Despite significant advances in the treatment of multiple myeloma (MM), most patients inevitably relapse and ultimately succumb to their disease. In particular, the treatment of patients who are triple-class refractory (TCR, i.e., refractory to an immunomodulatory agent, a proteasome inhibitor, and an anti-CD38 antibody) remains a significant challenge with a median overall survival of 9 months and limited therapeutic options available.^{1, 2} In 2021, the US Food and Drug Administration approved Idecabtagene vicleucel, the first chimeric antigen receptor-T-cell (CAR-T) therapy, for patients with relapsed/refractory multiple myeloma (RRMM) who have received at least four prior lines of therapy. Approval was based on the single arm phase II KARMMA study among 128 patients with heavily pretreated MM (84% TCR) demonstrating an overall response rate of 73%, complete response rate of 33%, and a median progression-free survival of 8.8 months; efficacy parameters considered transformative in this setting.³

While the FDA approval of CAR-T therapy among patients with RRMM represents a significant addition to the therapeutic arsenal, its immediate benefit in real-world patients with MM, particularly older adults, is unknown. Published real-world evidence suggests that 32–61% and 14–38% of patients with symptomatic MM receive second- and third-line therapy, respectively.⁴ Furthermore, the very reeligibility criteria used by the phase II KarMMa study may further limit the number of patients who are considered for this therapy.

We used the Flatiron Health Electronic Health Record (EHR)-derived deidentified database, a longitudinal database comprising patient-level structured and unstructured data, curated via technology-enabled abstraction, originating from approximately 280 cancer clinics across the United States (≈800 sites of care). Our primary study cohort included adults ≥18 years diagnosed with MM from January 2011 and December 2016, who received ≥1 line of therapy (LoT) within 90 days of diagnosis. We applied oncologist-defined, rule-based LoT algorithms, anchored to the addition of new anti-myeloma agents and gap periods during which no treatment was received. The primary outcome of interest was eligibility for CAR-T based on the current FDA approval, defined as a) patients who are initiating fifth line of therapy and b) meeting criteria for CAR-T therapy based on the key eligibility criteria used by the KarMMa (ECOG performance status (PS) 0-1, glomerular filtration rate > 45 mL/min, absolute neutrophil count ≥1250/mm³, platelet count ≥ 75 k/mm³, alanine aminotransferase/aspartate aminotransferase ≤2.5× upper limit, direct bilirubin ≤2× upper limit, and no active treatment for secondary cancer).³ The latter was defined based on documented ECOG PS or laboratory values within 90 days prior to initiation of the 5th LoT, whereas receipt of an anticancer agent not approved for MM during the fifth line was considered as evidence of active treatment for secondary cancer. All patients were followed up to the initiation of fifth-line therapy, death, or their last structured visit, whichever came earlier. Death was ascertained using a previously validated composite mortality variable using patient-level structured (EHR data, obituaries, and Social Security Death Index) and unstructured EHR data.⁵ Real world overall survival (rwOS) was defined as time from the date of diagnosis (index date) to death or last structured activity in the electronic health record. We computed the cumulative incidence of initiating fifth-line therapy with death prior to fifth-line as the competing risk, remaining patients were censored at their last structured activity in the EHR. We computed differences in cumulative incidence rates using univariate Fine-Gray's test. All p-values were two sided, and the level of significance was chosen as 0.05. Institutional Review Board approval of the study protocol was obtained prior to study conduct and included a waiver of informed consent.

