Both SARS-CoV-2 infection and vaccination have raised concern in immune mediated diseases, including autoimmune cytopenias (AIC, ie, autoimmune hemolytic anemia, AIHA; immune thrombocytopenia, ITP; autoimmune neutropenia, AIN; aplastic anemia, AA; and their combination, termed Evans syndrome, ES). Autoimmune cytopenias are highly heterogeneous conditions with variable severity and a clinical course. They are marked by several relapses often triggered by immune-activating events (infections, traumas, surgery) including vaccines.^{1, 2} Some reports of ITP and AIHA exacerbations after SARS-CoV-2 vaccines (mRNA vaccines Pfizer-Biontech and Moderna, and Adenovirus based vaccine AstraZeneca) have been described,^3-16 but evidence-based indications for their management are lacking.

Here we prospectively studied a large series of 108 patients with AIC (56 AIHA, 41 warm type, wAIHA and 15 cold, cAIHA, 38 ITP, 7 AIN, and 7 AA) prospectively followed at a reference hematologic center in Milan, Italy, who underwent SARS-CoV-2 vaccination from March, 24 until the end of June 2021. The study was conducted in accordance with Helsinki Declaration and each participant had given a written informed consent for data collection. Patients (median age 62 years, range 25–89 years, female/male ratio 1.7) were monitored with whole blood counts (and LDH levels in AIHA) the week before and the week after each vaccination dose. Importantly, ongoing AIC therapy (38% of cases, including steroids, cyclosporine, eltrombopag, and complement inhibitor sutimlimab) were kept stable within the 2 weeks before the first dose.

Table 1 summarizes hematologic trends and side effects observed after each dose in patients with ITP, AIHA, AIN, and AA. Seven patients had ES, of whom four ITP plus AIHA and one ITP plus neutropenia. Patients mainly received Pfizer-BioNtech vaccine (N = 90), followed by Moderna (N = 16), and Astra-Zeneca (N = 2). Hematological parameters showed a wide distribution both at baseline and after vaccines. To better investigate intra-patient variation we calculated the delta% change of Hb and LDH for AIHA and of PLT in ITP patients, after the first and the second vaccine dose (supporting information Figure S1).

Regarding AIHA, four elderly patients experienced a clinically significant Hb reduction requiring treatment adjustment. In detail, patient number one was a 79-year-old female with warm type AIHA (wAIHA) who experienced an Hb decrease from 10.4 to 9.1 g/dL (LDH 1.2 to 1.7 × ULN) after the first dose of Pfizer vaccine; she required a slight increase of steroid dose (to 5 mg day prednisone) and remained stable after the second dose. Patient number two was a 73-year-old male with wAIHA, off-therapy; he had a frank relapse 1 week after the first dose of Moderna vaccine (Hb 13.9 to 9.1 g/dL, LDH 1.1 to 1.6 × ULN), requiring prednisone 0.5 mg/kg day with prompt response. The patient refused to receive the second dose. Patient number three was a 77-year-old male suffering from cold type AIHA (cAIHA), who had an Hb drop from 9.3 to 7.2 g/dL 1 week after after the first dose of Moderna vaccine, requiring steroids, rituximab, and recombinant erythropoietin. The second dose was administered without adverse events. Patient number four was a 73-year-old man suffering from multi-relapsing wAIHA on low-dose steroids, who experienced a severe relapse (Hb 14 to 7.4 g/dL, LDH 1.1 to 2.3 × ULN) 5 days after the second dose of Pfizer vaccine. He required high dose intravenous steroids and hemolysis improved in about 1 week.

