Abstract
Background
Non-melanoma skin cancers (NMSC) are the most common malignancies worldwide. While early-stage lesions can be definitively treated with local therapies, advanced stage cutaneous squamous cell carcinoma (cSCC) often requires systemic treatments such as PD-1 inhibitors. These treatments may be administered for prolonged durations; this practice may lead to an unnecessary physical and financial toxicity. The purpose of this study was to evaluate the patterns of disease progression after anti-PD-1 therapy discontinuation in this group of patients.
Methods
This retrospective cohort study included patients diagnosed with advanced cSCC and treated with either cemiplimab or pembrolizumab from 2019 to 2024 at a single university-affiliated tertiary medical center.
Results
The cohort included 131 patients, with a 73% overall response rate. Among the 86 patients with either partial or complete response as the best response included in the final analysis, 40 (47%) patients had a treatment break for at least 3 months, and 46 (53%) continued without discontinuation to a maximal duration of 2 years. After a median follow-up of 29.9 months, 24 (60%) patients in the break group remained progression-free, systemic treatment-free, and alive throughout the follow-up. Four patients (10%) experienced disease progression. Among these, the best overall response was PR in three patients and CR in one patient. Nine (22.5%) patients died due to non-oncological reasons, two (5%) patients died from an unknown cause, and one (2.5%) due to treatment toxicity. The percentage of patients achieving CR was statistically significantly higher in the break group compared to the no-break group.
Conclusions
Our findings advocate for a more tailored approach to the duration of PD-1 inhibitor therapy in cSCC, potentially reducing burdens of overtreatment. Future studies regarding establishing robust predictors for safe treatment discontinuation are required to enhance decision-making in clinical practice.
Supplementary Information
The online version contains supplementary material available at 10.1007/s00262-025-04115-y.
Keywords: Cutaneous squamous cell carcinoma, PD-1 inhibitors, Anti-PD-1 therapy discontinuation
Introduction
Non-melanoma skin cancers (NMSC) are the most prevalent type of malignancy worldwide. An important and prevalent group of NMSC is cutaneous squamous cell carcinoma (cSCC) [1]. While early-stage lesions are often cured with local treatment such as surgical interventions or radiotherapy, patients suffering from advanced stage disease might not be suitable for definitive local treatment (due to medical inoperability, severe surgical morbidity, or distant metastatic spread), thus necessitating systemic therapy [1]. These more severe cases can display an aggressive clinical course, characterized by significant pain, bleeding, recurrent infection, and disfigurement—all of which markedly diminish patients’ quality of life [2].
Since the demonstration of cemiplimab and pembrolizumab as potent agents in treating cSCC in the EMPOWER-CSCC-1 and KEYNOTE-629 studies [3, 4], systemic anti-PD-1 immunotherapy has become a more appealing option as it can achieve symptom relief and improve outcome without the toxicity associated with chemotherapy. While treatment can provide significant clinical benefits, it may paradoxically deteriorate patients’ quality of life due to severe immune-related toxicity [5].
The optimal duration of anti-PD-1 therapy for cSCC remains undefined. In the EMPOWER-CSCC-1 study, patients were treated with cemiplimab every 2 or 3 weeks for up to 96 weeks, and in the KEYNOTE-629 study, pembrolizumab was administered every 3 weeks for up to 35 cycles (approximately 2 years). Both treatments were continued until disease progression or unacceptable toxicity was observed [3, 4]. The National Comprehensive Cancer Network (NCCN) guidelines recommend continuing treatment until disease progression or unacceptable toxicity occurs [6]. However, evidence supporting the benefit of prolonged treatment remains limited, and this approach carries a risk of overtreatment, with a chance of unnecessary physical and financial toxicity.
Unlike patients suffering from melanoma, NMSC patients tend to be older and frailer, making them potentially more susceptible to severe immune-mediated toxicity [7, 8]. Therefore, understanding the clinical sequelae and the benefit from prolonged PD-1 inhibitor therapy is critically important.
In this study, we evaluated the patterns of disease progression after anti-PD-1 therapy discontinuation in a large real-world cohort of patients with squamous cell carcinoma of the skin who achieved a partial or complete response to treatment.
Materials and methods
Patients
The study included all patients treated for cSCC with cemiplimab or pembrolizumab as first-line or advanced lines of treatment at a single university-affiliated tertiary medical center from 2019 to 2024. The final analysis included only those who received treatment for at least 1 month and achieved either a partial response (PR) or complete response (CR). Patients were divided into two groups. Those who discontinued treatment for at least 3 months were categorized into the “break” group, and patients who maintained treatment without any interruptions exceeding 2 months were placed in the "no-break" group. Patients with a treatment interruption of 2–3 months were excluded from both groups.