We selected 4522 eligible patients during the study period. The overall cohort had a median age at diagnosis of 69 y (IQR 61-77) and 49% with age 70y or older, 54% males, and 63% non-Hispanic whites. Overall, 20% had International Staging System-stage III disease, with 11% high-risk cytogenetics, 31% receiving initial therapy with proteasome inhibitor and immunomodulatory-based triplet therapy, and 28% receiving stem cell transplant. Over a median follow-up of 4.6 years, 2117 (46.8%) patients died, with a median rwOS of the entire cohort of 4.9 (95% CI 4.6–5.2) years. During this time period, a total of 656 (14.5%) of the initial cohort patients started fifth-line therapy, whereas 1793 (39.6%) died prior to starting fifth-line therapy. When accounting for death as a competing risk, the cumulative incidence of starting fifth-line therapy 8 years from diagnosis was 28% in the overall cohort (Figure 1A). However, achievement of 5th-line therapy at 8 years was much lower for older adults (35% for age < 70 years vs 20% for age ≥ 70 years; p value < .01, Figure 1B), while the risk of death before achieving 5th line-therapy was higher among older patients (37% for age < 70 years vs 63.2% for age ≥ 70 years; p value < .01). Among the patients initiating 5th-line therapy, only 44% would be eligible for CAR-T therapy based on the aforementioned eligibility criteria, which would translate to about 15.4% and 8.8% of newly diagnosed patients with MM <70 and ≥70 years, respectively.

Our study highlights that only a small proportion of patients with RRMM, particularly those ≥70 years at the time of diagnosis, are actually likely to benefit from the current CAR-T FDA approval. Ongoing studies are currently looking at these therapies in earlier lines of therapy so that more patients can potentially be benefited. Our findings are overall consistent with a recent study by Fonseca et al showing high attrition rates across lines of therapy among patients with newly diagnosed MM, with only 8% of nontransplant patients and 22% of total transplant patients receiving a fifth-line therapy, although we report slightly higher proportions.⁴ This difference is likely explained by the use of competing risk framework in our analysis, which takes into account the high number of deaths occurring prior to the initiation of fifth-line therapy. With the incorporation of CD38 monoclonal antibodies into earlier regimens, it is quite likely that most patients will have TCR myeloma by second or third line of therapy, and there is a large unmet need for novel therapies in this population. Therefore, approvals should be defined based on development of TCR rather than based on a set number of lines of therapy. Furthermore, our findings that majority of patients reaching 5th LoT would not meet eligibility criteria for CAR-T therapy calls upon studying the efficacy and safety of these therapies among patients with broader set of eligibility criteria than used by KarMMa study. The real-world data on the outcomes of patients who actually obtained commercial CAR-T will further clarify the eligibility criteria and outcomes of these heavily pretreatment patients. Lastly, clinical trials should continue to test CAR-T therapies in earlier lines of therapy, and results from several of these trials (NCT04855136, NCT03651128, and NCT04181827) are eagerly awaited.

Our study has several limitations. While not nationally representative, our data originated from 280 cancer clinics (≈800 practice sites), predominantly community cancer centers across 44 states in the United States. Our study cohort was derived from patients diagnosed in the pre-CART era, and it is possible that a greater proportion of patients may receive a 5th LoT than those reported in our study. Despite having more recent data (2016–2021), we were forced to select an earlier time-period to allow sufficient follow-up and more accurately ascertain the proportion of patients receiving 5th LoT. Since, our results were heavily influenced by high attrition rates; these findings may not only be unique to CART but for any novel therapies for MM approved in the later lines of therapy. Patients who transfer their care to a different practice site may have lost to follow-up, leading to an underestimation of the proportion of patients starting a LoT. Data on comorbidity and organ function (other than renal/hepatic) was not available; therefore, we were not able to adjudicate treatment ineligibility based on cardiac or pulmonary status as originally defined in KARMMA study.⁶ Lastly, eligibility status on some patients (≈20%) starting fifth-line therapy could not be determined due to missing data on performance status or laboratory data, and we restricted our analysis to complete cases only.

To summarize, less than 10% of older adults with newly diagnosed MM patients are expected to be eligible for CAR-T therapy based on the current FDA approval and eligibility criteria. Meanwhile, a much higher proportion of patients die before reaching CAR-T eligibility. These findings highlight the need to explore CAR-T cell therapy in earlier lines of disease and in a population that better represents real-world patients to expand the applicability of this novel treatment.

