Exploration of genetic characterization in hyperprogressive disease after immunotherapy retreatment in a patient with LCNEC: A case report

ABSTRACT

Immune checkpoint inhibitors (ICIs) have emerged as a promising therapeutic option for large cell neuroendocrine carcinoma (LCNEC). However, various studies have suggested a potential risk of hyperprogressive disease (HPD) in patients receiving ICI, which might be associated with gene alterations. Here, this is the first report on an unknown primary LCNEC patient who had achieved a long-term response from ICI treatment (atezolizumab), but developed HPD after tumor progression due to receiving another ICI agent (serplulimab). The mutation region of FAT4, SMARCA4, CYLD, CTNNB1, and KIT was altered prior to serplulimab treatment compared to before atezolizumab treatment. This case suggested a potential association between these mutated genes and HPD. Patients with the aforementioned genes should caution when selecting ICI treatment. These findings required further confirmation in a larger study cohort.

KEYWORDS:

• Large cell neuroendocrine carcinoma (LCNEC)
• immune checkpoint inhibitors (ICIs)
• retreatment with immunotherapy
• hyperprogressive disease (HPD)
• genetic characterization

Introduction

Large cell neuroendocrined carcinoma (LCNEC) is a rare and aggressive tumor with a short median survival time mostly because of limited effective therapies.¹ Although immune checkpoint inhibitors (ICIs) have revolutionized the treatment paradigm of several solid tumors, to date, studies on ICI treatment in LCNEC are nascent.^2,3

After ICI treatment, a small subset of patients experience a rapid type of tumor growth known as hyperprogressive disease (HPD).⁴ To diagnose HPD, at least three of the following criteria must be met: 1) the time from ICI treatment to tumor progression is less than 2 months; 2) tumor volume has increased by more than 50% compared to baseline; 3) the tumor growth kinetics ratio (TGKr) is greater than 2.⁵

Several studies have indicated that HPD is associated with various factors, including older age, poor performance status, higher tumor burden, as well as amplification of EGFR or MDM2/4, and immune-cell biomarkers.⁶ It is worth noting that patients older than 65 are more susceptible to developing HPD, and those with poor performance status (PS) are at a higher risk.^7,8 The mechanism connecting HPD and advanced age may be due to the immunosenescence, which causes a reduction generated by age-related thymic atrophy.⁹ A prospective study found a significant association between HPD status and a higher metabolic tumor burden at baseline, which negatively affected tumor response and survival, especially in cases where peripheral blood T-cell reinvigoration was insufficient.⁸ It has been suggested that MDM2/4 may contribute to HPD by inhibiting gene transactivation of the tumor suppressor p53, thereby promoting tumor growth.¹⁰ However, there is no clear consensus as to the association between HPD and the above-mentioned factors. Therefore, in this study, we aimed to analyze the characterization of somatic gene alterations in patients with unknown primary LCNEC who underwent initial ICI treatment with a long progression-free survival (PFS) but HPD occurred after retreatment with another ICI agent.

Case presentation

A 65-year-old man was diagnosed with an unknown primary LCNEC presenting with isolated bone metastases (fourth lumbar) on February 19, 2021. The patient underwent a resection of the fourth lumbar spine metastasis, and the tissue sections of bone metastases revealed that the positive rates of PDL1 (TPS) were approximately 30% by Immunohistochemistry (IHC). Patients with higher expression of immune checkpoint markers (such as PDL1) might be more responsive to immunotherapy.¹¹ Therefore, the patient received atezolizumab 1200 mg plus chemotherapy (etoposide and platinum) every 3 weeks for six cycles followed by maintenance atezolizumab. Meanwhile, stereotactic radiotherapy (20 Gy/10 Fr) was performed on the L4 vertebrae to control local bone metastases. CT reexamination showed good changes in the bone metastases, and the patient had a stable disease (SD) according to RECIST criteria 1.1 during March 2021 to August 2022.

After 18-months of maintenance treatment, the patient had recurrence in the fourth lumbar with psoas muscle invasion, and metastatic lesions in the common iliac nodes and the retroperitoneal lymph nodes. The patient then received atezolizumab and bevacizumab as second-line treatment, but discontinued therapy after three cycles due to a grade 3 liver injury. Meanwhile, spinal MRI demonstrated an increased size of the L4 and psoas muscle lesion over 2 months after drug interruption, which placed the patient at the risk of paraplegia. Therefore, the combination of palliative radiation to the psoas muscle sites and L4 replacement surgery was performed . The surgical specimens provided histological reconfirmation of metastatic LCNEC with a high tumor mutation burden (19.4 Muts/Mb).

