Epidermal growth factor receptor inhibitors in advanced cutaneous squamous cell carcinoma: A systematic review and meta‐analysis

Patients with advanced cutaneous squamous cell carcinoma (cSCC) who are not eligible for or who fail to respond to anti‐PD1 immunotherapy have few treatment options. Epidermal growth factor receptor (EGFR) inhibitors have been investigated as a therapeutic option for advanced cSCC; however, data are limited to small single‐arm trials or retrospective studies. A systematic review and meta‐analysis was conducted to PRISMA guidelines (CRD42023394300). Studies reporting on outcomes of EGFR inhibition in advanced cSCC were identified. Objective response rate (ORR), progression‐free survival (PFS), overall survival (OS) and adverse event (AE) rate were pooled using a random effects model and the inverse variance method. Twelve studies (six prospective, six retrospective) were identified, representing 324 patients. Pooled ORR was 26% (95% confidence interval [CI] 18–36), median PFS was 4.8 months (95% CI 3.9–6.6) and median OS was 11.7 months (95% CI 9.2–14.1). Any grade AEs occurred in 93% of patients (95% CI 85–97) while grade 3 and higher AEs occurred in 30% (95% CI 14–54). These results were similar between anti‐EGFR monoclonal antibodies (MAbs) and tyrosine kinase inhibitors (TKIs). EGFR inhibitors can be considered in patients with advanced cSCC who are contraindicated for or progress on first‐line anti‐PD1 immunotherapy. Future studies should evaluate their activity and safety following anti‐PD1, identify predictive biomarkers for their efficacy and explore combination approaches.

growth factor receptor (EGFR) is frequently overexpressed in highrisk (~35%) or metastatic cSCC (~80%), which portends poor prognosis. 6,7EGFR inhibitors have been investigated in advanced cSCC, but studies are limited by small sizes and heterogeneity of patient cohorts.We conducted a meta-analysis of EGFR inhibitors in advanced cSCC to benchmark their efficacy and allow for informed treatment selections and patient counselling.

| ME THODS
This meta-analysis was registered with PROSPERO (CRD42023 394300) and conducted according to PRISMA guidelines. 8Ovid Medline, EMBASE, PubMed, the Cochrane Register of Controlled Trials and Web of Science were searched from dates of inception to the 30 January 2023 applying a combination of MeSH search terms combining EGFR inhibitors and cSCC (Table S1).English language publications (randomised controlled trials, prospective single-arm trials and retrospective observational studies) were included if efficacy outcomes of EGFR inhibitors in ≥5 patients with advanced cSCC were reported.Corresponding authors were contacted for missing data.
Case reports, conference abstracts, editorials and review articles were excluded.Studies of EGFR inhibitors in the adjuvant/neoadjuvant settings or combined with other treatment modalities were also excluded.A structured data extraction table was populated recording study design, sample size, patient demographics, efficacy outcomes (ORR, progression-free survival [PFS] and overall survival [OS]) and adverse event (AE) rate.Quality of the studies was assessed using the National Institute of Health (NIH) assessment tool for interventional studies. 9Screening, data extraction and riskof-bias assessment were performed by two investigators (J.P.P. and A.R.) independently, with any disagreements resolved via consensus.
Due to expected study heterogeneity, random-effects models were pre-specified.Logit-transformed ORR were pooled using the inverse variance method.Heterogeneity of studies was evaluated using Cochran's Q (p < 0.05 suggesting lack of homogeneity) and I 2 statistic (<30% for low heterogeneity, 30%-60% for moderate heterogeneity and >60% for substantial heterogeneity).Published Kaplan-Meier PFS and OS curves were digitised to generate individual patient data using previously described methods, 10,11 then pooled using a multivariate extension of the DerSimonian-Laird method to construct summary PFS and OS plots.These plots were used to estimate median values and survival proportions.

