Zhu Qinze (朱千賾), Xio Hiyn (肖海燕), Liu Weiyi (劉為易), Qun Richeng (全日城),Tng Xudong (唐旭東), Lv Yn (呂 妍), Liu Chi (劉 馳), Li Liu (李 柳),Wng Hongzhi (王洪志), Guo Xioqing (郭小青), M Rou (麻 柔), Hu Xiomei (胡曉梅)*

aXiyuan Hospital, China Academy of Chinese Medical Sciences, Beijing, 100091, China

bGraduate School, China Academy of Chinese Medical Sciences, Beijing, 100700, China

ABSTRACT OBJECTIVE: To evaluated the benefits of oral administration of arsenic-containing Qinghuang Powder (QHP)compared with decitabine for patients with high/very-high (H/VH) risk myelodysplastic syndrome (MDS) according to the Revised International Prognostic Score System. METHODS: The OS (mOS) rate, annual OS rate and progression to acute myeloid leukemia (AML) in patients with H/VH MDS treated with QHP (QHP group, n = 27) and decitabine (decitabine group, n = 20) were retrospectively analyzed. The effects of prognostic factors of age, proportion of bone marrow blast,peripheral blood cell count, karyotype and Charlson Comorbidity Index (CCI) on OS were further analyzed. RESULTS: The mOS rate of QHP group (29 months) was significantly longer than that of the decitabine group (18 months) (P = 0.043). The OS rates of 1, 2, and 3 years were significantly higher in the QHP group (88.9%, 59.3%, 29.6%) than that in the decitabine group (70%, 25%, and 5%) (P = 0.01). There was no significant difference of 5-year OS rate between the 2 groups (P = 0.133).The effects of prognostic factors on mOS were further analyzed, and it was found that there was no significant difference of mOS rate between the QHP group (29 months) and the decitabine group (21 months) in the patients with age 65 years old(P = 0.673). The mOS rate was significantly longer in QHP group (28.5 months) than that in decitabine group (18 months) in the patients with age of ＜ 65 years old (P = 0.04). The proportions of bone marrow blast cells with10% or ＜ 10% had no significant effects on the mOS rate of patients in the 2 groups (P = 0.429, P = 0.183). In patients with HGB80 g/L, mOS rate was significantly longer in the QHP group (57 months) than that in the decitabine group (21 months) (P = 0.047), while in patients with HGB ＜ 80 g/L, there was no significant difference of mOS rate between the 2 groups (P = 0.265). In the patients with PLT ＜ 50×109 /L, the mOS rate was significantly longer in the QHP group (33 months) than that in the decitabine group(16 months) (P = 0.028). In the patients with PLT50×109 /L, there was no significant difference of the mOS rate between the 2 groups (P = 0.338). In the patients with ANC ＜ 0.8×109 /L, the mOS rate was significantly longer in the QHP group(20 months) than that in the decitabine group (7 months) (P = 0.014). In the patients with normal karyotype, the mOS was significantly longer in the QHP group (32 months) than that in the decitabine group (15 months) (P = 0.009). In the patients with abnormal karyotypes, there was no significant difference of the mOS rate between the 2 groups (P = 0.882). In the patients with good karyotypes, the mOS rate was significantly longer in the QHP group (37 months) than that in the decitabine group (20 months) (P = 0.019). In the patients with moderate/poor/very poor karyotype, there was no significant difference of the mOS rate between the 2 groups (P = 0.685). In the patients with CCI3, the mOS rate was significantly longer in the QHP group (34 months) than that in the decitabine group (10.5 months) (P = 0.017). In patients with CCI ＜ 3, there was no significant difference of the mOS rate between the 2 groups (P = 0.581). The proportion of progression to AML in the QHP group (18.8%) was significantly lower than that in the decitabine group (25%) (P = 0.03). CONCLUSION: Compared with decitabine, oral administration of arsenic-containing Qinghuang Powder could help patients to survive longer and decrease incidence of progression to acute myeloid leukemia in the treatment of patients with high/very high MDS.

