Frontline therapy of Philadelphia chromosome-positive acute lymphoblastic leukemia (Ph + ALL) consists of chemotherapy or blinatumomab in combination with a BCR::ABL1 tyrosine kinase inhibitor (TKI). Ponatinib is a potent BCR::ABL1 TKI that is active against T315I ABL1 mutations, which are the dominant mechanism of relapse with earlier-generation TKIs. Promising outcomes have been achieved with frontline ponatinib-based regimens in Ph + ALL. With hyper-CVAD plus ponatinib, a 6-year overall survival (OS) of 75% has been reported, which compares favorably to historical expectations with hyper-CVAD in combination with a first- or second-generation TKI.¹ In an updated analysis from an ongoing study of the chemotherapy-free combination of ponatinib and blinatumomab, the estimated 2-year OS rate is 89%, without the need for allogeneic hematopoietic stem cell transplantation (HSCT) in first remission in the vast majority of patients (<5% transplant rate for the study population).² Ponatinib was also recently shown to be superior to imatinib in a randomized phase III clinical trial in adults with newly diagnosed Ph + ALL.³ However, despite encouraging data reported with ponatinib-containing frontline regimens, some patients still relapse. The characteristics of patients with Ph + ALL who relapse following a frontline ponatinib-based regimen have not been systemically evaluated. In this analysis, we sought to evaluate the patterns of relapse in these patients and to describe their clinical outcomes following salvage therapy.

We retrospectively evaluated the outcomes of patients with Ph + ALL following the failure of a ponatinib-based regimen. To be eligible for this analysis, patients were required to have received ponatinib as frontline therapy and subsequently experienced hematologic/morphological and/or extramedullary relapse. Patients were categorized as either ponatinib-resistant (in cases where relapse occurred while on ponatinib) or ponatinib-intolerant (in cases where relapse occurred after discontinuation of ponatinib due to toxicity). Clinical outcomes were analyzed from the time of the first relapse following the use of a frontline ponatinib-based regimen (defined as the time of “ponatinib failure” for this analysis). Relapse-free survival (RFS) was calculated from the time of second complete remission (CR) or CR with incomplete hematologic recovery (CRi) until relapse or death from any cause, censored if the patient was alive at the last follow-up. OS was calculated from the time of ponatinib failure until death from any cause, censored if the patient was alive at the last follow-up.

Between March 2012 and September 2023, we identified 22 patients with Ph + ALL who experienced treatment failure after a frontline ponatinib-containing regimen (out of 201 patients with newly diagnosed Ph + ALL treated with ponatinib). The characteristics of the study population are summarized in Table S1. The median age of patients at the time of ponatinib failure was 49 years (range, 27–76 years). Sixteen patients (73%) received chemotherapy as part of frontline therapy and 6 (27%) received a frontline chemotherapy-free regimen. Four patients (18%) had received inotuzumab ozogamicin prior to relapse and 11 patients (50%) had received blinatumomab; all four patients received inotuzumab ozogamicin due to persistent or recurrent measurable residual disease (MRD). No patient had undergone prior allogeneic HSCT. Eighteen patients (82%) had achieved complete molecular response (CMR; defined as no detectable BCR::ABL1 transcripts with a sensitivity of at least 10⁻⁴) within 3 months of the start of therapy. Sixteen patients (73%) were ponatinib-resistant and 6 (27%) were ponatinib-intolerant. Among the ponatinib-resistant patients, 11 of 16 (69%) were receiving ponatinib at a dose of 15 mg daily at the time of relapse. Among the six ponatinib-intolerant patients, four were on dasatinib at the time of relapse, and one each was on imatinib and nilotinib. Reasons for ponatinib intolerance were: cardiovascular toxicity in four patients, venous thromboembolism in one patient, and liver function test abnormalities in one patient.