Clinical data
Data were collected from electronic medical records and included demographics, medical history, and response to PD-1 inhibitor treatment. Overall survival (OS) was defined as the time from the start of anti-PD-1 therapy to the last follow-up or death. Progression-free survival was defined as the time from start of anti-PD-1 therapy to time of progression, time of death, or last follow-up. Responses were classified as complete response (CR), partial response (PR), stable disease (SD), or progressive disease (PD) as classified by the oncologist, according to radiology reports. ORR was calculated as the percentage of patients who achieved CR or PR.
Statistical analysis
All statistical analyses were conducted using R software. Chi-square test and t-tests were employed to assess the relationships between various demographic characteristics, medical history, treatment patterns, and responses to therapy. To account for potential confounding variables, a propensity score matching (PSM) model was applied, adjusting for ischemic heart disease (IHD), diabetes mellitus (DM), chronic renal failure (CRF), chronic obstructive pulmonary disease (COPD), and immune suppression. The matched cohort was then used to evaluate the association between treatment break status and treatment response. Kaplan–Meier curves were used to evaluate survival. Follow-up time was calculated using the reverse Kaplan–Meier method. A p-value of < 0.05 was considered statistically significant. The study was approved by the Rabin Medical Center ethics board (IRB approval number RMC-21-0515), and informed consent was waived.
Results
Baseline patient characteristics
The cohort included 131 patients treated with cemiplimab or pembrolizumab for advanced stage cSCC. The average age at diagnosis for all patients was 78 years (range 51.4–98.7). 101 patients (77%) were male, and 30 (23%) were female. 35 (27%) of them were immune suppressed, and 25 (19%) patients had lesions that were either metastatic or of unknown primary origin. Data regarding stage were available for 88 patients in the cohort. Of them, 27 patients (30.7%) had stage IV disease, 20 patients (22.7%) had stage III, 25 patients (28.4%) had stage II, and 16 patients (18.2%) had stage I.
Response to PD-1 inhibitors
The overall response rate was 71.8%. Maximal response was CR in 46 patients (35.1%), 48 patients (36.6%) had a PR, 6 patients (4.6%) exhibited SD, and 16 patients (12.2%) showed PD as best response. 15 patients (11.5%) died before response assessment. Max response was not correlated with stage of disease (χ2 = 5.8, df = 12, p = 0.93). The full distribution is shown in Table 1 in the supplementary material. The median survival was 34.46 months, and median follow-up was 24.4 months. Median follow-up was 29.9 months in the break group and 22.6 months in the no-break group. The median PFS was 41.3 months in the break group and 17.6 months in the no-break group. Kaplan–Meier curves are shown in Fig. 2.
Table 1.
Baseline patient characteristics according to break and no-break groups
| No Break | Break | Total | ||
|---|---|---|---|---|
| Smoking (n = 86) | No | 32 (52.5%) | 29 (47.5%) | 61 |
| Yes | 14 (56%) | 11 (44%) | 25 | |
| ECOG (n = 83) | 0 | 9 (39.1%) | 14 (60.9%) | 23 |
| 1 | 16 (61.5%) | 10 (38.5%) | 26 | |
| 2 | 12 (60%) | 8 (40%) | 20 | |
| 3 | 6 (54.5%) | 5 (45.5%) | 11 | |
| 4 | 1 (33.3%) | 2 (66.7%) | 3 | |
| Gender (n = 86) | Male | 37 (53.6%) | 32 (46.4%) | 69 |
| Female | 9 (52.9%) | 8 (47.1%) | 17 | |
| DM (n = 86) | No | 24 (50%) | 24 (50%) | 48 |
| Yes | 22 (57.9%) | 16 (42.1%) | 38 | |
| IHD (n = 86) | No | 33 (54.1%) | 28 (45.9%) | 61 |
| Yes | 13 (42.6%) | 12 (57.4%) | 25 | |
| COPD (n = 86) | No | 41 (50.6%) | 40 (49.4%) | 81 |
| Yes | 5 (100%) | 0 (0%) | 5 | |
| CRF (n = 86) | No | 36 (55.4%) | 29 (44.6%) | 65 |
| Yes | 10 (47.6%) | 11 (52.4%) | 21 | |
| Cognitive (n = 86) | Normal | 35 (50.7%) | 34 (49.3%) | 69 |
| MCI | 4 (44.4%) | 5 (55.6%) | 9 | |
| Dementia | 7 (87.5%) | 1 (12.5%) | 8 | |
| Location (n = 86) | Head and Neck | 29 (55.8%) | 23 (44.2%) | 52 |
| Limbs | 6 (75%) | 2 (25%) | 8 | |
| Torso | 3 (50%) | 3 (50%) | 6 | |
| Systemic/Unknown | 8 (40%) | 12 (60%) | 20 |
Fig. 2.
