Cancers | Free Full-Text | Anti-PD-1 Therapy in Advanced Pediatric Malignancies in Nationwide Study: Good Outcome in Skin Melanoma and Hodgkin Lymphoma

Every year, approximately 400,000 cases of cancer are diagnosed in children and adolescents worldwide. Over the years, significant progress has been made in therapy. The current effectiveness of anti-cancer therapy is estimated to exceed 80% [1]. However, there are still malignancies in which the efficacy of therapy remains unsatisfactory. This category includes malignant bone tumors, high- and very-high-risk soft tissue sarcomas, selected brain tumors, and high-risk neuroblastoma [2,3]. Additionally, a challenge in therapy involves patients who do not respond to treatment, as well as those with disease progression and recurrence. Hence, there is a continuous need to explore new therapeutic options, with immunotherapy being one of them.

The success of immunotherapy aimed at blocking immune checkpoints has changed the outlook on the treatment of selected adult cancers, namely, malignant melanoma, non-small cell lung cancer, Hodgkin lymphoma, clear-cell renal cell carcinoma, and others. Currently, six immunotherapeutics that block the key pathway of inhibiting the activity of T lymphocytes are approved by the FDA and registered by the EMA: anti-PD-1 (programmed death 1) monoclonal antibodies—nivolumab, pembrolizumab, and cemiplimab; and anti-PD-L1 (programmed death ligand 1) antibodies—atezolizumab, durvalumab, and avelumab. Such therapies function by augmenting existing anti-tumor T-cell responses that have been rendered ineffective by inhibitory pathways. The interaction of PD-1 and PD-L1 molecules leads to the extinction of the anti-cancer activity of T lymphocytes. The idea of immunotherapy using antibodies against immune checkpoints is based on blocking one of these molecules, which restores the cytotoxic activity of T lymphocytes [4].

The role of immune checkpoint inhibitors (ICIs) in the treatment of childhood cancers is still evolving. Single phase-I/II trials published in 2020–2022, where pembrolizumab (NCT0233266849), nivolumab (NCT0230445848), avelumab (NCT0345182551), and atezolizumab (NCT0254160450) were evaluated as monotherapies against recurrent and refractory childhood cancers, supply the most data. ICIs have been most advantageous in the therapy of classical Hodgkin lymphoma (HL) in children. As in adults, inhibitors of PD-1/PD-L1 have resulted in objective response rates of 30–60% in children and adolescents with relapsed HL. Similar effectiveness has not been observed in other types of pediatric cancers, where only about 3% of patients experienced an objective response. No response was observed in solid tumors, such as central nervous system tumors, neuroblastoma, rhabdomyosarcoma, Ewing sarcoma, and osteosarcoma. This ineffectiveness of the therapy probably represents a principal difference in tumor immunobiology in children compared to adults. Childhood tumors are commonly regarded as “cold cancers”, with a low mutational load and limited infiltrating T lymphocytes [4,5].

Davis et al. presented the results of a multicenter phase-I–II study conducted in 23 hospitals in the USA, wherein they assessed the effectiveness of nivolumab in 72 patients, aged 1–30 years, diagnosed with the following cancers: melanoma, neuroblastoma, rhabdomyosarcoma, osteosarcoma, Ewing sarcoma, Hodgkin lymphoma, and non-Hodgkin lymphoma (NCT02304458). Responses were observed in individuals with Hodgkin lymphoma (30%; 3/10) and with non-Hodgkin lymphoma (10%; 1/10). All responders with objective responses (OR) had PD-L1 expression confirmed. OR were not observed in other tumor types. Mild toxicities were observed in most patients. The most common overall toxicities were anemia (47%) and fatigue (37%). Dose-limiting toxicities were present in 7% of patients [8].