Among the ponatinib-resistant patients, the median time from diagnosis to relapse was 24.1 months (range, 8.8–79.0 months). Among the ponatinib-intolerant patients, the median time from diagnosis to relapse was 31.4 months (range, 11.4–80.7 months) and the median time from ponatinib intolerance to relapse was 22.4 months (range, 6.1–69.1 months). Among 17 patients with MRD testing (PCR for BCR::ABL1) within 4 months prior to relapse, MRD recurrence or an increase in MRD levels was observed in eight patients (47%). The types of relapses observed in the cohort were: bone marrow only (n = 12, 54%), central nervous system (CNS)-only (n = 6, 27%), bone marrow with CNS, and/or non-CNS extramedullary involvement (n = 3, 14%), and non-CNS extramedullary-only (n = 1, 5%). All non-CNS extramedullary relapses occurred in the lymph nodes and were biopsy-proven. There was no difference in the number of doses of intrathecal chemotherapy for patients with CNS-only versus other types of relapses (median of 12 doses in both groups). Extramedullary-only relapses were more common in patients who received a frontline chemotherapy-free regimen than in those who received a chemotherapy-based regimen (four out of six [67%] v.s three out of 16 [19%], respectively; p = .05). ABL1 kinase domain testing was performed on 13 patients with bone marrow relapse as previously described⁴: five patients (38%) had no detectable ABL1 mutation, four patients (31%) had a T315I mutation, two patients (15%) had an E255K mutation, and one patient each (8%) had a E255V or F359K mutation. Among the four patients who developed a T315I mutation, three were ponatinib-intolerant and were receiving dasatinib at the time of relapse.

Overall, 20 patients (91%) received at least one subsequent therapy after ponatinib failure. Among these patients, the median number of subsequent therapies was 1 (range, 1–5); 7 of the 20 patients (35%) received two or more subsequent therapies after ponatinib failure. A variety of salvage regimens were used. All patients received a TKI with first salvage. The TKIs used were: ponatinib in eight patients (40%), dasatinib in seven patients (35%), bosutinib in four patients (20%), and olverembatinib in one patient (5%). One patient who was considered ponatinib-intolerant was rechallenged with ponatinib at the time of relapse due to T315I mutation; all other patients who received ponatinib in first salvage were classified as ponatinib-resistant. No patient received TKI monotherapy as first salvage. Thirteen patients (65%) received chemotherapy as part of first salvage therapy, six (30%) received blinatumomab, four (20%) received inotuzumab ozogamicin, and seven (35%) received CD19 chimeric antigen receptor (CAR) T-cells. Overall, 16 patients (80%) received blinatumomab, inotuzumab ozogamicin or CAR T-cells with first salvage. All seven patients who relapsed in the CNS received cranial and/or spinal irradiation (six with their first salvage regimen; and one as part of HSCT conditioning).

Overall, 18 patients (90%) achieved second remission following the first salvage therapy, including CR in 15 patients and CRi in three patients. All patients who received blinatumomab, inotuzumab ozogamicin, and/or CD19 CAR T-cells as part of their first salvage therapy achieved CR/CRi. Among 12 responding patients with bone marrow relapse, 10 (83%) achieved a CMR. Two patients did not respond to first salvage therapy, both of whom had bone marrow-only relapses. One patient received venetoclax plus ponatinib and the other received MOpAD (methotrexate, vincristine, pegylated-asparaginase, and dexamethasone), ofatumumab, and dasatinib as first salvage. Both patients received subsequent salvage regimens but did not respond to any of them.

The patient characteristics, responses to salvage therapy, and disposition for the study cohort are shown in Figure 1A and Table S2. The median duration of follow-up of the entire study population was 18.6 months (range, 1.1–112.9 months) from the time of ponatinib failure. Among the 18 responders to first salvage therapy, six patients (33%) relapsed in the absence of HSCT, one (6%) died in remission, five (28%) proceeded to HSCT in second remission, and six (33%) are still alive in continuous ongoing remission without HSCT. The recommendation for HSCT in second remission was individualized and no formal criteria were used. The median duration of remission of the six patients who remain in remission without HSCT is 5.3 months (range, 1.0–104.8 months). Among the five patients transplanted in second remission, one died from HSCT-related complications, one died from unknown causes, and three are still alive in continuous remission. Among the nine patients who are alive and in ongoing second remission, the median duration of remission was 14.2 months (range, 1.0–104.8 months).