Progression-free survival in the break and non-break groups
Out of 94 responding patients, 60 (63%) remained event-free and alive at the last follow-up. During the follow-up period, nine patients (10%) experienced PD. 13 (14%) died from causes unrelated to their oncological disease. Three patients (3.2%) died from treatment-related toxicity, and five patients (5.3%) died due to unknown reasons. For two (2.2%) patients treatment change duo to other cause then PD. Data were unavailable for two patients (2.2%).
Outcomes of treatment interruptions
A total of 118 patients received treatment for at least one month, with 91 (77%) achieving either PR or CR. Among these responders, 40 patients discontinued treatment for a minimum of 3 months (break group), while 46 maintained continuous treatment without any breaks exceeding two months to a maximal duration of two years (no-break group). Five patients had a treatment break of 2–3 months and were not included in the analysis. Flowchart is presented in Fig. 1. There were no differences in medical history between the groups regarding diabetes mellitus, ischemic heart disease, hypertension (HTN), chronic renal failure, smoking status, or cognitive status. Additionally, there were no significant differences in gender, location of primary tumor, or ECOG performance status. Chronic obstructive pulmonary disease (COPD) was more common in patients in the non-break group. Additional data in Table 1. After adjusting for IHD, DM, CRF, COPD, and immune suppression using a PSM model, the difference in CR rates between groups remained statistically significant. Specifically, 70% of patients in the break group achieved CR as the maximal response compared to 40% in the no-break group (χ2 = 6.93, p = 0.008).
Fig. 1.
Chart flow
In a subgroup analysis based on best response status, patients in the break group experienced a statistically significant prolongation of PFS. Among patients who achieved a maximal response of PR, the median PFS was 34.5 months in the break group compared to 15.7 months in the non-break group (p = 0.0088). For those with CR, the median PFS was 41.3 months in the break group compared to 24.4 months in the non-break group (p = 0.01).
Patients discontinued treatment for various reasons. The most common were non-specific fatigue (15 patients), immune-mediated toxicity (12 patients), completion of two years of treatment (8 patients), and non-oncological health issues (4 patients). For two patients, the reason was unknown, and one patient lacked reimbursement (Fig. 2).
Of the 40 patients in the break group, 24 (60%) remained progression-free, systemic treatment-free, and alive during the follow-up period. Four patients (10%) experienced disease progression. Among these, the maximal response was PR in three patients and CR in one patient. Nine (22.5%) patients died due to non-oncological reasons, two (5%) patients died from an unknown cause, and one (2.5%) due to treatment toxicity. Six (15%) patients resumed treatment: four due to PD and two due to unknown causes. Among the five patients with available response assessment, the ORR after resuming treatment was 80%. 30 (75%) patients experienced some grade of immune-mediated toxicity. A swimmer’s plot is presented in Fig. 3.
Fig. 3.
Swimmer plot of the 40 patients in the break group
In the no-break group, seven (15.2%) patients experienced PD during the follow-up period, and three patients died due to toxicity. 34 (73.9%) experienced some grade of immune-mediated toxicity. 11 patients died due to unknown causes, 14 died due to non-oncological reasons, five due to treatment-related toxicity, and six due to disease progression.
The main immune-related toxicities leading to discontinuation were dermatitis (5 patients) and pneumonitis (3 patients). Other immune-related toxicities that led to discontinuation in one patient each included thrombocytopenia, Raynaud syndrome, diabetes mellitus, myocarditis, and arthritis.
Discussion
This study describes our clinical experience with discontinuing PD-1 inhibitors in patients with advanced cSCC who have achieved a clinical response. These findings are particularly significant given the expanding use of immune checkpoint inhibitors (ICIs) in clinical practice for cSCC in both the palliative and neoadjuvant setting and the uncertainty surrounding the optimal duration of treatment [4, 9]. Given the basic characteristics of cSCC patients, different strategies for de-escalating treatment are of great importance. One example of de-escalation is the De-Squamate study, which demonstrates how neoadjuvant pembrolizumab can guide risk-adapted surgical and radiation therapy decisions based on clinical, radiological, and pathological responses [10].