| RE SULTS
Of 743 studies identified in the literature search, a total of 12 studies fulfilled the inclusion criteria (Figure S1).These reported on outcomes of EGFR inhibitors in 324 patients with advanced cSCC.Nine studies investigated anti-EGFR monoclonal antibodies (MAbs, n = 198), whereas three investigated anti-EGFR tyrosine kinase inhibitors (TKIs, n = 126).Six studies were prospective single-arm trials (n = 184) while six were retrospective observational studies (n = 140).Table 1 summarises the characteristics of the 12 included studies.
All included studies were of 'good' (n = 4) or 'fair' (n = 8) quality as assessed by the NIH tool (Table S2).Common potential sources of bias included lack of assessor blinding (n = 12) and lack of descriptions of eligibility screening (n = 10).Quality, scored using the NIH tool, was similar between prospective and retrospective studieswith median points, out of a possible 12, of 9 and 7.5, respectively, Mann-Whitney p = 0.08.Three studies, [12][13][14] all retrospective analyses of anti-EGFR MAbs (n = 44), did not report demographic or clinical characteristics of the cohorts.Based on reported data, 72% of patients were male, 84% had a baseline performance status of 0 or 1, 76% were systemic treatment-naïve, 41% had nodal or visceral metastases and 62% had primaries originating in the head or neck region (Table S3).MAbtreated patients were more likely to be systemic treatment-naïve (86% vs. 64%, p < 0.001) and less likely to have head/neck primaries (52% vs. 72%, p < 0.001) than TKI-treated patients (Table S4).
Similarly, baseline characteristics were comparable between prospective and retrospective study cohorts, apart from a slightly lower prevalence of systemic treatment-naïve patients in prospective studies (72% vs. 85%, p = 0.01) (Table S5).However, given the variability in data availability between studies, these subgroup comparisons should be interpreted with caution.
There were no significant differences for PFS and OS between MAb or TKI treatments (p = 0.27 and 0.42 respectively; Figure 2C,D) or between prospective and retrospective studies (p = 0.35 and 0.43 respectively; Figure S4).

These commonly included acneiform eruptions (8%-15%) with
MAbs 16,17 and diarrhoea (3%-17%) with TKIs. 15,18No treatment-related deaths secondary to EGFR inhibitors were reported.b Quality of studies as measured by the National Institute of Health assessment tool for interventional studies without control groups (Table S2).

F I G U R E 1
Forest plot of reported objective response rates to EGFR inhibitors in advanced cSCC and pooled results, stratified by anti-EGFR MAbs or TKIs.
although robust comparisons between them could not be made given the lack of randomised trials.Moreover, such comparisons should be made with caution due to the fewer studies investigating TKIs, all of which were prospective trials, whereas two thirds of the studies investigating MAbs were retrospective.
Only two studies, both of cetuximab, 17,19 were conducted entirely in systemic treatment-naïve patients, and no studies were conducted entirely in pretreated patients (Table 1).Therefore, we were unable to evaluate the effect of prior exposure to systemic therapies on the activity and safety of EGFR inhibitors for advanced cSCC.While there are no head-to-head randomised trials comparing EGFR inhibitors to anti-PD1 in cSCC, our pooled ORR and survival results with EGFR inhibitor monotherapy are inferior to those reported for anti-PD1 monotherapy. 2,3Variation in patient selection for anti-PD1 versus EGFR inhibitors may contribute to these differences.Although we were unable to calculate the median duration of response to EGFR inhibition due to limited data reporting, the lack of plateaus in the pooled PFS and OS Kaplan-Meier curves (unlike published curves for anti-PD1 2,3 ) suggests most responses are not durable.However, PD-1 inhibitors are relatively contraindicated in some cSCC patients-such as SOTRs who risk allograft rejection and those with uncontrolled autoimmune disorders. 5In these patients, EGFR inhibitors may be a viable treatment option, although we did not identify any dedicated studies in this population.
The efficacy of EGFR inhibition in combination with other treatments in advanced cSCC may be greater than with monotherapy.
For example, a retrospective study by Hourbeight et al. identified an ORR and median PFS of 52% and 6.9 months, respectively, to panitumumab plus radiotherapy. 20Cetuximab, which unlike other EGFR inhibitors is able to activate antibody-dependent cellular cytotoxicity against EGFR-expressing cancer cells, 21 has also been studied in combination with immunotherapy.This includes the I-TACKLE trial (cetuximab plus pembrolizumab) which reported a preliminary ORR of 38% in patients resistant to pembrolizumab alone 22 and an ongoing trial of cetuximab plus avelumab (NCT03944941).
However, combinations of anti-EGFR TKIs with anti-PD1 should be trialled with caution, given high rates of severe AEs observed in other disease settings.For example, the KEYNOTE-021 in EGFRmutant non-small cell lung cancer (NSCLC) observed a 71% incidence of grade 3+ liver toxicity with gefitinib plus pembrolizumab 23 ; while the TATTON trial investigating osimertinib plus durvalumab for the same indication reported a 29% incidence of grade 3+ pneumonitis. 24Timing of anti-EGFR TKIs sequential to anti-PD1 may also be relevant, with one retrospective analysis in NSCLC observing severe AEs with osimertinib occurring more frequently when patients received anti-PD1 within the preceding 12 months compared to >12 months-24% versus 0% respectively. 25Therefore, in patients with advanced cSCC who progress on anti-PD1, addition of/or transition to anti-EGFR MAbs may be preferable, although confirmatory trials are required.F I G U R E 3 Forest plot of pooled EGFR inhibitor-associated adverse events of (A) any grade and (B) grade 3 or above.

