Multiple myeloma (MM) is a heterogeneous disease, with an estimated 34 920 newly diagnosed cases in the United States in 2021 according to the American Cancer Society. Long-term survival has improved with the development of proteasome inhibitors (PI), immunomodulatory drugs (IMiDs), and monoclonal antibodies with overall survival (OS) exceeding 10 years for patients with standard-risk MM. However, outcomes for the 25%-45% of patients characterized as high-risk are poor, with a median OS of 18–24 months.^{1, 2} Relapse after autologous stem cell transplantation (ASCT) is inevitable with or without lenalidomide (len) maintenance.² Immune dysfunction in MM is, in part, due to functionally defective dendritic cells, increased T_H17 cells, myeloid devoid suppressor cells (MDSCs) and tumor-associated macrophages (TAMs) in the BM and peripheral blood. Since T-cell clonal expansion after ASCT has been associated with improved clinical outcomes,³ we hypothesized that by incorporating a checkpoint inhibitor into post-ASCT consolidation, it will reinvigorate a T-cell inflamed phenotype to target residual MM. We used len which has shown to enhance T cell and NK cell effector anti-MM responses, reducing PD-L1 expression on MM cells⁴ and combine it with pembrolizumab, a PD-1 inhibitor.

We used the combination of pembrolizumab (pembro) plus len and dexamethasone (dex) which has previously shown an acceptable safety profile and improved efficacy in previous studies⁵ and intended to enroll 58 patients to our an open-label, phase II single-arm trial (NCT02906332). However, all studies evaluating the combination of PD-1 or PDL-1 inhibitors with IMiDs were halted by the FDA on July 3, 2017 due to a higher incidence of immune-related toxicities and deaths in the pembro arm. At the time of the FDA hold, 12 subjects were enrolled. Follow-up was permitted.

Study procedures were as follows: Patients were enrolled between days +60 and + 180 following ASCT and received treatment with pembro-len-dex (pembro-rd) for a total of four cycles. The combination included pembro 200 mg IV on day 1; len 25 mg p.o. daily on days 1–14; and dex 40 mg daily at days 1, 8, 15 of a 21-day cycle for a total of two cycles and then an additional two cycles of pembro-rd at the same dose and frequency. After the completion of cycle four, each subject was followed for 30 days. Serious adverse events were collected for 90 days after the end of treatment. Subjects who discontinued treatment for reasons other than disease progression continued post-treatment follow-up. Participants included adult patients with measurable disease and hrMM defined by any of the following: ISS stage 3; del 13q by cytogenetics; FISH with 1q amplification, 1p deletion (del), p53 del, t (4;14), t (14;16), t (14;20), hypodiploidy; or a high-risk gene expression profile score. Patients were excluded if they had progression of disease at time of screening or if there was evidence of organ dysfunction. Minimal Residual Disease (MRD) assessments were performed 30 days after the fourth cycle for those achieving VGPR or better.

Peripheral blood samples were analyzed at screening, at cycle two, day 1 of treatment and at long-term follow-up. Cells were processed by Ficoll-Paque Plus (Fisher Scientific) separation. A portion of the cells underwent immune phenotyping by fluorescent-labeled monoclonal antibody (mAb) staining for T-cell subsets. This included mAb purchased from BD Biosciences for: CD3-Alexa Fluor 647 (#557706), CD4-APC-Cy7 (#557871), CD45RO-FITC (#555492), CD25-PerCP-Cy5.5 (#560503), CD127-FITC (#560549), CD8-PE (#555635), CCR7-PC7 (#557648), and matched fluorescent labeled isotype control mAb.

The intent-to-treat population consisted of all the enrolled patients who received at least one dose of study drug and was the basis for the analysis of efficacy endpoints. Demographic and clinical variables were summarized using median (IQR) for continuous variables and counts (percentages) for categorical variables. The Kaplan–Meier method was used to determine median PFS and OS, respectively. Responses to treatment by each cycle were descriptively summarized. Both AE and SAEs were tabulated descriptively. For T-cell subset phenotypic analysis, paired continuous data were compared using Mood's median test. A two-sided p < 0.05 was considered statistically significant. Data analysis was performed using R (ver. 4.0.2).

Results were as follows: Of the 12 patients, five (41.7%) were ISS 3, six (50%) had a p53 deletion by FISH, three (25%) had 1q21 gain, and two (17%) were considered ultra-high risk with combination of 1q21 gain and 13q deletion by FISH (Table S1). All patients received triplet or quadruplet induction. Five patients (42%) received induction bortezomib-lenalidomide-dex; four (33%) carfilzomib-lenalidomide-dex, two (24.9%) bortezomib-cyclophosphamide-dex and one bortezomib-cyclophosphamide-len-dex.

