Abstract
Head and neck squamous cell carcinoma (HNSCC) is a deadly disease with a poor prognosis due to late diagnosis and limited treatment options. Immunotherapy (IT) is emerging as a promising approach, especially after the failure of standard of care therapies (STs). The objective of this systematic review and meta-analysis was to evaluate whether the addition of IT to STs improves outcomes for patients with HNSCC, including overall survival (OS), progression-free survival (PFS), and quality of life (QoL). This review employed the Population Intervention Comparison and Outcome (PICO) framework to identify relevant search terms in electronic databases, and also included supplementary hand searches. Six primary research articles were selected using the Preferred Reporting Items for Systematic Review and Meta-analysis (PRISMA) flow chart, and were critically appraised. Data extraction from these studies was conducted, and a meta-analysis was performed to aid in the generation of forest plots. The addition of IT to standard anticancer therapies was found to enhance patient outcomes, such as OS, PFS, and QoL. The toxicity profile of IT was acceptable, with minimal treatment-related deaths. The most frequently observed adverse events (AE) were related to the skin, followed by hematological toxicities. Based on our analysis, the addition of IT to STs is a suitable treatment option and is supported by current research. However, further studies are needed to investigate factors that influence treatment effectiveness and to develop optimal therapies. To achieve this, we recommend a comprehensive treatment approach that involves the multidisciplinary team (MDT) and patient assessment tools.
- Head and neck cancer
- head and neck squamous cell carcinoma
- immunotherapy
- standard therapies
- survival
- quality of life
- review
Head and neck cancer (HNC) comprises 5% of all malignancies worldwide, with head and neck squamous cell carcinoma (HNSCC) being the most common subtype (1). Approximately 650,000 new cases of HNC are diagnosed annually, resulting in 350,000 deaths (2). The primary risk factors for HNC are smoking, alcohol consumption, and Human Papilloma Virus (HPV) infection (3). Diagnosing HNSCC is challenging, and it is often detected when the disease is locally advanced or metastatic (4). The standard treatments (STs) for HNC depend on the primary tumour’s location and traditionally include surgery, radiotherapy (RT), and chemotherapy (5). In the early stages (I and II), surgery and RT have high cure rates, with one- and two-year survival rates of 88.7% and 79.8%, respectively. However, relapses can still occur despite intensive treatment, leading to a poor prognosis (6). For cases where relapse occurs or upfront metastatic tumors are present, the aim of treatment is palliative rather than curative (7). For the last three decades, platinum-based chemotherapy has been the standard of care for these patients, with a median overall survival (OS) of approximately seven months (8). Immune checkpoint inhibitors (ICIs) were initially evaluated after frontline therapies failed, and have since transformed the outcomes for these patients (9).
Recent studies have suggested that combining ICIs with STs may improve patient outcomes compared to chemotherapy-based regimens (10). Research has also demonstrated that incorporating ICIs into radiotherapy (RT) enables modified fractionation, thereby decreasing the radiation dose delivered to the circulating blood supply (11). This technique preserves proximal lymphocytes while still promotes anti-programmed death-ligand 1 (PD-L1) antibody activity, thereby enhancing the effects of ICIs (11). Additionally, research has indicated that the combination of RT and ICIs can induce an abscopal effect. In this phenomenon, local treatment triggers a systemic anti-tumor immune response, affecting distant metastatic lesions that were not directly exposed to radiation (12, 13). This systematic literature review and meta-analysis aims to provide current evidence on the use of ICIs in combination with STs for patients with HNSCC, evaluating whether it improves survival outcomes and patients’ quality of life (QoL).
Methods
This study utilized a literature review methodology to gather data, following the requirements of the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines. The research question was formulated using the Population Intervention Comparison and Outcome (PICO) framework (14). The rationale for using PICO is that it helps create a clear, focused research question, which assists in the database search process by highlighting relevant keywords (15). The refined PICO question for this study was, “Does adding immunotherapy (I) to standard therapies (C) improve outcomes (O) for patients with head and neck squamous cell carcinoma (P)?”.