Concerning ITP, 4 patients experienced a clinically significant platelets (PLT) reduction. Patient number five and number six were two elderly male subjects (81-year-old and 78-year-old) on low-dose eltrombopag who experienced a severe relapse (PLT 21 and 29 × 10⁹/L, 80% and 90% decrease from pre-vaccine, respectively) with mucosal bleeding, 9 days after the first and 7 days after the second dose of Pfizer vaccine, respectively. Patient number six had a concomitant reactivation of his chronic bronchitis. Both were rescued by increasing eltrombopag dose and with the addition of prednisone 1 mg/kg day. Patient number five deferred the second dose due to hip fracture requiring surgery. Patient number 7 was a 63-year-old man on low-dose steroids plus eltrombopag and rivaroxaban for a previous pulmonary embolism. He experienced a PLT drop to 40 × 10⁹/L (40% reduction) 10 days after the first dose of Pfizer vaccine and required eltrombopag and steroid increase. The second dose was administered without adverse events. Patient number eight was a 70-year-old woman on steroid therapy for persistent ITP. Her PLT decreased to 20 × 10⁹/L (70% reduction) after the second dose of Pfizer vaccine and the patient was rescued with intravenous immunoglobulin and steroids increase.

Patients with AIN and AA had no significant changes in their hematologic values (one AIN had a neutrophil decrease by 30% but was consistent with previous oscillations) and required no treatment changes. Finally, the following non-hematologic adverse events were observed (all grade I): fever (11%), pain at the injection site (21%), arthralgia (3%), headache (2%), and vomiting (1%), without significant differences between the first and the second dose.

Our data show that SARS-CoV-2 vaccination may be associated with a clinically significant decrease of hematologic values in 7.4% of AIHA and ITP cases that were rapidly rescued by treatment adaptation. Mild decrease was observed in about 10% of AIHA and 20% of ITP, requiring no treatment change. Notably, AIC recrudescences were not predictable, since they occurred in both patients on active treatment and off therapy, independently from AIHA type, after either the first or the second dose, and regardless vaccine type. Regarding available literature, three AIHA patients developing/reactivating AIHA after SARS-CoV-2 vaccine have been reported so far: one cAIHA exacerbation after the first and second dose of Pfizer vaccine¹⁶ and two severe wAIHAs developing after the first and the second dose of Pfizer vaccine. All cases have been successfully managed with steroids and transfusion support.^{14, 15} More data are available for ITP, including 21 patients (15 Pfizer, four Moderna, two Johnson&Johnson) from nine case reports/series.^3-11 Most patients rapidly improved with steroids or adjustment of ongoing treatment, including TPO-RA that may reduce the need of immunosuppression. The latter may abate the immune response to vaccines,¹⁷ although further studies are needed to assess vaccination efficacy under immunosuppressive treatment. Additionally, two large epidemiological studies in Scotland and USA^{12, 13} estimated an incidence of ITP of 0.8 million doses with Pfizer BioNTech, and of 1.71 million doses of AstraZeneca. This incidence was inferior to that expected for primary ITP in the general population (3.3 cases per 100 000 adults per year).^{12, 13}

In conclusion, our study along with the evidence above, underlines that post-vaccine AIC flares are manageable and that the benefits of vaccination greatly outweigh the risks. The hematologic monitoring adopted in our survey (1 week before, 1 week after the first and the second dose) appears appropriate to promptly intercept and manage AIC reactivations. We also avoided modifications of the ongoing AIC treatment in the 2–3 weeks preceding vaccine to prevent AIC exacerbations unrelated to vaccination. One of the main concerns is whether to administer the second dose after an AIC flare following the first one. Apart from patients' refusal, the second dose should be carefully weighed, and may be feasible upon close monitoring of blood counts and therapy adjustment. All these measures will ensure a safe vaccination campaign in this patient population.