Median follow-up was 50 months. The median PFS from the date of ASCT for all patients with a median follow-up of 50.7 months (range: 6–55.5 months) was 27.7 months (95% CI: 22.6 months - not reached, Figure S1). The PFS rates at 1 and 2-year follow-up was 90.9% and 63.6%, respectively. Overall response rate was 100% at time of best response (Figure S2). The rate of stringent complete remission was 33% at the end of induction and 83.3% at the end of consolidation and 83.3% at the pre-specified 7th follow-up point (a median of 19 months after completion of treatment) (Figure 1). For all patients, over the 2-year follow-up period, 11/12 (91.7%) achieved a complete remission or better, and 10/12 (83.3%) achieved stringent complete remission (Table S2). Of the 11 patients who completed therapy, eight had MRD status assessed and among them, seven (87.5%) were MRD negative by multiparametric flow cytometry with a cut-off value of 10⁻⁵.

Adverse events (AEs) were monitored at every visit and graded according to the guidelines outlined in the NCI Common Terminology Criteria for Adverse Events (CTCAE) version 4.03. All patients had one or more AEs attributed to pem, len or dex. Of the 97 AEs reported, 6.2% were grade 3 and 93.8% were grade 1 or 2. The most common non-hematologic AEs were respiratory (cough) and gastrointestinal (diarrhea and constipation), all grade 1 or 2, while the most common hematologic AE was neutropenia with six occurrences observed among two patients, grade 1–3. Immune-mediated AEs included diarrhea in four patients, fever in one patient, maculopapular rash in one patient, and dyspnea in two patients. All were grade 1 or 2. There was one serious AE, H. influenza pneumonia requiring inpatient admission, which was unrelated to pembrolizumab (Table S3).

In comparison to the C1D1 start of the first treatment cycle, the median percent effector memory CD4+ T cells within the CD3 compartment decreased steadily through C4D1 (p = 0.037). The effector memory CD8 compartment in the CD3 population exhibited a small decrease at the time of the second cycle (C2D1) (p = 0.032) and returned to insignificantly increased levels by the third cycle (p = 0.199). The median percentage Memory regulator T cells was not significantly decreasing through the second cycle (C2D1) and third cycle (C3D1) (p = 1.0, p = 0.199, respectively) (Table S4).

Our results suggest that PD-1 blockade in combination with len, given as post-ASCT consolidation, was well tolerated and efficacious in 12 hrMM patients with a median PFS of 2.3 years, exceeding So, PFS in hrMM patients without len maintenance from historical controls.⁶ Immune mediated AEs were only grade 1 or 2 while the one SAE observed was unrelated to pembro.

Lymphocyte composition and function after ASCT guided optimal timing of immunotherapy and helped identifying potential markers of relapse. Regulatory T cells (Treg) decline as CD8(+) T cells expand during early lymphocyte recovery after ASCT, markedly reducing the Treg:CD8(+) effector T-cell ratio. These CD8(+) T cells can respond to autologous dendritic cells presenting tumor antigen in vitro as early as day +12 after transplant, becoming antigen-specific cytolytic T-lymphocyte effectors and thereby demonstrating preservation of cellular reactivity. A subpopulation of exhausted/senescent CD8(+) T cells, however, downregulates CD28 and upregulates CD57 and PD-1, characterizing immune impairment and relapse after ASCT. Relapsing patients have higher numbers of these cells at +3 months after transplant, but before detection of clinical disease, indicating their applicability in identifying patients at higher risk of relapse. Thus, PD-1 blockade also revives the proliferation and cytokine secretion of the hyporesponsive, exhausted/senescent CD8(+) T cells in vitro. Collectively, these results identify T-cell exhaustion/senescence as a distinguishing feature of relapse and support early introduction of immunotherapy to stimulate antitumor immunity after ASCT (Table S4).

The pembro-rd combination may have an important role as consolidation therapy after ASCT, aiming to achieve a change from an “anergic” to a more “active” immune milieu. This trial provides long-term follow-up of patients with hrMM who have received the combination of checkpoint inhibitors and IMiDS regarding safety and efficacy. While enrollment was limited after the FDA hold, the outcomes were significantly better than historical controls for unmaintained patients would have suggested. The combination of PD1 blockade and IMiDs as a short consolidative regimen offers the possibility of long-term disease control in a population of patients that do not typically achieve this, even with standard maintenance therapies.