To identify eligible primary research articles, we conducted a comprehensive search of four electronic databases. We selected Cumulative Index to Nursing and Allied Health Literature (CINAHL) Complete and MEDLINE Complete for their wide range of nursing, medical, and health-related topics. We included Embase and the Cochrane Central Register of Controlled Trials (CCRCT) for their specific focus on drug therapy research. The use of CCRCT can be criticized because it does not contain full-text articles, which limited our ability to gather studies. However, we included it because it provided valuable data to facilitate our understanding of the efficacy of immunotherapy (IT) and encourage further reading and research relevant to the review question.
Using different databases with various specialties and entering the keywords ensured a broad range of research was obtained (16). We performed four searches to gather a wealth of information on the effects of combining IT and STs on the outcomes of HNSCC patients. We began by conducting searches using different variations of the PICO keywords combined with the Boolean operator ‘OR’ and truncations to broaden the search (Table I). We then combined these searches with the Boolean operator ‘AND’ Table I), yielding 349 results before applying limiters (Table II).
Outline of PICO search strategy and results.
Inclusion, exclusion and limiting factors for research articles.
Hand searching was employed to supplement the database searches as research has shown that this method can help locate up-to-date articles, enabling a broader yet efficient search (17). In total, five studies were found through hand searching, and after applying filters, a total of 39 articles were identified, which were reduced to 32 after removing duplicates. The abstracts of these articles were screened using inclusion and exclusion criteria (Table II) to determine their eligibility for review. After duplicates were removed and limiters, inclusion, and exclusion criteria were applied, the articles were refined to six. These six full-text papers were critically evaluated for eligibility, and the methods were assessed using critical appraisal tools (Appendix I). For this review, the Critical Appraisal Skills Programme (CASP) for cohort studies and the Centre for Evidence-Based Medicine (CEBM) for prognostic studies appraisal tools were used (Appendix I) (18, 19). These tools are useful as they provide structured checklists allowing the simplified breakdown of research. The CASP was selected because four of the six studies evaluated were retrospective cohort reviews, making it the most appropriate appraisal tool (18). The CEBM tool was chosen because it analyzes studies that consider chosen variables and how these impact the outcome of a disease; thus, it was the most applicable for the non-randomized clinical trials examining the impact of IT plus or minus ST (19).
Data was extracted from the selected articles, including the authors, year of publication, country of origin, study design, sample size, methods of patient recruitment, and outcomes. A meta-analysis was performed using the extracted data. This review considered quantitative data, with all studies containing statistics regarding overall survival (OS) and progression-free survival (PFS). A forest plot was generated to visualize the data, facilitating the identification of statistical and clinical significance from the summarized evidence (20).
Results
The methodology for screening the selected articles is demonstrated using the PRISMA framework, as shown in Figure 1 (21). The 32 articles found through database and hand searches were initially screened by title and abstract for eligibility. Of these, eight full-text articles were assessed for eligibility for appraisal, and six articles were deemed appropriate for review. A summary of these articles can be seen in Table III.
The PRISMA flowchart illustrates the method used to select references through the search process. Revised from Moher et al. (21).
Summary of the reviewed papers.
Study characteristics. The studies included in this meta-analysis were conducted in various countries, including one in the USA (22), one in France (23), one in Germany (24), one multicentre study in France and Belgium (25), one in Spain (26), and one conducted internationally (27). Among the selected articles, four used a retrospective cohort study design (22-24, 26) and two were prognostic, non-randomised phase II studies (25, 27). Three studies employed mixed methods (22, 26, 27) and three used quantitative methods (23-25).
Table III presents the sample sizes, sex distribution, and median age of the participants in the reviewed studies. Vargo et al. (22) included 28 patients, Burgy et al. (23) included 59 patients, Milanovic et al. (24) included 23 patients, Guigay et al. (25) included 54 patients, Sosa et al. (26) included 33 patients, and Zandberg et al. (27) included 112 patients. The sex distribution and median age of the participants varied among the studies.