For the entire cohort, the median RFS was 14.7 months (1-year and 2-year RFS: 55% and 37%, respectively), and the median OS was 22.6 months (1-year and 2-year OS: 68% and 41%, respectively) (Figure 1B,C). Among patients who relapsed in the bone marrow (with or without concomitant extramedullary relapse) the median OS was 15.7 months, and the 2-year OS was 31%; in contrast, for patients with extramedullary-only relapses, the median OS was not reached, and the 2-year OS was 86% (Figure 1D). At last follow-up, four patients were alive and in ongoing second remission for at least 2 years after ponatinib failure. Three of these patients underwent HSCT following first salvage therapy; the other patient received blinatumomab plus dasatinib as first salvage and remains in remission 9 years after relapse without HSCT.

Here, we describe the largest cohort of patients with Ph + ALL relapsing after ponatinib-based frontline therapy. Among the 22 patients in our study, seven (32%) experienced extramedullary-only relapse (six of which were CNS-only), with higher rates of extramedullary-only relapses observed in patients who received a frontline chemotherapy-free regimen. This finding is consistent with reports from prospective studies of blinatumomab plus TKI regimens and highlights the need for measures to prevent CNS relapses in these patients.^{2, 5} In the current blinatumomab plus ponatinib clinical trial, the number of doses of prophylactic intrathecal chemotherapy has been increased from 12 to 15 and serial monitoring of the cerebrospinal fluid during maintenance is now being instituted. The use of systemic chemotherapy in patients at high risk for CNS relapse is also being considered. For patients relapsing in the bone marrow, the mechanism of resistance to ponatinib-based regimens remains poorly understood. Three of the 13 tested patients had E255K or E255V mutation which may confer resistance to ponatinib. However, in most other cases, no causative ABL1 mutation could be identified, suggesting that many relapses after ponatinib therapy may be driven by BCR::ABL1-independent mechanisms.

We observed a relatively high response rate to salvage therapy (90%), and the 2-year OS of 41% observed in our cohort suggests that some patients can have long-term survival even after ponatinib failure. Impressively, these outcomes appear superior to a previous report of patients with relapsed Ph + ALL, where second remission was achieved in 84% of patients and the 2-year OS was 20%.⁶ The better outcomes in our cohort may be due to the frequent utilization of immunotherapy-based salvage regimens at the time of relapse. It is notable that survival in patients who experienced bone marrow relapse appeared inferior to those with extramedullary-only relapse (2-year OS 31% versus 86%, respectively). While CNS relapses remain a particular concern with chemotherapy-free regimens, these findings suggest that durable remissions can be achieved with appropriate CNS-directed therapy, including intrathecal chemotherapy, craniospinal irradiation, CAR T-cells, and/or HSCT.

The appropriate consolidation for most patients with Ph + ALL who achieve second remission following ponatinib failure remains uncertain. In our study, only 28% of patients who achieved second remission underwent allogeneic HSCT. Unfortunately, the number of patients was too small to assess the relative benefit of consolidative HSCT in second remission. Three of four patients with long-term (i.e., >2 years) durable remissions after ponatinib failure underwent subsequent HSCT, which suggests that HSCT may be beneficial for some patients. However, longer-term follow-up of patients receiving novel approaches such as CAR T-cell therapy are needed to better assess the optimal consolidative strategy for patients with Ph + ALL who relapse following a ponatinib-based regimen.