On April 26, 2023, 4 weeks after surgery, the patient showed good postoperative recovery and started receiving serplulimab (200 mg) plus sovantinib (250 mg). After one cycle of the combination treatment, the patient presented with fever, fatigue, upper quadrant pain, and lower back pain, which gradually became aggravated. After 4 weeks, local MRI revealed an increase in the number and volume of metastases, and the appearance of new metastatic sites in the liver and vertebra, as well as the enlarged metastasis in the right adrenal gland (Figure 1a), compared with baseline imaging of PETCT assessed 5 weeks prior. The size of metastatic tumor in the right adrenal gland increased by 145% compared with baseline imaging; the progressive pace was 3.4-fold higher than before serplulimab treatment (Figure 1b). Additionally, neuron-specific enolase (NSE) antigens, before and after serplulimab, were 7.57 and 139 ng/ml, respectively (Figure 1c). Due to the hyperprogression of the disease, the discontinuation of serplulimab was decided. Unfortunately, the patient’s performance status deteriorated rapidly, and died of respiratory failure on the 47th day after administration of serplulimab. The overall treatment process diagram is shown in Figure 1d.

Figure 1. Case study of a LCNEC patient with HPD during immunotherapy.

(a) CT scans were performed 15 weeks before starting serplulimab treatment (column 1), at baseline (PETCT scans were performed 5 weeks before starting serplulimab, column 2), and at first evaluation (MRI scans were performed 4 weeks after starting serplulimab, column 3). Lines 1 and 2 revealed the changes in right adrenal gland and liver, respectively. New liver lesions appeared. The red arrows indicate tumor lesions. (b) Rate of change in growth pattern in the patient, who developed HPD to serplulimab. Compared with the tumor image (−15 weeks), the tumor lesions at baseline (−5 weeks) and at first evaluation (4 weeks after starting immunotherapy) showed approximately 33% and 227% increases (145% increase compared with baseline imaging), respectively; 3.4-fold increase in progressive pace compared with pre-serplulimab. (c) Changes in tumor-associated antigens before and after serplulimab. NSE, neuron-specific enolase. (d) Patient treatment chart.

Profiling of genomic characteristic

Although the patient had achieved a long-term response from atezolizumab (an anti-PD-L1 antibody) with a PFS of 2 years, HPD still occurred after the retreatment with immunotherapy, suggesting that gene changes during tumor progression may potentially participate in HPD. Thus, we compared genomic characteristics of tumor tissue specimens before atezolizumab and serplulimab treatment by next-generation sequencing (NGS), respectively (Supplemental Table 1). NGS mainly covered somatic and germline variants, tumor mutation burden (TMB), microsatellite instability (MSI), PD-L1 protein expression, human leukocyte antigen (HLA)-class I molecular genotyping, and HLA-I evolutionary divergence (HED) score. The NGS panels included a list of 674 and 708 cancer-related genes, respectively (Supplemental Tables 2 and 3). As shown in Table 1, before the atezolizumab treatment, the patient had single-nucleotide variants (SNVs) in FAT4 (exon1: R11G), SMARCA4 (exon19: V855D, E861*), and CYLD (exon10: M462I). Before serplulimab treatment, the mutation region of exons was changed in the above-mentioned three genes, and an additional two different genes were also identified, including CTNNB1 (exon10: V541L) and KIT (exon9: N463S). The latest studies and clinical evidence indicated that upregulation of the Wnt/β-catenin signaling pathway might participate in the immune checkpoint blockade (ICB)-associated HPD.¹² Previous studies also showed that SMARCA4, CTNNB1, and KIT could positively regulate Wnt/β-catenin signaling pathway activity, while FAT4 and CYLD negatively regulated it.^13–23 Those observations led us to speculate that the genes of FAT4, SMARCA4, CYLD, CTNNB, and KIT were involved in HPD.

Table 1. Genomic alterations in specimens before atezolizumab and serplulimab.

Timelines	Before atezolizumab (First surgery)	Before serplulimab (Second surgery)
gene	Alteration type	Alteration type
FAT4	exon1, R11G, SNV	exon2, R11G, SNV
SMARCA4	exon19, V855D, E861*, SNV	exon18, V855D, SNV
		exon18, E861* exon18 SNV
CYLD	exon10, M462I, SNV	exon9, M462I, SNV
CTNNB1		exon10, V541L, SNV
KIT		exon9, N463S, SNV

Discussion

In this case, a 65-year-old man was diagnosed with an unknown primary LCNEC presenting with isolated bone metastases. After failure of the first-line treatment with atezolizumab plus chemotherapy followed by maintenance atezolizumab and the  second-line treatment with atezolizumab plus bevacizumab, the patient received serplulimab plus sovantinib as a third-line treatment. After 4 weeks of immune therapy, the size of metastatic tumor in the right adrenal gland increased by 145% compared with baseline imaging and the progressive pace was 3.4-fold higher than before serplulimab treatment, which fully met the diagnostic criteria of HPD. The mutation region of FAT4, SMARCA4, CYLD, CTNNB, and KIT was changed before serplulimab treatment compared to atezolizumab treatment and might be associated with HPD via Wnt/β-catenin signaling pathways, which was never reported before.

HPD has been associated with a worse survival prognosis.²⁴ However, in our case, the patient survived for 28 months, which is significantly longer than the clinical median overall survival (mOS) of approximately 8–12 months.²⁵ The survival benefit may be attributed to initial immunotherapy of atezolizumab and the future of ICI therapy seems to be bright in LCNEC. The patient died unexpectedly less than 2 months after receiving serplulimab treatment. It was believed that this was due to the onset of HPD. So it is essential to correctly identify clinical or molecular factors associated with HPD and to promptly prevent adverse treatment outcomes.