Small sample sizes in research can be problematic when attempting to generalize findings to the larger population. However, in the studies by Burgy et al. (23) and Sosa et al. (26), the authors acknowledged the issue of small sample sizes. Sosa et al. (26) justified their small sample size by highlighting how their results were consistent with other similar studies. Burgy et al. (23) excluded certain patients from their study in order to prioritize patients’ safety, as there is currently no standard care for elderly HNSCC patients. Although Burgy et al. (23) specifically studied an elderly population, this should not be viewed as a limitation, as older adults are a group with a higher risk of developing HNSCC, thus identifying appropriate treatments for this population is critical.
An additional weakness of the reviewed research is that it did not account for patient characteristics such as socioeconomic status and ethnicity. These factors can play a significant role in the development of HNSCC, treatment effectiveness, and patient QoL (28, 29), which can impact the validity and generalizability of the findings to the entire HNSCC patient population.
Considering the PICO question (14), all six studies reviewed in this meta-analysis assessed the population of HNSCC patients who received IT interventions and evaluated the outcomes such as OS and PFS. In terms of comparison, two studies examined the effect of cetuximab as an adjuvant to chemotherapy (23, 25). Two studies investigated the impact of cetuximab with RT (22, 24), and Zandberg et al. (27) studied the IT agent durvalumab, a monoclonal antibody that targets the PD-L1 receptor. However, it can be argued that Zandberg et al.’s (27) study did not fully address the PICO question of this review as it did not assess the addition of IT to STs. Nonetheless, evaluating the efficacy of IT monotherapy may indicate whether IT is a useful addition to ST or if it could be used as a treatment alone. A limitation of this review is that the reviewed articles did not include a control ST arm for comparison with the IT and ST combination. Hence, the selected studies alone do not address the PICO question of this review.
All six reviewed articles employed data and statistical analysis methods to determine outcomes such as OS and PFS, using the Kaplan–Meier method (as detailed in Table III) (22-27). Two of the mixed methods studies utilized questionnaires to obtain qualitative data on patient QoL (22, 27). Vargo et al. (22) used the validated University of Washington Quality of Life Revised (UW-QoL-R) questionnaire, which measures patient-reported QoL (PR-QoL) with 12 head and neck and three global domain Likert-type questions. Zandberg et al. (27) employed the European Organisation for Research and Treatment of Cancer (EORTC) QoL questionnaire, containing 30 core and HNC-specific symptoms, with the corresponding HNC EORTC module used throughout treatment to give researchers a comparison point. Sosa et al. (26) informally asked participants about their perceptions of their pain and anxiety (whether they felt more, equal, or less) regarding their tumors at every three-week follow-up, evaluating adverse events (AE).
Key findings and themes. Efficacy. To determine whether adding IT to ST improves outcomes, it is important to establish efficacy. This meta-analysis measured OS and PFS, as shown in Table III, by using patient baseline values when treatment started. Forest plots for OS and PFS were created for Burgy et al. (23), Guigay et al. (25), Sosa et al. (26), and Zandberg et al. (27) (Figure 2 and Figure 3), while Vargo et al. (22) and Milanovic et al. (24) were excluded. Vargo et al. (22) measured the percentage of OS and PFS at different patient follow-up intervals, and Milanovic et al. (24) did not include confidence intervals, making their data incompatible. The forest plots show point estimates and interquartile range (IQR) for OS and PFS, with the studies using median and IQR data rather than standard deviation and means. This method is resilient to outliers and normal data, making it easier to identify anomalies (30).
Forest plot displaying the overall survival data.
Forest plot displaying the progression-free survival data.
The forest plot for OS, which is measured from patient baseline (before treatment) until death or the Kaplan–Meier estimation of this, indicates outcome homogeneity (i.e., consistency in the results across the studies) as the IQR overlap.
The forest plot for PFS, which measures the time from patient baseline until disease progression or the Kaplan–Meier estimate of this, also demonstrates consistency in the results of the four studies.
The consistency of outcomes observed between the reviewed studies indicates a high level of internal validity and reproducibility, making the findings applicable to the general population. However, a limitation of these studies is the absence of a comparison arm to compare the outcomes of IT combined with ST versus ST alone. Therefore, in the discussion section, it is necessary to compare the results of the reviewed articles with similar current research that has a comparison arm. This leads to data indirectness, which can affect data quality and applicability. However, since the reviewed articles have clinical homogeneity, meaning they assess similar population groups using similar treatment protocols that are not different enough to affect the results, a “similarity assumption” can be made. Nonetheless, data quality still needs to be downgraded accordingly (31).