The ORR in our cohort was 71.8%, with a control rate of 76.4%. Previous datasets have shown significant diversity in response patterns, with ORR ranging from 40 to 80% [11–13]. This variability may be attributed to factors such as differences in patient demographics, disease stages, primary sites, and stage in clinical course PD-1 inhibitors were employed.
The treatment break group comprised 40 patients, while the non-break group included 46 patients. It should be noted that only patients achieving a response to PD-1 inhibitors were included in the final two groups. The rationale behind this decision is that typically, the option to take a treatment break is considered only after a patient has shown a response to therapy. By focusing on this subset of patients, our aim was to ensure that the analysis mirrors the decisions made in real-world clinical practice, providing insights that are applicable to typical therapeutic scenarios. Additionally, it should be noted that only patients with at least a 3-month treatment break were included in the final analysis. This timeframe was chosen because it is commonly used in oncology clinical trials to effectively evaluate responses to treatment. To ensure this timeframe does not skew the outcomes, we summarized the clinical outcomes of the excluded patients in Table 1 in the supplementary material. None of these patients progressed until the last follow-up.
The clinical outcomes in the break group were remarkably positive, with a median OS of 41.3 months and 24 (60%) of patients remaining progression-free, not requiring systemic treatment, and alive throughout the follow-up period. Only four (10%) patients experienced disease progression and only for six patients treatment was resumed. Death due to disease progression was not observed, and there was only one death due to treatment toxicity. Notably, the PD as maximal response rate was significantly lower (15.2% vs. 10%), and PFS was markedly longer (41.3 vs. 17.6 months) in the break group.
It is important to emphasize that directly comparing outcomes between the break and non-break groups may be misleading. Typically, the decision to take a treatment break is a mutual choice between the primary oncologist and the patient. It is reasonable to assume that improved response characteristics, such as significant tumor shrinkage or symptom relief, can positively influence patients’ and clinicians’ perspectives, enhancing their confidence and willingness to consider a treatment break. This potential selection bias is evident in our study where 70% of patients in the break group had CR as their maximal response, compared to only 40% in the non-break group. Importantly, this difference remained statistically significant even after adjusting for key comorbidities—including IHD, DM, CRF, COPD, and immune suppression—using a PSM model, suggesting that the observed association is unlikely to be solely due to confounding.
This finding aligns with the previous research on the subject. Bailly-Caillé et al. reported that 66.7% of patients with advanced cSCC maintained their response two years after stopping cemiplimab [14]. Furthermore, a phase II clinical trial revealed that after receiving up to four doses of cemiplimab before surgery, 51% of patients achieved a complete pathological response, and 13% showed a major pathological response (≤ 10% viable tumor cells) [15]. An additional report found that in a group of patients who discontinued cemiplimab treatment, among those who discontinued after 12 months, only one (10%) patient died, suggesting that treatment for more than 1 year may be unnecessary [16]. These findings, together with our data, suggest that PD-1 inhibitors can rapidly induce significant tumor regression and that, for some patients, prolonged treatment may not be necessary.
One question arising from these conclusions is which parameters could help identify patients who will not be adversely affected by a treatment break. Among the four patients in the break group who progressed after discontinuing treatment, three (75%) had previously achieved only a PR and only one (25%) a CR response. Together with earlier observations of a high CR rate in the break group, these findings highlight maximal response as a potential parameter for patient selection. This finding is consistent with previous data on ICIs discontinuation in melanoma patients. Several real-world reports have shown that the likelihood of progression after discontinuation is significantly lower in patients achieving a CR [17–20].
However, given that the PFS remained significantly longer in the break group even in the sub-analysis by maximum response, it appears that additional parameters could help identify these patients. Another factor that can guide the decision to discontinue treatment is the pathological and histological characteristics of the tumor. Since tumor size, depth of invasion (DOI), and lymphovascular invasion (LVI) have been identified as predictive indicators of disease progression [21], these attributes may also affect the likelihood of progression following ICIs discontinuation. Unfortunately, this study did not uncover such parameters. Future research should aim to better characterize the patient population that would not be harmed by treatment discontinuation.