In a study reported by Davis et al., the effectiveness of ICIs in the treatment of cutaneous malignant melanoma has not been demonstrated, but it should be noted that only one analyzed patient (1/70) was diagnosed with melanoma [8]. Single clinical trials assessing the effects of immunotherapy and other drugs in the treatment of malignant melanoma in patients under 17 years of age failed due to low enrollment (NCT01696045, NCT01519323). Therefore, there are limited options for adjuvant therapy among pediatric patients with advanced melanoma [9]. We included nine patients with cutaneous malignant melanoma in our analysis. In seven/nine cases where anti-PD1 drugs were used as monotherapy, CR was achieved with a median OS of 30.6 months. Most of them were patients > 14 years of age for whom the biology of the tumors is probably proportional to that in the adult population, where perhaps the success achieved can be seen.

KEYNOTE-051 was a phase-I/II open-label trial where 154 children with melanoma or a PD-L1-positive, refractory, or relapsed solid tumor or lymphoma were recruited (NCT02332668). All participants, with the median age 13 years (range = 8–15), were treated with pembrolizumab. In 45% (69/154) of patients, grade 3–5 adverse events were observed, most commonly anemia (9%) and lymphopenia (6%); 3% (4/154) of individuals discontinued treatment due to serious adverse events; and 1% (2/154) of patients died (1 due to pleural effusion and pneumonitis and 1 due to pulmonary oedema). Of 15 children with refractory or relapsed Hodgkin lymphoma, 2 had complete responses (CR) and 7 had partial responses (PR). Thus, 60% (9/15) patients achieved an objective response (95% CI 32.3–83.7). Of 136 individuals with solid tumors and other lymphomas, 8 had partial responses; the diagnoses in these participants were adrenocortical carcinoma, mesothelioma, malignant ganglioglioma, epithelioid sarcoma, lymphoepithelial carcinoma, and malignant rhabdoid tumor. The proportion of children with an objective response was 5.9% (95% CI 2.6–11.3) [10,11].

Front-line anthracycline-based chemotherapy platforms are associated with the most robust outcome for patients with classical Hodgkin lymphoma [12]. As salvage therapy, the combination of brentuximab vedotin plus bendamustin is a valid option for patients with c-HL with age ≥ 18 years [13]. CheckMate 744 was a phase-II study, where 44 patients with a median age 16 years (range = 5–30) with relapsed/refractory classical Hodgkin lymphoma without complete metabolic response before autologous hematopoietic cell transplantation (auto-HCT) were evaluated for a risk-stratified, response-adapted approach with nivolumab plus brentuximab vedotin (NCT02927769). Patients received four induction cycles with nivolumab plus brentuximab vedotin; those without complete metabolic response (Deauville score > 3) received brentuximab vedotin plus bendamustine intensification. Complete metabolic response rate was observed in 59% of individuals after induction with nivolumab plus brentuximab vedotin [14,15,16]. Sun et al. performed a systematic review and meta-analysis of 20 prospective studies assessing the effectiveness and safety of PD-1 and PD-L1 inhibitors in relapsed and refractory Hodgkin lymphoma. The authors analyzed a total of 20 studies involving 1440 adult patients of which 19 studies concerned treatment with PD-1 inhibitors and 1 study with a PD-L1 inhibitor. The pooled ORR for the 19 studies in which it was an indicator of final effectiveness was 79% (95% CI 73–85). The pooled CR rate for the 20 studies was 44% (95% CI 34–55). PR was analyzed in the 19 included studies, and the pooled PR rate was 34% (95% CI 26–42) [5,17]. The conclusions from the available literature are consistent with the treatment results achieved in our study group among individuals with relapsed and refractory Hodgkin lymphoma. There is a high probability that the observation related to a better treatment effect in the group of older patients (>14 years of age) correlates with histopathological diagnoses more typical for this age (nodular sclerosis Hodgkin lymphoma and malignant melanoma of the skin).

Neuroblastoma is the most common extracranial solid tumor in children, and about 50% of patients have metastatic or refractory disease. The prognosis has partially improved after adding the anti-disialoganglioside antibody dinutuximab beta to the multimodal therapy, but it is still unsatisfactory. Preclinical studies in mouse models showed that dinutuximab beta resulted in an upregulation of the PD-1 checkpoint in neuroblastoma cell lines, and combined treatment with this antibody and an anti-PD1 drug in the tested mice showed a synergistic effect [18,19]. Ehlert et al. presented two cases of patients, a 4-year-old female and 17-year-old male with refractory metastatic neuroblastoma, for whom the combination of dinutuximab beta with nivolumab led to a complete and a very good partial remission [20]. The effects of ICIs in our group of patients with neuroblastoma were poor, although complete remission was achieved in two/eight individuals. The older age of the patients (untypical for neuroblastoma) who responded well to treatment is noteworthy.