The risk of bias can affect the grade and applicability of research findings. Among the reviewed studies, Vargo et al. (22), Burgy et al. (23), Milanovic et al. (24), and Sosa et al. (26) have low to moderate risk of bias (see Appendix II), while Guigay et al. (25) and Zandberg et al. (27) have low risk of bias (see Appendix III). This suggests that the findings of the reviewed articles are valid and transferable to the HNSCC patient population.
Quality of life. The six papers reviewed have highlighted determinants of QoL, which is a holistic assessment of patients’ overall wellbeing. Two key determinants identified were AE and treatment regimen, which impact attrition rate. Three studies used qualitative assessments of patients’ QoL (22, 26, 27) and all showed improvements in patients’ QoL. For example, Sosa et al. (26) found that “all responding patients improved the feeling of local pain and anxiety caused by the tumour”. Zandberg et al. (27) found clinically significant improvements in global health status, physical functioning, and fatigue. The increase in global and physical functioning is particularly meaningful as it suggests an overall improvement in the patient’s health status. Clinically meaningful improvements were also seen in HNC-specific symptoms such as taste and smell, swallowing, mouth pain, and speech.
Furthermore, Vargo et al. (22) found that 56% of patients reported that their QoL had improved or remained the same throughout the trial. PR-QoL remained stable throughout the entirety of the head and neck-specific and general health-related criteria for the duration of survey examination. Mean scores obtained from UW-QoL-R responses were compared, using the Wilcoxon signed-rank test, to patient baseline scores and a p-value of less than 0.05 was deemed statistically significant. Thus, the specific UW-QoL-R result for pain (p=0.034) and activity (p≤0.41) when compared from patient baseline to one year showed a statistically significant improvement, which is arguably clinically significant. This suggests that the combination of IT and ST may improve patients’ QoL.
Adverse events. All six articles reviewed reported on AE and treatment-related adverse events (TRAE). Three studies reported treatment-related deaths (23-25). Burgy et al. (23) had the highest percentage of treatment-related deaths (5%), all due to sepsis. Milanovic et al. (24) reported a patient death due to anaphylaxis during initial intravenous administration of cetuximab. Guigay et al. (25) reported a death of unknown attribution. The impact of TRAE and treatment-related death on study attrition rates needed to be better addressed by the reviewed articles, as this may consequently bias results, reducing the reliability of the research findings.
Four of the studies utilized the Common Terminology Criteria for Adverse Events (CTCAE) to evaluate adverse events and toxicities (23, 24, 26, 27), while Vargo et al. (22) and Guigay et al. (25) employed the National Cancer Institute’s CTCAE v3.0 to grade AE. However, Vargo et al.’s (22) AE were recorded by physicians rather than patients at baseline, which they acknowledged as a limitation as it may lead to an underestimation of toxicity. To mitigate potential bias, Vargo et al. (22) also used PR-QoL to assess AE.
One limitation of this review is that it compares studies that use different grading systems for AE. This is problematic because there may be differences in the standardized terms used for each tool to record and grade the AE, and research has shown that grading tools such as CTCAE are often misused (32). However, Zhang et al. (32) concluded that this is clinically insignificant because, despite variations, the reader’s ability to understand the treatment toxicity profile is not affected. Figure 4 depicts the most frequently occurring AE of any grade, which include mucositis (22-26), skin (22, 24, 25), and haematological toxicities (23, 25-27).
Most commonly occurring grade 1-4 adverse events.
Common grade 3 and 4 toxicities reported in the reviewed studies were related to blood (23, 25-27), infection (22, 23, 25), and skin (24-26) and are illustrated in Figure 5. According to the Cancer Therapy Evaluation Program’s (33) CTCAE, grade 3 and 4 toxicities are classified as “severe AE” and “life-threatening or disabling AE”, respectively, indicating a significant impact on patient QoL. Therefore, it is promising to note the relatively low occurrence of these toxicities in the 6 studies. For instance, in Vargo et al.’s (22) study, 57% of patients experienced no acute and 80% had no late toxicities, whereas in Milanovic et al. (24), no grade 4 toxicities were reported.
