A retrospective analysis of response rates and predictors of response to rituximab in management of primary immune thrombocytopenia: Second line and beyond

INTRODUCTION

Thrombocytopenia is a common hematological abnormality defined by a platelet count of <150 × 10⁹/L and needs evaluation once it is <100 × 10⁹/L. Causes of thrombocytopenia are increased destruction, reduced production, or peripheral pooling of platelets. Immune thrombocytopenia (ITP), previously known as ITP purpura, is a diagnosis of exclusion, and this hematological disorder is characterized by immune-mediated increased destruction along with decreased production of platelets.^[1] ITP can be primary (idiopathic) or secondary (due to human immunodeficiency virus [HIV], hepatitis C virus [HCV], or autoimmune disorders). Based on the duration of symptoms, it can be newly diagnosed or recent onset (<3 months), persistent (3–12 months), or chronic (>12 months). Pediatric ITP usually recovers in up to 80% of cases; contrary to this, adult ITP usually has a chronic relapsing-remitting course.^[1]

Steroids, intravenous immunoglobulins, and anti-RhD are the first-line treatment options for primary ITP, while secondary ITP needs the treatment of the underlying cause. Treatment indications are based on symptoms rather than platelet count, but it is prudent to start treatment once platelet count is below <30 × 10⁹/L because platelet count below this threshold is associated with increased bleeding. Only 60–70% of patients respond to initial steroid treatment, and most of them relapse during steroid tapering. Steroids cannot be given for more than 4–6 weeks due to associated side effects.^[1]

Trombopoietinmimetics (TPO-mimetics), eltrombopag and romiplostim, rituximab, and splenectomy are feasible options in the second line.^[2] Splenectomy is the most effective option but is usually deferred in the 1^st year of diagnosis and is associated with perioperative complications as well as increased incidence of infections in the postoperative period. Eltrombopag and romiplostim have very high complete response (CR) rates, but they are costly and need to be given for an indefinite period. Long-term remission after stopping these drugs is reported in a very small percentage of patients. Azathioprine and mycophenolate mofetil have low response and complete remission rates and take 3–6 months for optimal responses to occur.

Rituximab is a monoclonal, chimeric antibody targeted against CD20, an antigen present in B-lymphocytes. It has been found to be effective in primary and secondary ITP as a single agent and in primary ITP in combination therapy for upfront treatment of acute ITP. It is not recommended as the first-line treatment modality for ITP yet. Most studies use either low-fixed dose (100 mg) or standard dose (SD) (375 mg/m²) weekly ×4 doses along with steroids as one of the second-line treatment options for ITP.^[2-4] Various studies from the West have been reported, but data on the role and place of rituximab in the management of ITP have not been studied adequately in India. We retrospectively analyzed the rituximab data of the past 8 years to see the response rates and its predictors, side effects, and long-term efficacy of rituximab in primary ITP patients.

MATERIAL AND METHODS

• Type of Study: Retrospective analysis and single-centre study

• Department and Institute: Department of Clinical Hematology, King George’s Medical University Lucknow

• Duration: Patient data form year 2012 to 2019 were analyzed.

RESULTS

Twenty-five patients receiving rituximab, post-first-line treatment for primary ITP, between year 2012 and 2019, were retrospectively analyzed in this study. Baseline characteristics of the patients are given in Table 1.

Overall response rate (ORR), CR and partial response (PR)

Two patients (8%) received rituximab as the second-line treatment, 14 (56%) as the third-line treatment, and 9 (36%) as the fourth-line treatment. Figure 1 shows the treatment pattern of all 25 patients, along with a line of treatment and responses.

Close

Figure 1:

Showing recruitment of the patients for the study. Each number is consists of number of the patients. Boxes in red, yellow and green colours shows no response, partial response and complete response, respectively. (AZA: Azathioprine, DANA: Danazol, MTCO: Mycophenolate Mofetil, ELTRO: Eltrombopag, RITU: Rituximab, ROMI: Romiplostim, SPLEN: Splenectomy, IVIg: Intravenous Immunoglobuline, VIN: Vincristine).

Export to PPT

Response rates were defined as standard criteria for response in ITP. (R) Overall, the response was seen in 18 patients (72%), while seven patients did not get any response. At the end of 2 weeks from the first dose, ORR was 52%, which increased to 80% (CR-48%, PR-32%) at 4 weeks. At 8 weeks, 50% (four out of eight) of patients achieving PR lost the response. Figure 2 shows the monitoring of the pattern of response over a period of 6 months. Considering the sustained rise of platelet counts >50 × 10⁹/L, as one of the important response criteria in ITP studies, 40% of patients could achieve this response at 2 weeks, which increased to 60% at week 4 and then sustained at 40% at 6 months and later follow-ups. At a median follow-up of 47.6 months, 10 patients (40%) were still in CR, and 5 patients (20%) were maintaining PR with ORR of 60%. No response was seen in 7 patients (28%), and 3 patients (12%) relapsed with overall failure rates of 40%.

Close

Figure 2:

Stacked bar chart showing dynamic changes in the response to the rituximab therapy over time in comparison to the final response (CR: Complete response, PR: Partial response, NR: No response, wk: Weeks, mn: Months).

Export to PPT

We analyzed the factors determining the CR rates at the last follow-up, including age, sex, dose of rituximab (low dose [LD] vs. high dose), lines of treatment (≤2 lines vs. >2 lines of treatment), early (<12 months) or late (>12 months) initiation of rituximab therapy, and CR at 2 weeks and CR at 4 weeks. Age, sex, dose of rituximab, and lines of treatment did not affect the final CR rates, but there was a trend toward better CR rates in patients age> 20 years and female sex. CR at 2 and 4 weeks was significantly associated with the long-term CRs. Table 2 shows the various factors and their significance in determining the long-term CRs.