KEYWORDS: Myelodysplastic syndrome; Qinghuang Powder; Realgar; Arsenic; Decitabine; Survival period

According to the current guidelines, recommended first-line treatment for high/very-high (H/VH)risk myelodysplastic syndromes (MDS) includes demethylating agents (e.g., azacitidine and decitabine),chemotherapy, supportive care, and hematopoietic stem cell transplantation (allo-HCT)[1]. However, allo-HCT is only indicated for patients younger than 65 years old who are also physically fit when there is a donor source, while demethylating agents (e.g., azacitidine and decitabine)have better effects than conventional chemotherapy only in improving hematologic response or increasing survival time[2], but the survival rate of 1 and 2 years is only 58% and 42%[3], and survival rate of 3 years is only 17.3%[4]. Qinghuang Powder is an oral arsenic-containing prescription of traditional Chinese medicine (TCM), and it was found in previous studies that Qinghuang Powder has significant therapeutic effects on MDS and is suitable for each subtype of MDS[5–10], and its efficacy is related to the genetic type of MDS[11], and its mechanism lies not only in removing the abnormal hypermethylation of MDS[12], but also in its ability to induce apoptosis of MDS/AML and promote erythroid differentiation[13,14].Long-term use is highly safe[15-17]. This study evaluated the benefits of oral administration of arsenic-containing Qinghuang Powder compared with decitabine in patients with high-risk/very high-risk myelodysplastic syndrome(MDS) based on the Revised International Prognostic Score System (R-IPSS)[18].

GENERAL INFORMATION

A total of 47 patients of high/very-high (H/VH) risk MDS patients treated in Xiyuan Hospital from December 2011 to December 2018 were selected, including 28 males and 19 females, aged from 27 to 87 years old, with a median age of 65 years old. There were 27 patients treated with Qinghuang Powder (QHP group) and 20 patients treated with decitabine (decitabine group). See Table 1.

METHODS

Treatment

QHP group: Take 1 capsule of Qinghuang Powder(Realgar 0.1 g; Indigo Naturalis 0.2 g) orally after meals per day for over 3 months.

Decitabine group: 7 of them take decibine 20 mg/m2, d1-d5; 13 take decitabine combined with small dose of cytarabine 12.5 mg, q12h, d1-d14, for over 4 courses.

Table 1. General information

Statistical methods

SPSS22.0 statistical software was used for data processing. The measurement data was performed by t-test; non-normally distributed data and counting data were performed by non-parametric test, and the median expression was described by quartile method [M (Q25,Q75)]. Survival analysis was performed using the Kaplan-Meier method for statistics and plotting survival curves. Survival time refers to the time from diagnosis to death or last follow-up. Patients lost to follow-up were treated according to censored data, and the loss rate was approximately 5%.

RESULTS

Comparison of OS rate between the 2 groups

To understand the overall survival situation of the 2 groups, the mOS and annual OS rates between the 2 groups were compared. The results showed that the mOS rate was 29 months (16, 41.25) in the QHP group and 18 months (10, 23.75) in the decitabine group. The mOS rate of patients in the QHP group was significantly higher than that of decitabine group (P = 0.043). See Figure 1A.The 1-, 2-, 3-, and 5-year OS rates were 88.9%, 59.3%,29.6%, and 7.4% in the QHP group, and 70%, 25%, 5%,and 5% in the decitabine group, respectively. The 1-, 2-,and 3-year OS rates in the QHP group were significantly higher than those in the decitabine group (P = 0.01);there was no significant difference in the 5-year OS rate between the 2 groups (P = 0.133). See Figure 1B.

Figure 1. Comparison of survival between the 2 groups. A: OS rate; B: annual OS rate

Effects of age on OS rate in both groups

To understand the effects of age on OS rate in both groups, the patients were divided into those aged65 years old and ＜ 65 years old. The results showed that in patients aged65 years old, the mOS rate was 29 months (15, 41) in QHP group and 21 months (13, 25) in decitabine group, respectively. There was no significant difference in mOS between the 2 groups (P = 0.673). See Figure 2A. In patients aged ＜ 65 years old, the mOS rate was 28.5 months (12.8, 37) in QHP group and 18 months(8.5, 19) in decitabine group, respectively. The mOS rate of patients in the QHP group was significantly higher than that in decitabine group (P = 0.04). See Figure 2B.