尽管多发性骨髓瘤 (MM) 的治疗取得了重大进展，但大多数患者不可避免地会复发并最终死于疾病。特别是对三类难治性（TCR，即免疫调节剂、蛋白酶体抑制剂和抗 CD38 抗体难治性）患者的治疗仍然是一个重大挑战，中位总生存期为 9 个月，治疗有限可用的选项。^{1, 2}2021 年，美国食品和药物管理局批准了首个嵌合抗原受体 T 细胞 (CAR-T) 疗法 Idecabtagene vicleucel，用于治疗至少接受过四线治疗的复发/难治性多发性骨髓瘤 (RRMM) 患者. 批准基于单臂 II 期 KARMMA 研究，该研究在 128 名接受过重度治疗的 MM（84% TCR）患者中进行，总体缓解率为 73%，完全缓解率为 33%，中位无进展生存期为 8.8 个月；在这种情况下被认为具有变革性的功效参数。³

虽然 FDA 对 RRMM 患者的 CAR-T 疗法的批准是对治疗武器库的重要补充，但它对现实世界的 MM 患者，特别是老年人的直接益处尚不清楚。已发表的真实世界证据表明，32-61% 和 14-38% 的有症状 MM 患者分别接受二线和三线治疗。⁴此外，II 期 KarMMa 研究使用的重新合格标准可能会进一步限制考虑接受这种治疗的患者数量。

我们使用了 Flatiron Health 电子健康记录 (EHR) 衍生的去识别数据库，这是一个纵向数据库，包括患者级别的结构化和非结构化数据，通过技术支持的抽象进行管理，来自美国大约 280 个癌症诊所（约 800 个关心）。我们的主要研究队列包括从 2011 年 1 月到 2016 年 12 月诊断为 MM 的≥18 岁的成年人，他们在诊断后 90 天内接受了≥1 线治疗 (LoT)。我们应用了肿瘤学家定义的、基于规则的 LoT 算法，以添加新的抗骨髓瘤药物和未接受治疗的间隔期为基础。感兴趣的主要结果是基于当前 FDA 批准的 CAR-T 资格，³、血小板计数≥75 k/mm ³、丙氨酸氨基转移酶/天冬氨酸氨基转移酶≤2.5×上限，直接胆红素≤2×上限，继发癌无积极治疗）。³后者是根据第 5 次 LoT 开始前 90 天内记录的 ECOG PS 或实验室值定义的，而在第 5 行期间收到未批准用于 MM 的抗癌剂被认为是继发性癌症积极治疗的证据。对所有患者进行随访，直至开始五线治疗、死亡或最后一次结构化就诊，以较早者为准。使用先前验证的复合死亡率变量确定死亡，该变量使用患者级别的结构化（EHR 数据、讣告和社会保障死亡指数）和非结构化 EHR 数据。⁵真实世界总生存期 (rwOS) 定义为从诊断日期（索引日期）到死亡或电子健康记录中最后一次结构化活动的时间。我们计算了开始五线治疗的累积发生率，在五线之前死亡作为竞争风险，其余患者在 EHR 中的最后一次结构化活动中被审查。我们使用单变量 Fine-Gray 检验计算了累积发病率的差异。所有p值都是两侧的，显着性水平选择为 0.05。在研究进行之前获得机构审查委员会对研究方案的批准，并包括放弃知情同意。