Grade 3 and 4 toxicities.
Treatment regimen and attrition. Table IV summarizes the treatment regimens and attrition rates observed in the included review articles. Noting inconsistencies in the treatment regimes for the analysed studies is essential as this affects validity. Clinical homogeneity, which refers to the similarity and comparability of the method of intervention implementation and outcome measurement, can be observed between the six studies, thus suggesting that their findings are reliable and replicable.
Attrition rates and treatment regimens of the reviewed papers.
Considering attrition rates, Vargo et al. (22) and Milanovic et al. (24) had the best attrition rates, with all patients completing the regimen; 20 out of 23 patients completing the ST element, and 23 out of 23 completing the IT part of the regimen, respectively. Both studies had the same adjuvant therapy (RT and cetuximab). Two of the studies showed that patients had an allergic response to cetuximab (24, 26). The most common reason for discontinuation of the treatment was disease progression, followed by adverse events and unacceptable treatment toxicity, treatment-related death, and personal decision.
Discussion
The findings of this review can be grouped into four main discussion points, namely efficacy, QoL including AE, treatment regimen, and attrition. Additionally, recommendations for practice, research, and education are evaluated.
The four reviewed articles that assessed the combination of IT and ST (23-26) showed similar values for OS and PFS as current research that evaluated the same combination (34). Vermorken et al.’s research (34) also found significant improvements in OS and PFS between adjuvant therapies versus the ST alone arm, providing further support for the combination of IT and ST. Figure 6 illustrates a statistically significant difference in OS between the combination of IT and ST compared to ST alone in four of the reviewed studies and the EXTREME trial. Conversely, Zandberg et al. (27) found no significant difference in OS. They only examined IT monotherapy, suggesting that the improvement in OS is due to the combination of IT and ST rather than single modality ST or IT.
Forest plot showing overall survival in reviewed studies versus current research.
The majority of the reviewed studies on adjuvant ST and IT showed an improvement in PFS compared to ST alone, as depicted in Figure 7. However, Zandberg et al. (27) reported a decrease in PFS compared to ST alone. Both Szturz et al.’s (35) EXTREME trial and Guigay et al. (25) showed statistically significant improvements in PFS compared to ST. While Burgy et al. (23) and Sosa et al. (26) also demonstrated improvement, it was not statistically significant.
Forest plot showing progression-free survival in reviewed studies versus current research.
It is crucial to differentiate between statistical and clinical significance, especially in the context of survival data (36). This is because a patient adding a month to their overall survival may not be statistically significant, but the clinical benefit of gaining this extra month to spend with their loved ones outweighs this statistical insignificance. Moreover, a treatment that leads to a statistically significant increase in overall survival may not be appropriate if it is accompanied by intolerable toxicity that diminishes the patient’s QoL (37). Hence, a comprehensive evaluation of significance is necessary while making treatment recommendations.
Burgy et al. (23) and Guigay et al. (25) investigated the use of adjuvant cetuximab and chemotherapy, with Guigay et al. (25) reporting the longest OS (14 months) and PFS (6.1 months) among all the reviewed studies. The National Institute for Health and Care Excellence (NICE) also recommends this combination and suggests the use of cetuximab and platinum-based chemotherapy followed by maintenance cetuximab until disease progression for recurrent HNSCC (38). Therefore, the review’s findings are consistent with NICE’s treatment recommendation, indicating that adding IT to the ST regimen can improve patient outcomes.
Additional support for the use of adjuvant IT and ST comes from a case report of a patient who participated in the EXTREME trial and achieved a disease-free survival of eight years, with a relatively low to moderate incidence of adverse events except for grade 3 stomatitis (39). However, it is important to note that case studies have limitations as their findings may not be generalizable to the larger population due to individual differences (40).