SARS-CoV-2 感染和疫苗接种都引起了对免疫介导疾病的关注，包括自身免疫性血细胞减少症（AIC，即自身免疫性溶血性贫血，AIHA；免疫性血小板减少症，ITP；自身免疫性中性粒细胞减少症，AIN；再生障碍性贫血，AA；以及它们的组合，称为埃文斯综合征，ES）。自身免疫性血细胞减少症是高度异质性的疾病，其严重程度和临床病程各不相同。它们的特点是多次复发，通常由包括疫苗在内的免疫激活事件（感染、创伤、手术）引发。^{1, 2 已经描述了}一些关于 SARS-CoV-2 疫苗（mRNA 疫苗 Pfizer-Biontech 和 Moderna，以及基于腺病毒的疫苗阿斯利康）后 ITP 和 AIHA 恶化的报告，^3-16但缺乏对其管理的循证指征。

在这里，我们前瞻性地研究了在意大利米兰的参考血液学中心前瞻性随访的 108 名 AIC 患者（56 名 AIHA、41 名暖型、wAIHA 和 15 名冷型、cAIHA、38 名 ITP、7 名 AIN 和 7 名 AA）从 2021 年 3 月 24 日至 2021 年 6 月底接受了 SARS-CoV-2 疫苗接种。该研究是根据赫尔辛基宣言进行的，每位参与者都已签署书面知情同意书以收集数据。在每次接种疫苗前一周和后一周用全血计数（和 AIHA 中的 LDH 水平）监测患者（中位年龄 62 岁，范围 25-89 岁，女性/男性比例 1.7）。重要的是，正在进行的 AIC 治疗（38% 的病例，包括类固醇、环孢素、艾曲波帕和补体抑制剂 sutimlimab）在首次给药前 2 周内保持稳定。

表 1 总结了 ITP、AIHA、AIN 和 AA 患者每次给药后观察到的血液学趋势和副作用。7 名患者患有 ES，其中 4 名 ITP 加 AIHA，1 名 ITP 加中性粒细胞减少症。患者主要接受辉瑞-BioNtech 疫苗（N  =90），其次是 Moderna（N  =16）和阿斯利康（N  =2）。血液学参数在基线和疫苗接种后均表现出广泛分布。为了更好地研究患者内的变异，我们计算了第一次和第二次疫苗接种后 AIHA 和 ITP 患者 PLT 的 Hb 和 LDH 变化 delta%（支持信息图 S1）。

关于 AIHA，四名老年患者经历了临床上显着的 Hb 降低，需要调整治疗。详细来说，第一例患者是一位 79 岁的暖型 AIHA (wAIHA) 女性，在第​​一剂辉瑞疫苗后，Hb 从 10.4 降至 9.1 g/dL（LDH 1.2 至 1.7 × ULN）；她需要稍微增加类固醇剂量（至每天 5 毫克泼尼松），并在第二次给药后保持稳定。第二名患者是一名 73 岁男性，患有 wAIHA，停止治疗；他在第一剂 Moderna 疫苗（Hb 13.9 至 9.1 g/dL，LDH 1.1 至 1.6 × ULN）后 1 周明显复发，需要泼尼松 0.5 mg/kg 天并迅速反应。患者拒绝接受第二剂。第三名患者是一名 77 岁男性，患有感冒型 AIHA (cAIHA)，Hb 从 9.3 降至 7。2 g/dL 在第一剂 Moderna 疫苗后 1 周，需要类固醇、利妥昔单抗和重组促红细胞生成素。第二次给药没有出现不良事件。4 号患者是一名 73 岁男性，在服用低剂量类固醇后患有多次复发 wAIHA，在第二次给药后 5 天出现严重复发（Hb 14 至 7.4 g/dL，LDH 1.1 至 2.3 × ULN）辉瑞疫苗。他需要大剂量静脉注射类固醇，溶血在大约 1 周内得到改善。3 × ULN) 第二剂辉瑞疫苗后 5 天。他需要大剂量静脉注射类固醇，溶血在大约 1 周内得到改善。3 × ULN) 第二剂辉瑞疫苗后 5 天。他需要大剂量静脉注射类固醇，溶血在大约 1 周内得到改善。