我们在研究期间选择了 4522 名符合条件的患者。整个队列诊断时的中位年龄为 69 岁（IQR 61-77），49% 的患者年龄在 70 岁或以上，54% 为男性，63% 为非西班牙裔白人。总体而言，20% 患有国际分期系统 III 期疾病，11% 为高危细胞遗传学，31% 接受蛋白酶体抑制剂和基于免疫调节的三联疗法的初始治疗，28% 接受干细胞移植。在 4.6 年的中位随访期间，2117 名 (46.8%) 患者死亡，整个队列的中位 rwOS 为 4.9 (95% CI 4.6-5.2) 年。在此期间，共有 656 名（14.5%）初始队列患者开始了五线治疗，而 1793 名（39.6%）在开始五线治疗之前死亡。在将死亡视为竞争风险时，在整个队列中，诊断后 8 年开始五线治疗的累积发生率为 28%（图 1A）。然而，老年人在 8 岁时达到 5 线治疗的比例要低得多（年龄 < 70 岁为 35%，年龄 ≥ 70 岁为 20%；p值 < .01，图 1B），而老年患者在达到 5 线治疗前的死亡风险较高（年龄 < 70 岁为 37%，年龄 ≥ 70 岁为 63.2%；p值 < .01）。在开始 5 线治疗的患者中，根据上述资格标准，只有 44% 的患者有资格接受 CAR-T 治疗，这意味着新诊断的 MM <70 岁和≥70 岁的患者中分别有 15.4% 和 8.8%，分别。

我们的研究强调，只有一小部分 RRMM 患者，尤其是诊断时 ≥ 70 岁的患者，实际上可能受益于当前的 CAR-T FDA 批准。目前正在进行的研究正在研究早期治疗中的这些疗法，以便更多的患者可能受益。我们的研究结果总体上与 Fonseca 等人最近的一项研究一致，该研究显示新诊断的 MM 患者的跨线治疗流失率很高，只有 8% 的非移植患者和 22% 的移植患者接受了五线治疗，尽管我们报告的比例略高。⁴这种差异很可能通过我们分析中使用竞争风险框架来解释，该框架考虑到在开始五线治疗之前发生的大量死亡。随着 CD38 单克隆抗体纳入早期治疗方案，大多数患者很可能会通过二线或三线治疗患有 TCR 骨髓瘤，并且该人群对新疗法的大量需求未得到满足。因此，批准应该基于 TCR 的发展而不是基于固定数量的治疗线来定义。此外，我们发现大多数达到第 5 次 LoT 的患者将不符合 CAR-T 治疗的资格标准，这要求在具有比 KarMMa 研究使用的更广泛的资格标准的患者中研究这些疗法的有效性和安全性。实际获得商业 CAR-T 的患者结果的真实数据将进一步阐明这些重度预处理患者的资格标准和结果。最后，临床试验应继续在早期治疗线中测试 CAR-T 疗法，并且热切期待其中几个试验（NCT04855136、NCT03651128 和 NCT04181827）的结果。

我们的研究有一些局限性。虽然没有全国代表性，但我们的数据来自 280 家癌症诊所（约 800 个诊所），主要是美国 44 个州的社区癌症中心。我们的研究队列来自在 CART 时代之前诊断出的患者，与我们研究中报告的患者相比，可能有更大比例的患者接受第 5 次 LoT。尽管有更近期的数据（2016-2021 年），但我们被迫选择更早的时间段，以便进行充分的随访并更准确地确定接受第 5 次 LoT 的患者比例。因为，我们的结果受到高流失率的严重影响；这些发现可能不仅是 CART 独有的，而且对于后期治疗中批准的任何新的 MM 疗法也是如此。将护理转移到不同诊所的患者可能会失去随访，导致低估开始 LoT 的患者比例。没有关于合并症和器官功能（肾/肝除外）的数据；因此，我们无法根据最初在 KARMMA 研究中定义的心脏或肺部状态来判定治疗不合格。⁶最后，由于体能状态或实验室数据缺失，一些患者（≈20%）开始五线治疗的资格状态无法确定，我们将分析限制为仅完成病例。

总而言之，根据目前 FDA 的批准和资格标准，预计不到 10% 的新诊断为 MM 患者的老年人有资格接受 CAR-T 治疗。同时，更高比例的患者在达到 CAR-T 资格之前死亡。这些发现强调了在早期疾病和更能代表现实世界患者的人群中探索 CAR-T 细胞疗法的必要性，以扩大这种新疗法的适用性。