted ethically, with all study procedures being performed in accordance with the requirements of the World Medical Association’s Declaration of Helsinki.

In accordance with national laws, no informed consent was required from the participants due to the retrospective nature of the study and the absence of identifying information.

History

Received: August 8, 2024

Accepted: December 5, 2024

Figures and tables

Figure 1. Overall survival (5-year OS 59.6%).

Figure 2. Disease-free survival (5-year DFS 47.5%).

Figure 3. Overall survival stratified according to H-index cutoff of 3.6 (p = 0.5)

Table I. Demographic and clinical characteristics of the study population.

Characteristic	N
Age median, range	66 (30-83)
Gender
Male	69 (70%)
Female	30 (30%)
Smoking (current/former)	63 (63%)
Alcohol (current/former)	14 (14%)
Pathologic grading
G1	6 (6%)
G2	65 (65%)
G3	28 (28%)
Primary site
Oral cavity	86 (87%)
Oropharynx p16-	9 (9%)
Hypopharynx	1 (1%)
Other	3 (3%)
Stage
II	12 (12%)
III	32 (32%)
IVA	38 (38%)
IVB	17 (17%)
Extranodal extension (ENE)	28 (28%)
Surgical margins (R1/R2)	29 (29%)
Perineural invasion (PNI)	49 (49%)
Lymphovascular invasion (LVI)	31 (31%)
Surgical complications	26 (26%)
Adjuvant treatment
Radiotherapy (RT)	40 (40%)
Chemoradiotherapy (CRT)	38 (38%)

Table II. Univariate analysis estimated from Cox proportional hazard model.

	Overall survival		Disease-free survival
	HR (95% CI)	p value	HR (95% CI)	p value
Stage (IV vs II-III)	5.11 (1.22-21.50)	0.026	4.35 (1.34-14.19)	0.015
ENE (present vs absent)	2.11 (1.12-3.96)	0.02	2.15 (1.22-3.76)	0.007
Margins (R1/R2 vs R0)	1.24 (0.64-2.40)	0.518	1.71 (0.97-3.02)	0.062
PNI (present vs absent)	1.92 (1.02-3.60)	0.042	2.02 (1.15-3.53)	0.014
Monocytes 103μL (continuous variable)	4.18 (0.84-20.83)	0.081	4.65 (1.01-21.49)	0.049
H-index (continuous variable)	1.15 (0.99-1.34)	0.055	1.22 (1.06-1.41)	0.006
LVI (present vs absent))	1.31 (0.69-2.48)	0.402	1.46 (0.84-2.55)	0.174
Neutrophils 103μL (continuous variable)	1.06 (0.91-1.24)	0.458	1.14 (0.97-1.34)	0.105
Lymphocytes 103μL (continuous variable)	0.82 (0.49-1.37)	0.453	0.90 (0.60-1.37)	0.645
Albumin (continuous variable)	0.54 (0.23-1.24)	0.148	0.55 (0.25-1.19)	0.132
Haemoglobin (continuous variable)	0.91 (0.72-1.13)	0.414	1.01 (0.82-1.25)	0.885

Table III. Multivariate analysis estimated from Cox proportional hazard model.

	Overall survival		Disease-free survival
	HR (95% CI)	p value	HR (95% CI)	p value
ENE (present vs absent)	1.97 (0.97-4)	0.06	2.15 (1.15-4.01)	0.016
PNI (present vs absent)	2.53 (1.26-5.09)	0.009	2.60 (1.42-4.74)	0.002
Stage (IV vs II-III)	2.87 (0.65-12.79)	0.165	2.51 (0.74-8.54)	0.142
H-index (continuous variable)	1.15 (0.99-1.34)	0.068	1.26 (1.08-1.47)	0.003
H-index considered as continuous variable.