S U PP O RTI N G I N FO R M ATI O N
Additional supporting information can be found online in the Supporting Information section at the end of this article.Table S1 Search strategy used to identify investigating EGFR inhibitors in advanced cutaneous SCC.
Table S2 Assessment of study quality using the National Institute of Health Quality Assessment tool for interventional studies with no control groups*.
Table S3 Comparison of baseline demographics in studies by type of therapy and study design.
Table S4 Difference in cohort variables between MAb and TKItreated groups.
All analyses were conducted using R version 4.2.2 (R Foundation for Statistical Computing, Vienna, Austria) with the meta (version 6.1.0)and RISCA (version 1.0.3)packages.
Our results suggest that EGFR inhibitors are associated with modest, but clinically meaningful, benefits in patients with advanced cSCC-with pooled ORR of 26%, median PFS of 4.8 months and OS of 11.7 months.Although, there were no studies reporting on the efficacy of EGFR inhibitor monotherapy after failure of anti-PD1, our data may support EGFR inhibitor monotherapy in the second line given the paucity of effective therapies.Our analysis suggests both anti-EGFR MAbs and oral TKIs have activity in advanced cSCC, TA B L E 1 Summary of included studies reporting on outcomes of EGFR inhibitors in advanced cutaneous squamous cell carcinoma.
However, ORR and median PFS in the treatment-naïve cohorts reported by Maubec et al. (28%, 4.1 months) and Preneau et al. (33%, 5.5 months) appear comparable to the pooled outcomes across cohorts irrespective of prior lines of therapy (26%, 4.8 months).Future studies investigating the activity of EGFR inhibitors following failure of anti-PD1, the more likely sequence in the contemporary management paradigm for advanced cSCC, 2 may find it helpful compare response and survival outcomes against the benchmarks established in our analysis.

F I G U R E 2
Summary of Kaplan-Meier curves for (A) progression-free survival across entire cohort; (B) overall survival across entire cohort; (C) progression-free survival stratified by MAb or TKI EGFR inhibitors; and (D) overall survival stratified by MAb or TKI EGFR inhibitors.
EGFR inhibitors can be considered for patients with advanced cSCC in whom PD-1 inhibitors are contraindicated, or who progress on immunotherapy.Our analysis provided pooled efficacy outcomes for 324 patients with cSCC, larger than any individual trial of EGFR inhibitors.However, limitations included varying patient selection criteria, publication bias and an inability to stratify outcomes based on prognostic features such as sex, stage or lines of prior systemic therapy.Future studies directly comparing MAbs and TKIs, particularly after failure of anti-PD1 and in cohorts not amenable to immunotherapy such as SOTRs would be informative.It is also unclear if EGFR expression is associated with efficacy of EGFR inhibitors in cSCC as this was not reported in any study, highlighting the need to define predictive biomarkers for these agents.These will be essential to integrate EGFR inhibitors into the contemporary treatment paradigm for advanced cSCC and inform evidence-based treatment selections for patients.AUTH O R CO NTR I B UTI O N SJames P. Pham contributed to the conceptualization, data curation, investigation, writing-original draft, and writing-review and editing.Anthony Rodrigues, Simone M. Goldinger and Hao-Wen Sim contributed to the data curation, investigation, writing-original draft, and writing-review and editing.Jia Liu contributed to the conceptualization, data curation, investigation, supervision, writing-original draft, and writing-review and editing.ACK N O WLE D G E M ENTSOpen access publishing facilitated by the University of New South Wales, as part of the Wiley-University of New South Wales agreement via the Council of Australian University Librarians.

Figure S1
Figure S1 Identification process of articles for inclusion from the literature search.

Figure S2
Figure S2 Pooled objective response rates of EGFR inhibitors for advanced cSCC, stratified by prospective or retrospective study design.

Figure S3
Figure S3 Pooled objective response rates of EGFR inhibitors for advanced cSCC stratified by prospective or retrospective study design and restricted to monoclonal antibody interventions.

Figure S4
Figure S4 Pooled Kaplan-Meier curves for (a) progression-free survival and (b) overall survival outcomes of EGFR inhibition in advanced cSCC, stratified by prospective or retrospective study design.

Study Design Cohort size (first-line a ) EGFR inhibitor Outcomes available for meta-analysis Quality (score b )
Abbreviations: MC, multi-centre; NR, not reported; ORR, objective response rate; OS, overall survival; P, prospective; PFS, progression-free survival; R, retrospective; SC, single-centre.a Number and percentage of study cohort receiving EGFR inhibitors as the first-line systemic treatment.

Table S5
Difference in cohort variables between prospective and retrospective studies.