The findings from the two reviewed articles on combination RT and IT (22, 24) show promise, but there is currently limited clinical guidance on the use of this approach for palliating HNSCC. Palliative RT and IT can be justified by a case study results of a patient with inoperable salivary gland mucoepidermoid carcinoma who underwent re-RT and cetuximab, resulting in a complete response after 25 months with good tolerability (41). However, the transferability of these findings to HNSCC is difficult due to differences in individual cell types’ responses to therapy, such as differences in oxygen supply, tumour proliferation, and density (42, 43). This highlights a weakness in the reviewed articles’ methodology as tumour sites with potentially different responses to treatments were not distinguished between. This could affect the reliability and replicability of the review findings. Nonetheless, all the tumours in the reviewed studies were of the same histological type, which reduces this bias.
Zandberg et al.’s (27) study on IT alone had the lowest median OS (7.1 months) and PFS (2.1 months) among the reviewed studies, indicating that the combination of IT and ST is more effective than either IT or ST monotherapy. However, this contradicts the findings of NICE (44) that single IT increased OS by 2.6 months compared to chemotherapy, upon which their recommendation for nivolumab as a second-line treatment for patients who recurred after platinum-based chemotherapy was based. Similarly, the United States Food and Drug Administration recommends second-line nivolumab and pembrolizumab (45), but concerns have been raised about nivolumab’s efficacy in low-PD-L1-expressing HNSCC. Zandberg et al.’s (27) methodology highlighted the PD-L1 tumor expression issue since only patients with highly expressed PD-L1 tumors were included in their research on durvalumab treatment, making the research focused on the sub-group most likely to respond to this therapy. Therefore, their results may be biased, unrepresentative, and lack generalizability.
Prioritizing patient well-being is a nursing care standard, and preserving QoL is an essential outcome of anticancer therapy (46, 47). In the case of HNSCC, symptom relief (6), improving QoL, and disease control are often the main objectives for patients (48). Thus, addressing these outcomes is necessary for adjuvant immunotherapy to be appropriate and improve patient outcomes. Three reviewed studies demonstrated increased patient QoL (22, 26, 27). These functional improvements are supported by the findings of Harrington et al. in their Phase III study Checkmate 141, who found both statistically and clinically significant improvements in role and social functioning, fatigue, appetite, pain, and sensory problems, among others, in the immunotherapy versus standard therapy arm (49). Zandberg et al. (27) and Harrington et al. (49) had similar results using the same assessment and treatment method, suggesting these findings are generalizable. However, the EXTREME trial – contradictory to Vargo et al. (22) and Zandberg et al. (27) – found non-statistically significant improvement compared to baseline in the immunotherapy and standard therapy arm compared to standard therapy alone for general QoL (50). However, the HNSCC-specific symptoms of pain, swallowing issues, speech, and social eating were significantly improved, which was supported by the review findings (22, 26, 27). In conclusion, this suggests that adding immunotherapy to standard therapy specifically improves HNSCC-related symptoms and patient QoL compared to standard therapy alone.
When considering the benefits of IT on patient outcomes, efficacy is a crucial factor. However, a treatment may be effective, but if it is not safe, then it may not be suitable (51). Three of the reviewed studies reported treatment-related deaths, but grade 3 and 4 toxicities were relatively infrequent (23-25). Vargo et al. (22), Burgy et al. (23), and Sosa et al. (26) all concluded that IT is an acceptable addition to ST in the treatment of HNSCC patients, who are often immunocompromised and thus vulnerable (52). Guigay et al. (25) also supported the tolerability of the IT and ST safety profile for HNSCC. Their findings are consistent with those of Vermoken et al. (6), who commented on the satisfactory adverse event profile in the EXTREME trial, noting no cetuximab-related deaths. These findings suggest that the addition of IT to ST may reduce toxicity compared to ST alone.
None of the reviewed articles stated that patients discontinued treatment due to the intensive treatment regimen (Table IV). However, the effect of treatment plans on patient QoL should not be ignored, as decreased quality of life can affect patients’ treatment adherence and, consequently, treatment effectiveness. It could also potentially lead to a waste of limited healthcare resources (53-55).
The results regarding treatment attrition rates for IT plus ST versus ST alone are conflicting. Milanovic et al. (24) reported that all 23 patients completed the prescribed IT, but only 20 out of 23 received the ST However, Vargo et al. (22) had a 100% completion rate. Mésia et al. (50) also support the reviewed studies’ findings, reporting higher dropout rates in their chemotherapy arm compared to the IT and chemotherapy combination. However, Lee et al. (56) found that the addition of IT to induction chemotherapy reduced completion rates. The impact of IT on ST completion rate requires further study to determine a causal link that could potentially improve treatment attrition rates and optimize patient outcomes.