多发性骨髓瘤 (MM) 是一种异质性疾病，据美国癌症协会估计，2021 年美国有 34 920 例新诊断病例。随着蛋白酶体抑制剂 (PI)、免疫调节药物 (IMiDs) 和单克隆抗体的发展，标准风险 MM 患者的总生存期 (OS) 超过 10 年，长期生存率得到改善。然而，25%-45% 的高危患者的预后较差，中位 OS 为 18-24 个月。^{1, 2}自体干细胞移植 (ASCT) 后复发是不可避免的，无论是否有来那度胺 (len) 维持。² MM 的免疫功能障碍部分是由于功能缺陷的树突细胞，T _H增加17 细胞、骨髓缺乏抑制细胞 (MDSC) 和 BM 和外周血中的肿瘤相关巨噬细胞 (TAM)。由于 ASCT 后 T 细胞克隆扩增与改善临床结果相关^3，我们假设通过在 ASCT 后巩固中加入检查点抑制剂，它将重振 T 细胞炎症表型以靶向残留 MM。我们使用的 len 已显示可增强 T 细胞和 NK 细胞效应子抗 MM 反应，降低 MM 细胞上的 PD-L1 表达⁴并将其与 pembrolizumab（一种 PD-1 抑制剂）结合使用。

我们使用了 pembrolizumab (pembro) 加 len 和地塞米松 (dex) 的组合，该组合之前在之前的研究中显示出可接受的安全性和更高的疗效⁵并打算招募 58 名患者参加我们的开放标签 II 期单臂试验(NCT02906332)。然而，由于 pembro 组免疫相关毒性和死亡的发生率较高，所有评估 PD-1 或​​ PDL-1 抑制剂与 IMiD 组合的研究于 2017 年 7 月 3 日被 FDA 停止。在 FDA 持有时，招募了 12 名受试者。允许随访。

研究程序如下：患者在 ASCT 后的 +60 至 +180 天之间入组，并接受 pembro-len-dex (pembro-rd) 治疗，共四个周期。该组合包括第 1 天的 pembro 200 mg IV；在第 1-14 天每天口服 25 毫克；在 21 天周期的第 1、8、15 天每天 dex 40 mg，总共两个周期，然后以相同的剂量和频率再进行两个周期的 pembro-rd。在第四周期完成后，每个受试者都被跟踪了 30 天。在治疗结束后 90 天内收集严重不良事件。因疾病进展以外的原因停止治疗的受试者继续治疗后随访。参与者包括患有可测量疾病和由以下任何一项定义的 hrMM 的成年患者：ISS 阶段 3；del 13q 通过细胞遗传学；具有 1q 扩增、1p 缺失 (del)、p53 del、t (4;14)、t (14;16)、t (14;20)、亚二倍体的 FISH；或高风险基因表达谱评分。如果患者在筛选时出现疾病进展或有器官功能障碍的证据，则被排除在外。对于达到 VGPR 或更好的患者，在第四个周期后 30 天进行最小残留病灶 (MRD) 评估。

在筛选、第二周期、治疗第 1 天和长期随访时分析外周血样本。通过 Ficoll-Paque Plus (Fisher Scientific) 分离处理细胞。通过荧光标记的单克隆抗体 (mAb) 染色 T 细胞亚群，对部分细胞进行免疫表型分析。这包括从 BD Biosciences 购买的 mAb：CD3-Alexa Fluor 647 (#557706)、CD4-APC-Cy7 (#557871)、CD45RO-FITC (#555492)、CD25-PerCP-Cy5.5 (#560503)、CD127 -FITC (#560549)、CD8-PE (#555635)、CCR7-PC7 (#557648) 和匹配的荧光标记同种型对照 mAb。

意向治疗人群由所有接受至少一剂研究药物的入选患者组成，是疗效终点分析的基础。使用连续变量的中位数 (IQR) 和分类变量的计数（百分比）来总结人口统计学和临床​​变量。Kaplan-Meier 方法分别用于确定中位 PFS 和 OS。描述性地总结了每个周期对治疗的反应。AE 和 SAE 均以描述方式制成表格。对于 T 细胞亚群表型分析，使用 Mood 中位数检验比较成对的连续数据。两侧p  < 0.05 被认为具有统计学意义。使用 R (ver. 4.0.2) 进行数据分析。

结果如下：在 12 名患者中，5 名 (41.7%) 是 ISS 3，6 名 (50%) 有 FISH 的 p53 缺失，3 名 (25%) 有 1q21 增益，2 名 (17%) 被认为是超FISH 结合 1q21 增益和 13q 缺失的高风险（表 S1）。所有患者均接受三联或四联诱导。5 名患者 (42%) 接受了诱导硼替佐米-来那度胺-dex；四种（33%）卡非佐米-来那度胺-dex、两种（24.9%）硼替佐米-环磷酰胺-dex和一种硼替佐米-环磷酰胺-len-dex。