费城染色体阳性急性淋巴细胞白血病 (Ph + ALL) 的一线治疗包括化疗或博纳吐单抗联合 BCR::ABL1 酪氨酸激酶抑制剂 (TKI)。 Ponatinib 是一种有效的 BCR::ABL1 TKI，对 T315I ABL1 突变具有活性，T315I ABL1 突变是早期 TKI 复发的主要机制。基于 ponatinib 的一线治疗方案在 Ph + ALL 中取得了有希望的结果。据报道，hyper-CVAD 加 ponatinib 的 6 年总生存 (OS) 为 75%，这优于 hyper-CVAD 联合第一代或第二代 TKI 的历史预期。 ¹ 在一项正在进行的 ponatinib 和 blinatumomab 免化疗组合研究的最新分析中，预计 2 年 OS 率为 89%，且无需进行同种异体造血干细胞移植 (HSCT)绝大多数患者首次缓解（研究人群的移植率<5%）。 ² 最近，在一项针对新诊断 Ph + ALL 成人的随机 III 期临床试验中，帕纳替尼也被证明优于伊马替尼。 ³ 然而，尽管含有 ponatinib 的一线治疗方案报告了令人鼓舞的数据，但一些患者仍然复发。尚未对接受基于 ponatinib 的一线治疗方案后复发的 Ph + ALL 患者的特征进行系统评估。在这项分析中，我们试图评估这些患者的复发模式，并描述他们在挽救治疗后的临床结果。

我们回顾性评估了 Ph + ALL 患者在基于 ponatinib 的治疗方案失败后的结局。为了符合这项分析的资格，患者需要接受普纳替尼作为一线治疗，并随后经历血液学/形态学和/或髓外复发。患者被分类为帕纳替尼耐药（在使用帕纳替尼期间发生复发的情况）或帕纳替尼不耐受（由于毒性而停用帕纳替尼后发生复发的情况）。从使用基于普纳替尼的一线治疗方案后第一次复发的时间（在本分析中定义为“普纳替尼失败”的时间）开始分析临床结果。无复发生存期 (RFS) 是从第二次完全缓解 (CR) 或 CR 伴不完全血液学恢复 (CRi) 的时间开始计算，直到复发或因任何原因死亡，如果患者在最后一次随访时还活着，则进行审查。 OS 是从 ponatinib 失败时开始计算，直到因任何原因死亡为止，如果患者在最后一次随访时还活着，则进行审查。

2012 年 3 月至 2023 年 9 月期间，我们确定了 22 名 Ph + ALL 患者在接受含 ponatinib 的一线治疗方案后经历治疗失败（201 名新诊断 Ph + ALL 患者接受 ponatinib 治疗）。表 S1 总结了研究人群的特征。 ponatinib 失败时患者的中位年龄为 49 岁（范围：27-76 岁）。 16 名患者 (73%) 接受化疗作为一线治疗的一部分，6 名患者 (27%) 接受一线无化疗方案。 4 名患者 (18%) 在复发前接受了伊珠单抗奥佐米星治疗，11 名患者 (50%) 接受了 blinatumomab 治疗；所有四名患者均因持续或复发性可测量残留病灶 (MRD) 接受了伊珠单抗奥佐米星治疗。没有患者之前接受过同种异体 HSCT。 18 名患者 (82%) 在治疗开始后 3 个月内实现了完全分子缓解（CMR；定义为未检测到 BCR::ABL1 转录物，灵敏度至少为 10 ⁻⁴ ）。 16 名患者 (73%) 对 ponatinib 耐药，6 名患者 (27%) 对 ponatinib 不耐受。在 ponatinib 耐药的患者中，16 名患者中有 11 名 (69%) 在复发时接受了每天 15 mg 剂量的 ponatinib。在六名帕纳替尼不耐受的患者中，四名在复发时服用达沙替尼，各一名服用伊马替尼和尼洛替尼。普纳替尼不耐受的原因包括：4 名患者出现心血管毒性、1 名患者出现静脉血栓栓塞、1 名患者出现肝功能检查异常。