Effects of bone marrow blasts on OS rate in both groups

Figure 2. Comparison of survival between the 2 groups. A:65 years old; B: ＜ 65 years old

To understand the effects of the proportion of bone marrow blasts on patients' mOS rate, the proportion of bone marrow blasts was divided into10% and ＜ 10%.The results showed that in patients with10% bone marrow blasts, the mOS was 19 months (13, 35) in QHP group and 18 months (10, 32) in decitabine group,respectively, and there was no significant difference in mOS between the 2 groups (P = 0.429). See Figure 3A.In patients with ＜ 10% bone marrow blasts, mOS was 33 months (12, 57) in QHP group and 21 months (7,25) in decitabine groups, respectively, and there was no significant difference in mOS rate between the 2 groups(P = 0.183). See Figure 3B.

Effects of HGB level on OS rate in both groups

To understand the effects of HGB level on OS rate in the 2 groups, the patients were divided into HGB80 /L and HGB ＜ 80 g/L. The results showed that among the patients with HGB80 g/L, the mOS was 57 months (19, 68) in the QHP group and 21 months (16, 24) in decitabine group respectively, and the mOS in the QHP group was significantly better than that in decitabine group (P = 0.047). See Figure 4A. In patients with HGB ＜ 80 g/L, mOS was 28 months (13, 33)in QHP group and 23 months (7, 25) in decitabine groups respectively, and there was no significant difference in mOS rate between the 2 groups (P = 0.265). See Figure 4B.

Effects of PLT levels on OS rate in both groups

Figure 3. Comparison of survival between the 2 groups. A:10%; B: ＜ 10%

Figure 4. Comparison of survival rates between the 2 groups. A: HGB 80 g/L; B: HGB ＜ 80 g/L

To understand the effects of PLT levels on OS rate in the 2 groups, the patients were divided into PLT50×109/L and PLT ＜ 50×109/L. The results showed that among the patients with PLT50×109/L, the mOS was 34 months (19, 57) in QHP group and 25 months(10, 32) in decitabine group respectively, and there was no significant difference in mOS rate between the 2 groups (P = 0.338). See Figure 5A. In patients with PLT ＜ 50×109/L, mOS rate was 33 months (13, 35) in QHP group and 16 months (10, 21) in decitabine group respectively, and there was significant difference in mOS between the 2 groups (P = 0.028). See Figure 5B.

Figure 5. Comparison of patient survival between the 2 groups. A: PLT 50×109 /L; B: PLT ＜ 50×109 /L

Effects of ANC level on OS rate in both groups

To understand the effects of ANC on mOS rate in the 2 groups, the patients were divided into those with ANC0.8×109/L and those with ANC ＜ 0.8×109/L.The results showed that among patients with ANC0.8×109/L, the mOS was 33 months (20, 46)in QHP group and 25 months (16, 32) in decitabine group, and there was no significant difference in mOS rate between the 2 groups (P = 0.129). See Figure 6A. In patients with ANC ＜ 0.8×109/L, the mOS rate was 20 months (12, 35) in QHP group and 7 months (6, 15) in decitabine group, and there was significant difference in mOS rate between the 2 groups (P = 0.014). See Figure 6B.

Effects of karyotype on OS rate in both groups

To understand the effects of karyotype on mOS rate in both groups, the patients were divided into normal karyotype and abnormal karyotype. The results showed that in patients with normal karyotype, the mOS rate was 32 months (17, 57) in QHP group and 15 months (10, 23)in decitabine group, and there was significant difference in mOS rate between the 2 groups (P = 0.009). See Figure 7A. In patients with abnormal karyotypes, the mOS rate was 35 months (13, 39) in QHP group and 32 months(18, 35) in decitabine group, and there was no significant difference in mOS rate between the 2 groups (P = 0.882).See Figure 7B.

Figure 6. Comparison of survival between the 2 groups. A: ANC 0.8×109 /L; B: ANC ＜ 0.8×109 /L

Figure 7. Comparison of survival between the 2 groups. A: Normal chromosomes; B: Chromosomal abnormalities; C: Prognosis good karyotype; D: Prognosis moderate/poor/very poor karyotype

To understand the effects of chromosome prognostic karyotype on mOS rate in the 2 groups, the patients were divided into good prognostic karyotype and moderate/poor/very poor prognostic karyotype, and the results showed that among the patients with good karyotype prognosis, the mOS rate was 37 months (20, 53) in QHP group and 20 months (7, 32) in decitabine group, and there was significant difference in mOS rate between the 2 groups (P = 0.019). See Figure 7C. In patients with intermediate/poor/very poor prognosis karyotype, mOS rate was 17 months (13,20) in QHP group and 18 months(16, 32) in decitabine group, and there was no significant difference in mOS rate between the 2 groups (P = 0.685).See Figure 7D.