To date, some clinical indicators have been found to be associated with HPD including older age, poor performance, and higher tumor burden.²⁵ In our case, TMB increased from 11.7 to 19.4 Muts/Mb before atezolizumab and serplulimab treatment, respectively, which may be the potential biomarker associated with HPD.

Furthermore, it has been suggested that genomic alterations such as MDM2/MDM4 and EGFR amplifications are significantly correlated with HPD.²⁶ To compare genomic characteristics of tumor tissue specimens before atezolizumab and serplulimab treatment, we conducted next-generation sequencing. However, we did not find any alterations in MDM2/4 or EGFR amplification in our case. It appears that HPD still requires further exploration and investigation.

Recent studies have shown that HPD may be related to the acetylation and activation of the β-catenin signaling pathway.¹² The activation of the Wnt/β signaling pathway can facilitate resistance to ICIs by promoting immune evasion and hindering T-cell mediated anti-tumor responses.²⁷ A study on the activation of the Wnt/β-catenin signaling pathway and immune exclusion found an inverse correlation between Wnt/β-catenin levels and the T-cell-inflamed phenotype, which is correlated with the efficacy of ICB.²⁸ Furthermore, some studies have shown that the activation of Wnt-β-catenin signaling pathway was associated with immune escape in mouse models and had a negative impact on PFS with immunotherapies.²⁹ Previous studies have shown that certain gene mutations such as FAT4, SMARCA4, CYLD, CTNNB1, and KIT, which were observed in our case, affect the Wnt/β-catenin signaling pathways.^13–23 Brg1/SMARCA4, CTNNB1, and KIT actively regulate Wnt/β-catenin signaling. SMARCA4 recruit lysine demethylase 4 (KDM4) to activate β-catenin target genes to potentiate Wnt/β-catenin pathway.^16,17 Activation of KIT promotes β-catenin nucleation to increase Wnt/β-catenin signaling pathway.^22,23 Mutations in CTNNB1 disrupt the β-catenin degradation domain and lead to β-catenin stabilization and translocation into the nucleus, activating the β-catenin pathway.^20,21 FAT4 and CYLD negatively regulate Wnt/β-catenin signaling. FAT4 is a 543-kD protein and usually acts as a tumor suppressor.^13,14 FAT4 downregulation promotes the activation of Wnt/β-catenin signaling.¹⁵ CYLD regulates Wnt/β-catenin signaling negatively by promoting deubiquitination of Dishevel (Dvl).^18,19 As described in the literature, the high β-catenin signature is a potential biomarker and target for HPD.¹² So we speculate that alterations in FAT4, SMARCA4, CYLD, CTNNB1, and KIT, which activate Wnt/β-catenin signaling pathway, may also be the potential predictive factors for the development of HPD.

HPD is a novel pattern of progression characterized by an unexpected rapid acceleration of tumor growth, occurring in a subset of patients after initial ICI treatment.³⁰ This case report describes the first instance of an LCNEC patient who achieved a superior therapeutic response from an initial ICI treatment, but developed HPD after the retreatment with another ICI agent. In a case reported in the literature, a patient with metastatic breast cancer developed HPD after treatment with pembrolizumab.³¹ However, after retreatment with atezolizumab, a partial response was observed. This suggested a potential association between the sequence and interactions of PD-1 and PD-L1 inhibitors, and the occurrence of HPD. Although there is a lack of conclusive data from prospective trials to support the risks of PD-1 and PD-L1, this case might be a clinical reference as a relevant addition to the risk factors library of HPD.

Conclusion

This is the first report on an LCNEC patient who had achieved a long-term response from ICI treatment, but developed HPD after tumor progression due to receiving another ICI agent. This case suggested that genes of FAT4, SMARCA4, CYLD, CTNNB, and KIT might be associated with HPD. Patients with the above-mentioned genes should be more careful in choosing ICI treatment. These findings should be further confirmed in a larger study cohort.

Author contributions

L Shou participated in the clinical management of the patient and conception and edited the manuscript. Q Shu participated in the clinical management of the patient and provided the funding. Y Zhang participated in the clinical management of the patient, manuscript writing, and revision. J Yang contributed to the literature review and illustration. T Shao contributed to the literature review and illustration. J Chen contributed to the illustration. All authors reviewed the manuscript.

Informed consent statement

Informed consent was obtained from the subject involved in the study.

Institutional review board statement

This study was conducted in accordance with the principles of the Declaration of Helsinki and the Patient provided written informed consent. IRB approval is not required at our institution for case reports.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

Data sharing is not applicable to this article.

Supplementary material

Supplemental data for this article can be accessed on the publisher’s website at https://doi.org/10.1080/21645515.2024.2313281.

Additional information

Funding

This work was supported by the Projects of Department of Science and Technology of National Administration of Traditional Chinese Medicine and Administration of Traditional Chinese Medicine of Zhejiang Province (GZY-ZJ-KJ-23012).