This report should be interpreted with caution due to several limitations. First, the retrospective design and lack of randomization limit our ability to establish causal relationships. Second, the response rates observed in our cohort were relatively favorable compared to previously published data, potentially limiting the generalizability of our findings to populations with less robust responses. Additionally, our analysis included only patients who achieved a response to PD-1 inhibitor therapy, further restricting the applicability of our conclusions to the broader population of patients with advanced cSCC. Lastly, pathological and molecular correlates that could help stratify patients for safe treatment discontinuation were not available in this dataset and should be prioritized in the future studies.
To conclude, our report supports the hypothesis that prolonged PD-1 inhibitor therapy may not provide additional benefit to patients suffering from cSCC who experience significant response. As it remains uncertain whether rechallenging therapy is as effective as continuous treatment, decisions regarding discontinuation should be made with caution and individualized based on clinical context and patient preferences. Future studies with larger cohorts are needed to better guide treatment decisions and optimize patient outcomes.
Supplementary Information
Below is the link to the electronic supplementary material.
Author contributions
Itamar Averbuch: Conceptualization, Methodology, Formal analysis, Writing—Original Draft, Visualization. Nofar Edri: Resources, Data Curation, Methodology Nethanel Asher: Funding acquisition, Supervision, Conceptualization, Methodology Gal Markel: Funding acquisition, Supervision, Conceptualization, Methodology Daniel Hendler: Resources, Data Curation. Hadas Ditzian Kugler: Methodology Eyal Yosefof: Resources, Data Curation, Conceptualization, Methodology Noga Kurman: Supervision, Conceptualization, Methodology, Writing—Original Draft, Funding acquisition.
Data availability
Data are available from the corresponding author upon reasonable request.
Declarations
Conflict of interest
Gal Markel reports receiving personal fees from MSD and Roche; grants and personal fees from BMS and Novartis; personal fees and stock options from 4C Biomed; and stock options from Nucleai, Biond Biologics, and Ella Therapeutics, outside the submitted work. Nethanel Asher reports receiving honoraria for professional services, including speaking engagements, lectures, or advisory board roles from MSD, Medison, and Taro. Additionally, Nethanel Asher is involved in an advisory board and steering committee for EUMelaReg. Itamar Averbuch, Nofar Edri, Daniel Hendler, Hadas Ditzian Kugler, Eyal Yosefof and Noga Kurman report no conflict of interest.
Footnotes
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Shalhout SZ, Kaufman HL, Emerick KS, Miller DM (2022) Immunotherapy for nonmelanoma skin cancer: facts and hopes. Clin Cancer Res 28(11):2211–20 [DOI] [PubMed] [Google Scholar]
- 2.McDowell L, Yom SS (2019) Locally advanced non-melanomatous skin cancer: Contemporary radiotherapeutic management. Oral Oncol 99:104443 [DOI] [PubMed] [Google Scholar]
- 3.Hughes BGM, Guminski A, Bowyer S, Migden MR, Schmults CD, Khushalani NI et al (2025) A phase 2 open-label study of cemiplimab in patients with advanced cutaneous squamous cell carcinoma (EMPOWER-CSCC-1): Final long-term analysis of groups 1, 2, and 3, and primary analysis of fixed-dose treatment group 6. J Am Acad Dermatol 92(1):68 [DOI] [PubMed] [Google Scholar]
- 4.Hughes BGM, Munoz-Couselo E, Mortier L, Bratland GR, Roshdy O et al (2021) Pembrolizumab for locally advanced and recurrent/metastatic cutaneous squamous cell carcinoma (KEYNOTE-629 study): an open-label nonrandomized multicenter phase II trial. Ann Oncol 32(10):1276–85 [DOI] [PubMed] [Google Scholar]
- 5.Averbuch I, Salman S, Shtamper N, Doweck I, Popovtzer A, Merkel G et al (2023) First-Line Programmed Death-1 Inhibitor Treatment for Locoregionally Advanced or Metastatic Cutaneous Squamous Cell Carcinoma – A Real-World Experience from Israel. Front Oncol 13:87 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Schmults CD, Blitzblau R, Aasi SZ, Alam M, Andersen JS, Baumann BC et al (2021) NCCN Guidelines® Insights: squamous cell skin cancer, version 1.2022. J Natl Compr Canc Netw 19(12):1382–94 [DOI] [PubMed] [Google Scholar]