Effects of CCI on OS rate in both groups.

To understand the effects of CCI level on mOS rate in both groups, the patients were divided into those with CCI3 and those with CCI ＜ 3. The results showed that in patients with CCI3, the mOS rate was 34 months(13, 42) in QHP group and 15.5 months (9.3, 24.3) in decitabine group, and the mOS rate was significantly better in QHP group than that in decitabine group (P = 0.017).See Figure 8A. In patients with CCI ＜ 3, the mOS rate was 28.5 months (14, 32.8) in QHP group and 21 months (11.5,25.3) in decitabine group, and there was no significant difference in mOS rate between the 2 groups (P = 0.581).See Figure 8B.

Comparison of progression rates to acute myeloid leukemia between the 2 groups

Figure 8. Comparison of survival between the 2 groups. A: CCI 3; B: CCI ＜ 3

To understand the progression to AML in the 2 groups, the incidence of AML in the 2 groups was compared. The results showed that of the 27 patients in QHP group, 5 progressed to AML, and the incidence of AML was 18.8%; of the 20 patients in decitabine group, 5 progressed to AML, and the incidence of AML was 25%.The incidence of AML in QHP group was significantly lower than that in decitabine group (P = 0.03). See Figure 9.

Figure 9. Comparison of progression rate to acute myeloid leukemia between the 2 groups

DISCUSSION

MDS is the result of a multistep process that includes changes in epigenetic mechanisms, such as DNA methylation, and for such reasons demethylating drugs are potential agents and decitabine has been widely used to treat MDS[19,20].

For patients with high or very high-risk (H/VH)MDS, current recommended first-line therapies include demethylating agents (e.g., azacitidine and decitabine),chemotherapy, supportive care, or hematopoietic stem cell transplantation (allo-HCT), but allo-HCT is only indicated for patients younger than 65 years old who are also in good physical condition and have a donor source[1].

IPSS-R risk stratification is a direct prognostic factor for MDS, and patients with H/VH MDS have a poor prognosis, with mOS rate of 1.6 years (19.2 months) and 0.8 years (9.6 months)[1]. The mOS rate of high-risk MDS patients treated with chemotherapy is 12.7 months[21],while those treated with azacitidine is 18 months[22],and the combination of the protein deacetylase inhibitor Pracinostat with azacitidine did not improve outcomes for patient with high-risk MDS, with an mOS rate of only 16 months[23]. This study showed that the mOS rate of decitabine group was 18 months, which was similar to the above reports, while the mOS rate of QHP group was 29 months, which was significantly higher than that of decitabine group (P = 0.043). Azacitidine has been used to treat high-risk MDS with 1- and 2-year OS rates of 58%and 42% respectively[3], and 3-year OS rate of 17.3%[4].Japanese studies have shown 2-year OS rates of patients with H/VH MD were 35% and 12% respectively[24]. In this study, the 1-, 2-, and 3-year OS rates in the decitabine group were 70%, 25%, and 5% respectively, while the 1-,2-, and 3-year OS rates in the QHP group were 88.9%,59.3%, and 29.6% respectively, which were significantly higher than those in the decitabine group (P = 0.01).

Age is the basis for the choice of treatment for patients with H/VH MDS[1], and in patients under 65 years old with good physical fitness, hematopoietic stem cell transplantation (allo-HCT) can be considered when there is a donor source. The results of this study showed that in patients with H/VH MDS aged under 65 years old, the mOS rate in the QHP group (28.5 months) was significantly longer than that in the decitabine group (18 months) (P = 0.04), providing an alternative treatment for patients with H/VH MDS without donor source.For patients with H/VH MDS aged over 65 years old or those who are in poor health, only supportive care or demethylation can be selected which also needs certain economic conditions. It has been reported that[4]age over 65 years old is one of the poor prognostic factors for demethylating treatment of H/VH MDS. The results of this study showed that in patients with H/VH MDS aged over 65 years old, there was no significant difference in mOS rate between the QHP group (29 months) and the decitabine group (21 months) (P = 0.695), providing an effective treatment for patients with H/VH MDS who are facing economic difficulties.