关于 ITP，4 名患者出现了临床上显着的血小板 (PLT) 减少。5 号和 6 号患者是两名接受低剂量艾曲波帕治疗的老年男性受试者（81 岁和 78 岁），他们经历了严重的复发（PLT 21 和 29 × 10 ⁹/L，分别比接种疫苗前减少 80% 和 90%），并分别在第一剂辉瑞疫苗后 9 天和第二剂后 7 天出现粘膜出血。6 号患者的慢性支气管炎同时复发。两者均通过增加艾曲波帕剂量和添加泼尼松 1 mg/kg 天而获救。由于髋部骨折需要手术，五号患者推迟了第二次给药。7 号患者是一名 63 岁男性，因既往肺栓塞，服用低剂量类固醇加艾曲波帕和利伐沙班。他经历了 PLT 下降到 40 × 10 ⁹/L（减少 40%）在第一剂辉瑞疫苗和需要艾曲波帕和类固醇增加后 10 天。第二次给药没有出现不良事件。8 号患者是一名 70 岁的女性，因持续性 ITP 接受类固醇治疗。在第二剂辉瑞疫苗后，她的 PLT 下降至 20 × 10 ⁹ /L（下降 70%），患者通过静脉注射免疫球蛋白和类固醇增加获救。

AIN 和 AA 患者的血液学值没有显着变化（一个 AIN 的中性粒细胞减少了 30%，但与之前的波动一致）并且不需要改变治疗。最后，观察到以下非血液学不良事件（均为 I 级）：发烧 (11%)、注射部位疼痛 (21%)、关节痛 (3%)、头痛 (2%) 和呕吐 (1%) )，第一剂和第二剂之间没有显着差异。

我们的数据显示，在 7.4% 的 AIHA 和 ITP 病例中，SARS-CoV-2 疫苗接种可能与血液学值在临床上显着下降有关，这些病例通过治疗适应迅速获救。在大约 10% 的 AIHA 和 20% 的 ITP 中观察到轻度下降，不需要改变治疗。值得注意的是，AIC 复发是不可预测的，因为它们发生在接受积极治疗和停止治疗的患者中，独立于 AIHA 类型，在第一剂或第二剂后，无论疫苗类型如何。关于现有文献，迄今为止已报告了三名 AIHA 患者在接种 SARS-CoV-2 疫苗后出现/重新激活 AIHA： 第一剂和第二剂辉瑞疫苗后出现 1 例 cAIHA 恶化¹⁶在第一剂和第二剂辉瑞疫苗后出现两个严重的 wAIHA。所有病例均已通过类固醇和输血支持成功治疗。^{14, 15}更多关于 ITP 的数据是可用的，包括来自 9 个病例报告/系列的 21 名患者（15 名辉瑞、4 名 Moderna、2 名强生）。^3-11大多数患者在使用类固醇或调整正在进行的治疗后迅速改善，包括可能减少免疫抑制需要的 TPO-RA。后者可能会减弱对疫苗的免疫反应，¹⁷尽管需要进一步研究以评估免疫抑制治疗下的疫苗接种效果。此外，苏格兰和美国的两项大型流行病学研究^12、13估计辉瑞 BioNTech 的 ITP 发生率为 80 万剂，阿斯利康的 ITP 发生率为 171 万剂。这一发病率低于普通人群中原发性 ITP 的预期发病率（每年每 10 万成人 3.3 例）。^12、13

总之，我们的研究与上述证据一起强调，疫苗接种后的 AIC 发作是可控的，并且疫苗接种的好处大大超过了风险。我们调查中采用的血液学监测（第一次和第二次给药前 1 周，第一次和第二次给药后 1 周）似乎适合及时拦截和管理 AIC 再激活。我们还避免在疫苗接种前 2-3 周对正在进行的 AIC 治疗进行修改，以防止与疫苗接种无关的 AIC 恶化。一个主要问题是是否在第一次 AIC 发作后进行第二次给药。除了患者拒绝外，应仔细权衡第二次剂量，并在密切监测血细胞计数和治疗调整后可能可行。