Cacciotti et al. conducted a retrospective review of data on 11 pediatric patients with relapsed or refractory CNS tumors treated with ICIs at Boston Children’s Hospital in 2018–2019. Diagnoses included high-grade glioma (n = 5), diffuse intrinsic pontine glioma (DIPG) (n = 2), craniopharyngioma (n = 1), ependymoma (n = 1), non-germinomatous germ cell tumor (n = 1), and high-grade neuroepithelial tumor (n = 1). In the entire group of patients, nine participants received combination therapy with ipilimumab and nivolumab, and two patients received pembrolizumab or nivolumab as monotherapy. All individuals had previously undergone radiotherapy. The median duration of immunotherapy was 6.1 months (range = 1–25). In seven patients, therapy was discontinued due to disease progression, and in two participants, unacceptable toxicities occurred (hypertransaminasemia and colitis). Responses observed included partial response (n = 3), stable disease (n = 7), and progressive disease (n = 1), with a durable response observed in two patients [21].

In another retrospective analysis, treatment with nivolumab outcome was assessed. The study group consisted of 10 children, aged 2 to 17 years, with recurrent or refractory brain tumors like high-grade glioma (n = 5), low-grade glioma (n = 1), pineoblastoma (n = 1), medulloblastoma (n = 1), ependymoma (n = 1), and CNS embryonal tumor (n = 1) treated at Rady Children’s Hospital San Diego from 2015 to 2017. In three cases (two with high-grade glioma and one with CNS embryonal tumor), a partial response to treatment at the primary tumor site was observed. Median survival for PD-L1-positive individuals was 13.7 weeks versus 4.2 weeks for PD-L1-negative patients (ρ = 0.08). The authors suggest that the use of ICIs in pediatric brain tumor patients should be limited to those with elevated PD-L1 expression [22].

The use of ICIs as a last-chance treatment in our patients with other advanced cancers (Table 4) was usually supported by single case reports or studies on small groups of patients in which this type of treatment proved effective [23,24,25,26,27,28,29,30,31,32,33]. We acknowledge several limitations in our study, including the retrospective nature of the study, the diversity of the types of diseases, and the age of the patients in the cohort group. However, our data contribute significant information that might aid in the development of prospective validation studies.

Adverse events were reported in 30.9% (13/42) of patients in our study group, of which just over half (7/13) had CTCAE grade 3/4 toxicities. Compared to the above-mentioned prospective studies using ICIs in pediatric patients [8,10], the frequency of side effects in our patients was significantly lower. In four clinical trials, where nivolumab (NCT0230445848), pembrolizumab (NCT0233266849), atezolizumab (NCT0254160450), and avelumab (NCT0345182551) were tested as monotherapies against refractory and recurrent pediatric malignancies, usually mild systemic effects (fatigue and fever, grade 1/2) were observed, and the most common immune-related adverse event was hepatic toxicity (elevated transaminases, grade 1/2) [4]. There is a risk associated with the retrospective nature of our analysis, and it is possible that mild side effects such as fatigue, fever, anemia, or elevated liver function tests that did not require therapeutic interventions were missed. It should be emphasized that both our data and the data collected from the experience of other authors indicate that ICIs are well tolerated in the group of pediatric patients, and severe immunological toxicities are observed significantly less often than in adults. However, this is not necessarily a positive. Numerous data suggest that the increased occurrence of immunological toxicities correlates with the effectiveness of the therapy [34]. We observed this phenomenon in one of our patients with recurrent melanoma, in whom after seven doses of pembrolizumab, immunological side effects in the form of arthritis and SIRS forced the premature termination of immunotherapy. Despite significantly shorter treatment than planned, the child achieved 5-year PFS.