The proportion of bone marrow blasts represents how much the disease progresses. Studies[24]have suggested that the proportion of bone marrow blasts is one of the independent prognostic factors for poor OS in MDS patients. The results of this study showed no significant difference in mOS rate between the 2 groups with a bone marrow blast proportion of10% or ＜ 10%(P = 0.429), which is consistent with previous findings[9].It is suggested that Qinghuang Powder is suitable for MDS patients whose disease is in period of progression.

Peripheral blood cell counts are associated with the prognosis of MDS. A multicenter retrospective study showed[25]that low levels of ANC, HBG as well as PLT counts were associated with an increased incidence of infection in patients with H/VH MDS treated with azacitidine. Another study concluded[24]that low level of PLT is one of the independent prognostic factors for poor OS in MDS patients. The data from this study showed that in patients with HGB80 g/L, the mOS rate in the QHP group (57 months) was significantly longer than that in the decitabine group (21 months) (P = 0.047); in patients with ANC ＜ 0.8×109/L, the mOS rate in the QHP group (20 months) was significantly longer than that in the decitabine group (7 months) (P = 0.014); and in patients with PLT ＜ 50×109/L, the mOS rate in the QHP group (33 months) was significantly longer than that in the decitabine group (16 months) (P = 0.028). The results of this study suggest that Qinghuang Powder could further improve OS for H/VH MDS patients with HGB80 g/L or ANC ＜ 0.8×109/L or PLT ＜ 50×109/L.

MDS is a heterogeneous disease and the karyotype has independent prognostic significance[26]. A poor prognosis karyotype predicts poor OS[24]. This study showed that in patients with good prognosis karyotype,the mOS rate of QHP group (37 months) was significantly longer than that of decitabine group (20 months) (P =0.019), and the results showed that H/VH MDS patients with good prognosis karyotype who chose Qinghuang Powder could have better OS than decitabine; among patients with moderate/poor/very poor prognosis karyotype, the mOS rate of QHP group (17 months) was not significantly different from that of decitabine group(18 months) (P = 0.685), suggesting that it would also be a good choice to receive the treatment of Qinghuang Powder for those patients with H/VH MDS who received decitabine unconditionally.

The Charlson Comorbidity Index (CCI)combines age and comorbidity. The higher CCI means older patient, more comorbidities, and worse organ functions, suggesting that it is more difficult to perform hematopoietic stem cell transplantation, and the tolerance to chemotherapy will be poor. It has been shown in studies[27]that CCI is associated with poor OS in MDS patients. The results of this study showed that in patients with CCI3, the mOS rate was significantly longer in the QHP group (34 months) than in the decitabine group (15.5 months) (P = 0.017), suggesting that for those patients with H/VH MDS who were older and had more comorbidities, Qinghuang Powder is recommended.

The progression of high-risk/very high-risk MDS to AML is still worthy of research. A large series of case analyses showed[24]that the incidence of progression to AML in patients with high-risk and very high-risk MDS was 56% and 40% respectively. This study showed that the incidence of progression to AML in the QHP group(18.8%) was significantly lower than that in the decitabine group (25%) (P = 0.03).

Previous studies have shown that Qinghuang Powder not only has significant efficacy on early MDS[7,9,10,16], but also has long-term efficacy in patients with MDS with promyelocytic hyperplasia[9], and it not only has significant effects in the treatment of lowrisk MDS[7,8,10,16], but also can help patients with highrisk/very high-risk MDS survive for a long time[7,8,10,16].The studies in vitro revealed that the active component of Qinghuang Powder not only removes MDS aberrant hypermethylation[10], but also induces apoptosis and promote erythroid differentiation[11,12]in MDS/AML cells, which may relate to the mechanism of efficacy of Qinghuang Powder in the treatment of high-risk/veryhigh-risk MDS.

In conclusion, compared with decitabine, oral administration of Qinghuang Powder, an arseniccontaining prescription of traditional Chinese medicine,can help patients with high-risk/very-high-risk MDS achieve longer survival time and lower rates of progression to AML in patients, especially for patients aged under 65 years old, HGB80 g/L, ANC ＜ 0.8×109/L,PLT ＜ 50×109/L, with good prognosis karyotype, and CCI3.

ACKNOWLEDGMENT

This work is supported by National Natural Science Foundation of China (81673821); National Natural Science Foundation of China (81774142); Central Level Research Institute of Public Welfare Research Funds (ZZ10-016); National TCM Clinical Research Base Business Construction Second Batch of Scientific Research Projects (JDZX2015264).