在普纳替尼耐药患者中，从诊断到复发的中位时间为 24.1 个月（范围为 8.8-79.0 个月）。在帕纳替尼不耐受的患者中，从诊断到复发的中位时间为31.4个月（范围为11.4-80.7个月），从帕纳替尼不耐受到复发的中位时间为22.4个月（范围为6.1-69.1个月）。在复发前 4 个月内进行 MRD 检测（针对 BCR::ABL1 的 PCR）的 17 名患者中，有 8 名患者 (47%) 观察到 MRD 复发或 MRD 水平升高。在队列中观察到的复发类型为：仅骨髓（n = 12，54%）、仅中枢神经系统（CNS）（n = 6，27%）、骨髓伴 CNS 和/或非 CNS髓外受累 (n = 3, 14%)，和仅非 CNS 髓外受累 (n = 1, 5%)。所有非中枢神经系统髓外复发均发生在淋巴结，并经过活检证实。仅 CNS 复发患者与其他类型复发患者的鞘内化疗剂量没有差异（两组中位数均为 12 剂）。接受一线无化疗方案的患者比接受化疗方案的患者中仅髓外复发更常见（分别为六分之四 [67%] 和 16 分之三 [19%]；p = .05）。如前所述，对 13 名骨髓复发患者进行了 ABL1 激酶结构域检测 ⁴ ：5 名患者（38%）没有检测到 ABL1 突变，4 名患者（31%）有 T315I 突变，2 名患者（ 15%）有 E255K 突变，各一名患者（8%）有 E255V 或 F359K 突变。在出现 T315I 突变的 4 名患者中，有 3 名患者对 ponatinib 不耐受，并在复发时正在接受达沙替尼治疗。

总体而言，20 名患者 (91%) 在 ponatinib 失败后接受了至少一种后续治疗。在这些患者中，后续治疗的中位数为 1 次（范围为 1-5 次）； 20 名患者中有 7 名 (35%) 在 ponatinib 失败后接受了两种或多种后续治疗。使用了多种抢救方案。所有患者均在首次抢救时接受了 TKI。使用的 TKI 为：8 名患者（40%）使用普纳替尼（ponatinib），7 名患者（35%）使用达沙替尼（dasatinib），4 名患者（20%）使用博舒替尼（bosutinib），以及 1 名患者（5%）使用 olverembatinib。一名被认为对 ponatinib 不耐受的患者因 T315I 突变而在复发时再次接受 ponatinib 治疗；所有其他在第一次挽救中接受普纳替尼治疗的患者均被归类为普纳替尼耐药。没有患者接受 TKI 单药治疗作为首次挽救。 13 名患者 (65%) 接受化疗作为首次挽救治疗的一部分，6 名患者 (30%) 接受博纳吐单抗治疗，4 名患者 (20%) 接受伊珠单抗奥佐米星治疗，7 名患者 (35%) 接受 CD19 嵌合抗原受体 (CAR) T 细胞治疗。总体而言，16 名患者 (80%) 首次挽救时接受了 blinatumomab、inotuzumab ozogamicin 或 CAR T 细胞治疗。所有 7 名中枢神经系统复发患者均接受了颅脑和/或脊柱放疗（其中 6 名患者接受了首次挽救方案；1 名患者作为 HSCT 预处理的一部分）。

总体而言，18 名患者 (90%) 在第一次挽救治疗后实现了第二次缓解，其中 15 名患者获得 CR，3 名患者获得 CRi。所有接受 blinatumomab、inotuzumab ozogamicin 和/或 CD19 CAR T 细胞作为首次挽救治疗一部分的患者均达到 CR/CRi。在 12 名有骨髓复发反应的患者中，10 名 (83%) 达到了 CMR。两名患者对第一次挽救治疗没有反应，两人都出现了仅骨髓复发。一名患者接受维奈托克加普纳替尼治疗，另一名患者接受 MOpAD（甲氨蝶呤、长春新碱、聚乙二醇化天冬酰胺酶和地塞米松）、奥法木单抗和达沙替尼作为首次挽救治疗。两名患者均接受了随后的挽救方案，但对任何方案都没有反应。

研究队列的患者特征、对挽救治疗的反应和处置如图 1A 和表 S2 所示。从 ponatinib 失败之日起，整个研究人群的中位随访时间为 18.6 个月（范围为 1.1-112.9 个月）。在对第一次挽救治疗有反应的 18 名患者中，6 名患者 (33%) 在没有进行 HSCT 的情况下复发，1 名患者 (6%) 在缓解期死亡，5 名患者 (28%) 在第二次缓解时接受 HSCT，还有 6 名患者 (33%) 在缓解期死亡。在不进行 HSCT 的情况下仍处于持续缓解状态。第二次缓解时 HSCT 的建议是个体化的，没有使用正式的标准。 6 名未接受 HSCT 仍保持缓解的患者的中位缓解持续时间为 5.3 个月（范围：1.0-104.8 个月）。第二次缓解期移植的5例患者中，1例死于HSCT相关并发症，1例死因不明，3例持续缓解期仍存活。在 9 名存活且正在进行第二次缓解的患者中，中位缓解持续时间为 14.2 个月（范围：1.0-104.8 个月）。