Vargo et al. (22) described how their treatment plan, which used target RT and took advantage of radiobiological benefits, resulted in a more effective regimen that delivered fewer large fractions, thereby reducing treatment time and acute toxicities compared to standard RT. Addeo et al. (57) also supported the use of IT for treatment optimization, as they found that the use of maintenance IT after adjuvant IT and chemotherapy improved effectiveness, convenience, tolerance, and compliance with the treatment plan. However, further research is necessary to weigh the costs and benefits of optimizing treatment.
The review findings, along with current research, have identified areas for development in practice, education, and further study. Table V summarizes the strength of the recommendation to implement IT as an adjuvant to ST in the treatment of HNSCC (58). The analysis of these findings suggests that IT as an adjuvant to ST can address the issue of poor outcomes for HNSCC patients compared to ST alone by improving OS, PFS, and QoL. Although IT is expensive (59), the unit cost per benefit is low. Therefore, this review argues that the addition of IT to ST is appropriate.
GRADE evidence to decision; the approach used for making a healthcare recommendation – Adapted from Alonso-Coello et al (59).
More data is required to comprehend why some patients respond better to treatment than others and how to overcome potential barriers such as drug resistance (60). Investigations into the impact of HPV status, which has been shown to be a confounding variable for positive treatment outcomes, may help identify ways to increase treatment efficacy, such as survival and response rates. This is supported by Zandberg et al. (27), who reported a 29.4% response rate in HPV-positive tumors compared to 10.8% in non-HPV-positive tumors. Similar results were published by Lee et al. (56). Additionally, further research into utilizing the IT and ST regimen to improve HNSCC patients’ QoL and functionality would aid them in continuing to participate in activities of daily living, which is a primary nursing goal (61).
Two of the studies reviewed in this analysis used a multidisciplinary team (MDT) approach for treatment and assessment (22, 24). Research in oncology suggests that MDT input can improve outcomes and overcome barriers to patients receiving holistic care (48, 62). An MDT approach is particularly important for the treatment of HNSCC, which is complex and requires a range of expertise (63). Therefore, the utilization of an MDT approach is recommended in this review. Additionally, the implementation of systematic assessment tools for patients, such as the PR-QoL (22), could streamline the treatment process, ensure holistic assessments, and improve patient outcomes. These tools could be incorporated through nursing education and training.
One weakness of this review is the limited number of studies included, with only six selected in total. This was due to strict inclusion and exclusion criteria, such as study design, which meant that most of the available research did not meet these criteria. Furthermore, four of the studies were retrospective in design, which can lead to selection bias and small sample sizes. Retrospective studies like those reviewed in this article are more susceptible to selection bias because the participants must be representative of the same population, such as having HNSCC (64). However, retrospective cohort study designs have the benefit of using archived data, making them quicker and cheaper to conduct than follow-up prospective cohort studies (65). Moreover, they validate evidence and confirm relationships between treatment, such as IT, and data, like OS and PFS, from other weaker studies, which then safely facilitates further prospective research (65).
A general limitation of this study is that it only reviewed research published up to 2019, which threatens the comparability and applicability of the findings to current treatment recommendations (66). Since then, subsequent research has been published, including the EAGLE and Keynote-048 phase III studies by Ferris et al. (67) and Harrington et al. (68), respectively, which discuss the impact of IT and ST in the treatment of HNSCC. Ferris et al. (67) found no statistically significant differences in OS in IT monotherapy (durvalumab and dervalumab-tremelimumab) versus ST alone, similar to Zandberg et al. (27). However, they noted increased response and survival rates between 12 and 24 months for patients receiving durvalumab, indicating its clinical activity. Both Zandberg et al. (27) and Ferris et al. (67) examined IT monotherapy alone, suggesting that the significant improvement in OS in the other studies evaluated in this paper is due to the combination of IT and ST rather than single modality ST or IT. This deduction is supported by Harrington et al.’s Keynote-048 study, which found that pembrolizumab monotherapy and pembrolizumab-chemotherapy improved OS and PFS compared to ST second-line taxanes alone (68). Evaluating this current research against the original studies analysed in this review allows for comparison of findings across different conditions, interventions, and time, which enables a discussion of whether one can be confident in the initial conclusions drawn. The homogeneity of subsequent research findings with the original evidence analysed demonstrates its quality, as the outcomes are clearly replicable over time, and thus have external validity. Therefore, it is fair to infer that these results are reliable despite the latest research (69).