- 7.Yakobson A, Abu Jama A, Abu Saleh O, Michlin R, Shalata W (2023) PD-1 inhibitors in elderly and immunocompromised patients with advanced or metastatic cutaneous squamous cell carcinoma. Cancers (Basel) 15(16):4041 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Averbuch I, Salman S, Shtamper N, Doweck I, Popovtzer A, Markel G et al (2023) First-line programmed death-1 inhibitor treatment for locoregionally advanced or metastatic cutaneous squamous cell carcinoma – A real-world experience from Israel. Front Oncol 13:1117804 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Migden MR, Khushalani NI, Chang ALS, Lewis KD, Schmults CD, Hernandez-Aya L et al (2020) Cemiplimab in locally advanced cutaneous squamous cell carcinoma: results from an open-label, phase 2, single-arm trial. Lancet Oncol 21(2):294–305 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Ladwa R, Lee JH, Porceddu SV, McGrath ML, Cooper C, Liu H et al (2024) A phase 2 study of de-escalation in resectable, locally advanced cutaneous squamous cell carcinoma (cSCC) with the use of neoadjuvant pembrolizumab: De-Squamate. J Clin Oncol 42:9514–9514 [Google Scholar]
- 11.Miodovnik M, Dolev Y, Buchen R, Brezis MR, Nikolaevski-Berlin A, Finkel I et al (2024) Squamous cell carcinoma arising in chronically damaged skin (Marjolin’s Ulcer): still an unmet need in the era of immunotherapy. Oncologist. 10.1093/oncolo/oyae326 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Salzmann M, Leiter U, Loquai C, Zimmer L, Ugurel S, Gutzmer R et al (2020) Programmed cell death protein 1 inhibitors in advanced cutaneous squamous cell carcinoma: real-world data of a retrospective, multicenter study. Eur J Cancer 138:125–32 [DOI] [PubMed] [Google Scholar]
- 13.Patrinely JR, Dewan AK, Johnson DB (2020) The role of anti-PD-1/PD-L1 in the treatment of skin cancer. BioDrugs 34(4):495–503 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Bailly-Caillé B, Levard R, Kottler D, Dompmartin A, L’Orphelin JM (2024) Long-term survival after anti-PD-1 discontinuation in advanced cutaneous squamous cell carcinoma (cSCC): a proof of concept of benefit of concomitant cemiplimab and radiotherapy. Cancer Immunol Immunother 73(7):118 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Gross ND, Miller DM, Khushalani NI, Divi V, Ruiz ES, Lipson EJ et al (2022) Neoadjuvant cemiplimab for stage II to IV cutaneous squamous-cell carcinoma. New Engl J Med. 10.1056/NEJMoa2209813 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Mallardo D, Sparano F, Vitale MG, Trojaniello C, Fordellone M, Cioli E et al (2024) Impact of cemiplimab treatment duration on clinical outcomes in advanced cutaneous squamous cell carcinoma. Cancer Immunol Immunother 73(8):1–10. 10.1007/s00262-024-03728-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Asher N, Israeli-Weller N, Shapira-Frommer R, Ben-Betzalel G, Schachter J, Meirson T et al (2021) Immunotherapy discontinuation in metastatic melanoma: lessons from real-life clinical experience. Cancers (Basel) 13(12):3074 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.van Zeijl MCT, van den Eertwegh AJM, Wouters MWJM, de Wreede LC, Aarts MJB, van den Berkmortel FWPJ et al (2022) Discontinuation of anti-PD-1 monotherapy in advanced melanoma-Outcomes of daily clinical practice. Int J Cancer 150(2):317–26 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Rubatto M, Fava P, Stanganelli I, Ribero S, Pigozzo J, Di Giacomo AM et al (2023) Discontinuation of anti-PD1 in advanced melanoma: an observational retrospective study from the Italian melanoma intergroup. Eur J Cancer 187:25–35 [DOI] [PubMed] [Google Scholar]
- 20.Jansen YJL, Rozeman EA, Mason R, Goldinger SM, Geukes Foppen MH, Hoejberg L et al (2019) Discontinuation of anti-PD-1 antibody therapy in the absence of disease progression or treatment limiting toxicity: clinical outcomes in advanced melanoma. Ann Oncol 30(7):1154–61 [DOI] [PubMed] [Google Scholar]
- 21.Roccuzzo G, Orlando G, Rumore MR, Morrone A, Fruttero E, Caliendo V et al (2024) Predictors of recurrence and progression in poorly differentiated cutaneous squamous cell carcinomas: insights from a real-life experience. Dermatology 240(2):329–36 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data are available from the corresponding author upon reasonable request.