对于整个队列，中位 RFS 为 14.7 个月（1 年和 2 年 RFS：分别为 55% 和 37%），中位 OS 为 22.6 个月（1 年和 2 年 OS：68% 和分别为 41%）（图 1B、C）。在骨髓复发（伴或不伴髓外复发）的患者中，中位 OS 为 15.7 个月，2 年 OS 为 31%；相反，对于仅髓外复发的患者，中位 OS 未达到，2 年 OS 为 86%（图 1D）。在最后一次随访中，四名患者仍然活着，并且在 ponatinib 失败后至少 2 年持续处于第二次缓解状态。其中三名患者在第一次挽救治疗后接受了 HSCT；另一名患者接受了 blinatumomab 加达沙替尼作为首次挽救，在复发后 9 年仍保持缓解，未进行 HSCT。

在这里，我们描述了接受基于 ponatinib 的一线治疗后 Ph + ALL 复发的最大患者群体。在我们研究的 22 名患者中，7 名 (32%) 经历了单纯髓外复发（其中 6 例仅发生中枢神经系统复发），在接受一线无化疗方案的患者中观察到单纯髓外复发率更高。这一发现与 blinatumomab 加 TKI 方案的前瞻性研究报告一致，并强调需要采取措施预防这些患者的 CNS 复发。 ^{2, 5} 在目前的blinatumomab加ponatinib临床试验中，预防性鞘内化疗的剂量已从12剂增加到15剂，并且正在开始维持期间对脑脊液的连续监测。还正在考虑对中枢神经系统复发高风险患者使用全身化疗。对于骨髓复发的患者，对基于普纳替尼的治疗方案的耐药机制仍然知之甚少。 13 名接受测试的患者中有 3 名患有 E255K 或 E255V 突变，可能会对普纳替尼产生耐药性。然而，在大多数其他情况下，无法识别出致病性 ABL1 突变，这表明 ponatinib 治疗后的许多复发可能是由 BCR::ABL1 独立机制驱动的。

我们观察到对挽救治疗的反应率相对较高（90%），并且在我们的队列中观察到的 2 年 OS 为 41%，这表明一些患者即使在 ponatinib 失败后也可以长期生存。令人印象深刻的是，这些结果似乎优于之前复发性 Ph + ALL 患者的报告，其中 84% 的患者实现了第二次缓解，2 年 OS 为 20%。 ⁶ 我们的队列中更好的结果可能是由于在复发时频繁使用基于免疫治疗的挽救方案。值得注意的是，经历过骨髓复发的患者的生存率似乎低于仅发生髓外复发的患者（2 年 OS 分别为 31% 和 86%）。虽然中枢神经系统复发仍然是无化疗方案的一个特别令人担忧的问题，但这些研究结果表明，通过适当的中枢神经系统定向治疗，包括鞘内化疗、颅脊髓照射、CAR T 细胞和/或 HSCT，可以实现持久缓解。

对于大多数在普纳替尼失败后获得第二次缓解的 Ph + ALL 患者，适当的巩固治疗仍不确定。在我们的研究中，只有 28% 获得第二次缓解的患者接受了同种异体 HSCT。不幸的是，患者数量太少，无法评估巩固 HSCT 在第二次缓解中的相对益处。 ponatinib 失败后获得长期（即 >2 年）持久缓解的四分之三的患者接受了随后的 HSCT，这表明 HSCT 可能对某些患者有益。然而，需要对接受 CAR T 细胞治疗等新方法的患者进行长期随访，以更好地评估接受基于 ponatinib 的方案后复发的 Ph++ALL 患者的最佳巩固策略。