Lastly, a potential limitation of this review is that all the studies analysed were conducted in high-income countries, which may limit the generalizability of the findings to low and middle-income countries (70). This is particularly problematic since IT is expensive (59) and may not be affordable or appropriate in resource-limited settings. Additionally, none of the included studies were conducted in the UK. Vargo et al.’s (22) study was conducted in the USA, which could be a concern given the differences in healthcare policies and guidelines between countries. For instance, in the USA, cetuximab is approved as a first-line monotherapy for HNSCC, whereas in the UK, it is approved as a first-line combination therapy with platinum chemotherapy (8). However, since these studies focused solely on the effects of IT on patient outcomes, the findings are still relevant.
It is worthy to be mentioned that sex-specific medicine, an evolving field in healthcare, has garnered substantial recognition and significance in recent years. In a retrospective case-matched analysis, there was an observed tendency for female patients to exhibit a higher 5-year OS probability compared to their male counterparts across different therapeutic regimens for HNSCC (71). Furthermore, real-world data have been recently published regarding the efficacy and safety of palliative first-line ICIs in platinum-sensitive patients with recurrent or metastatic HNSCC treated outside of a clinical trial, thus closely mirroring clinical practice. The data revealed a disease control rate of nearly 50% and a response rate of approximately 20% in a disease with a historically unfavourable prognosis. Notably, these treatments were relatively well-tolerated by patients with multiple comorbidities (72).
Finally, we should report that no direct comparison has been made so far to determine the superiority between pembrolizumab monotherapy and pembrolizumab combined with chemotherapy for treating recurrent or metastatic HNSCC. Patients who experience adverse events with ICIs, specifically immune-related adverse events, tend to exhibit more favorable prognoses and may also demonstrate long-term maintenance of efficacy. In a recently published study, the addition of chemotherapy did not contribute to an improvement in prognosis (73). Therefore, when contemplating the long-term treatment of patients with recurrent or metastatic HNSCC, opting for pembrolizumab monotherapy may be more favourable than combination therapy. Moreover, the use of monotherapy has the potential to mitigate additional adverse effects associated with combination therapy.
Conclusion
Based on the selected articles, it can be concluded that IT is a useful adjunct to ST in the treatment of HNSCC. This conclusion is supported by current research and guidelines that suggest combining IT and ST leads to improved patient outcomes such as OS and PFS. Moreover, the results indicate that IT is a safe and appropriate addition to ST, with acceptable adverse effects and toxicities observed and minimal treatment-related deaths.
Furthermore, incorporating IT into HNSCC treatment has been demonstrated to enhance patient QoL by improving general function and addressing head and neck specific symptoms such as tumour-related pain and anxiety, swallowing difficulties, and appetite issues. Therefore, from a patient-centred perspective, the combination of IT and ST provides a clinically significant benefit.
However, additional research is required to understand the reasons for variations in patients’ responses to treatment and to identify and overcome any potential barriers to therapy success. The implementation of policies related to patient assessment and multidisciplinary team collaboration will also enhance the approach to HNSCC treatment.
Footnotes
Authors’ Contributions
DE – conceptualization, writing – original draft preparation, writing – review and editing; AG – conceptualization, data curation, writing – review and editing, visualization; MM – methodology, validation, resources; JAPF – software, formal analysis, project administration; ER – validation, investigation; SB – conceptualization, validation, writing – review and editing, supervision, funding acquisition.
Conflicts of Interest
The Authors declare that they have no conflicts of interest in relation to this study.
- Received January 30, 2024.
- Revision received February 6, 2024.
- Accepted February 7, 2024.
- Copyright © 2024 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
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).