中位随访时间为 50 个月。中位随访时间为 50.7 个月（范围：6-55.5 个月）的所有患者自 ASCT 之日起的中位 PFS 为 27.7 个月（95% CI：22.6 个月 - 未达到，图 S1）。1 年和 2 年随访的 PFS 率分别为 90.9% 和 63.6%。最佳反应时的总体反应率为 100%（图 S2）。严格完全缓解率在诱导结束时为 33%，在巩固结束时为 83.3%，在预先指定的第 7 次随访时为 83.3%（治疗完成后中位时间为 19 个月）（图 1） . 对于所有患者，在 2 年随访期间，11/12 (91.7%) 达到完全缓解或更好，10/12 (83.3%) 达到严格完全缓解（表 S2）。在完成治疗的 11 名患者中，8 名进行了 MRD 状态评估，其中 7 名 (87.^－5．

在每次访问时监测不良事件 (AE) 并根据 NCI 不良事件通用术语标准 (CTCAE) 4.03 版中概述的指南进行分级。所有患者都有一个或多个归因于 pem、len 或 dex 的 AE。在报告的 97 起 AE 中，6.2% 为 3 级，93.8% 为 1 或 2 级。最常见的非血液学 AE 为呼吸道（咳嗽）和胃肠道（腹泻和便秘），均为 1 或 2 级，而最常见的是血液学 AE 是中性粒细胞减少症，在两名患者中观察到 6 次，1-3 级。免疫介导的 AE 包括 4 名患者的腹泻、1 名患者的发热、1 名患者的斑丘疹和 2 名患者的呼吸困难。均为 1 级或 2 级。有一个严重的 AE，H.流感肺炎需要住院，这与派姆单抗无关（表 S3）。

与第一个治疗周期的 C1D1 开始相比，CD3 隔室内的效应记忆 CD4+ T 细胞百分比在 C4D1 期间稳步下降（p  = 0.037）。CD3 群体中的效应记忆 CD8 区室在第二个周期 (C2D1) 时表现出小幅下降 ( p  = 0.032)，并在第三个周期恢复到微不足道的增加水平 ( p  = 0.199)。在第二个周期 (C2D1) 和第三个周期 (C3D1) 中，记忆调节器 T 细胞的中位数百分比没有显着下降（分别为p  = 1.0，p  = 0.199）（表 S4）。

我们的结果表明，PD-1 阻断联合 len 作为 ASCT 后巩固治疗，在 12 名 hrMM 患者中具有良好的耐受性和有效性，中位 PFS 为 2.3 年，超过历史对照中没有 len 维持的 hrMM 患者的 PFS . ⁶免疫介导的 AE 仅为 1 或 2 级，而观察到的一个 SAE 与 pembro 无关。

ASCT 后淋巴细胞的组成和功能指导了免疫治疗的最佳时机，并有助于识别潜在的复发标志物。随着 CD8(+) T 细胞在 ASCT 后早期淋巴细胞恢复期间扩增，调节性 T 细胞 (Treg) 下降，显着降低 Treg:CD8(+) 效应 T 细胞比率。这些 CD8(+) T 细胞早在移植后第 +12 天就可以对体外呈递肿瘤抗原的自体树突细胞作出反应，成为抗原特异性溶细胞 T 淋巴细胞效应物，从而证明细胞反应性的保留。然而，一个耗竭/衰老的 CD8(+) T 细胞亚群下调 CD28 并上调 CD57 和 PD-1，这是 ASCT 后免疫损伤和复发的特征。复发患者在移植后 +3 个月，但在检测到临床疾病之前，这些细胞的数量更多，表明它们可用于识别具有较高复发风险的患者。因此，PD-1 阻断还可以在体外恢复低反应性、耗竭/衰老的 CD8(+) T 细胞的增殖和细胞因子分泌。总的来说，这些结果将 T 细胞耗竭/衰老确定为复发的一个显着特征，并支持在 ASCT 后尽早引入免疫疗法以刺激抗肿瘤免疫（表 S4）。

pembro-rd 组合可能在 ASCT 后作为巩固治疗发挥重要作用，旨在实现从“无反应性”到更“活跃”的免疫环境的转变。该试验为接受检查点抑制剂和 IMiDS 联合治疗的 hrMM 患者提供关于安全性和有效性的长期随访。虽然在 FDA 暂停后注册人数有限，但结果明显好于未维持患者的历史对照所表明的。PD1 阻断剂和 IMiD 的组合作为一种短期巩固方案，为通常无法实现这一目标的患者群体提供了长期疾病控制的可能性，即使采用标准维持治疗也是如此。