Table 2: Showing comparison of the various variables with the Complete Response at the last follow up. Fisher exact test was used. Parenthesis correspond to the percentage of the patients.

	CR (%)	PR or NR (%)	Total (%)	P-value
Age
<20 years	1 (11.1)	9 (56.2)	10 (40.0)	0.055
20–40 years	4 (44.4)	5 (31.2)	9 (36.0)
>40 years	4 (44.4)	2 (12.5)	6 (24.0)
Sex
Female	7 (77.8)	10 (62.5)	17 (68.0)	0.661
Male	2 (22.2)	6 (37.5)	8 (32.0)
Rituximab therapy
Early (<6 months)	2 (22.2)	7 (43.8)	9 (36.0)	0.401
Late (>6 months)	7 (77.8)	9 (56.2)	16 (64.0)
Line of treatment
≤2 lines	7 (77.8)	9 (56.2)	16 (64.0)	0.401
>2 lines	2 (22.2)	7 (43.8)	9 (36.0)
Rituximab dose
Low dose	4 (44.4)	9 (56.2)	13 (52.0)	0.44
High dose	5 (55.6)	7 (43.8)	12 (48.0)
CR at 2 week
Yes	6 (66.7)	1 (6.2)	7 (28.0)	0.003
No	3 (33.3)	15 (93.8)	18 (72.0)
CR at 4 week
Yes	7 (77.8)	5 (31.2)	12 (48.0)	0.033
No	2 (22.2)	11 (68.8)	13 (52.0)

Fisher exact test was used. Parenthesis corresponds to the percentage of the patients. CR: Complete response, PR: Partial response, NR= No response

We analyzed PFS and the factors influencing it. Median PFS was 19 months, and at 1 and 2 years, it was 33.3% and 30.4%, respectively. Early responses, including overall response at 2 weeks, 4 weeks, early use of rituximab at <12 months, and age >20 years, had significantly better PFS. Sex, lines of treatment, and rituximab dose (SD vs. LD) did not affect the PFS. Figure 3 shows the Kaplan–Mayer curves of PFS in patients receiving rituximab.

Close

Figure 3:

Kaplan Meir curves show progression-free survival for an overall response (complete response and partial response). A: Age Group, B: Rituximab used <12 or ≥12 months of diagnosis; C: Response at two weeks; D: Response at four weeks. Log Rank (Mantel- Cox) test was applied and a P value <0.05 was considered significant. While, sex, Line of treatment(second line or later), Rituximab Dose (low or high), Rituximab used (within six months or later of diagnosis) were not found significant.

Export to PPT

At the median follow-up of 47.6 months, mortality was 8%. One patient died due to refractory disease and central nervous system hemorrhage, and another patient died due to cellulites related sepsis and renal failure.

DISCUSSION

Initially, Stasi et al. studied the role of rituximab in 2001 in 25 cases of chronic ITP treated with 4 weekly infusions at a dose of 375mg/m².^[2] ORRs and sustained responses were 52% and 28%, respectively. No clinical or laboratory parameters predicted responses, but females and young age fared better. Miyakawa et al. studied the role of SD rituximab weekly for 4 weeks in patients with primary ITP who have relapsed or are refractory to previous lines of treatment.^[3] The primary endpoint was to study the proportion of patients maintaining platelet count >50 × 10⁹/L at 24 weeks. Out of 27 patients, 30.8% were able to maintain a platelet count above the target point. Similarly, in our study, 40% of patients were able to maintain a platelet count ≥50 × 10⁹/L at 24 weeks.

Another important multicenter study by Dierickx et al. studied the role of rituximab in autoimmune hemolytic anemia and ITP in Belgium.^[4] Out of a total of 40 patients with ITP, 12 had secondary ITP, and only 28 were primary ITP. Ninety-five per cent of patients received SD rituximab weekly for 4 weeks. ORR after the first course of rituximab was 70%, with 62.5% CR and 7.5% PR. Median PFS was 24.1 months, and at the end of 1 and 2 years, PFS was 70% and 44%. Age, sex, prior therapy, and duration from diagnosis to rituximab did not affect the outcome. Our patients also had an overall response of 80% at 4 weeks and 60% at 6 months. Median PFS was 19 months, and at 1 and 2 years, it was 33.3% and 30.4%, respectively. Early responses, including overall response at 2 weeks, 4 weeks, early use of rituximab at <12 months, and age >20 years, had significantly better PFS. Sex, lines of treatment, and rituximab dose (SD vs. LD) did not affect the PFS, although there was a trend toward better response in the female sex. At a median follow-up of 47.6 months, ORR was 60%, with CR at 40% and PR in 20% of patients, indicating that these patients are responsive to further lines of treatment, including thrombopoietin receptor agonists and splenectomy. Table 3 shows the most relevant studies with rituximab doses at 375 mg/m² and available data of overall response, response at different time points, and response at the last follow-up.^[5-10]

LD rituximab has been studied as the first-line treatment along with high-dose dexamethasone as well as post-first-line treatment of ITP. A study by Zhou et al. evaluated the role of rituximab and high-dose dexamethasone as the first-line treatment options with 100% ORR at day 28 and responses of 83.3% and 61.5% at 6 months and 12 months, respectively, with adverse event incidence of 11.1%.^[11] Zaja et al. studied LD rituximab in 48 patients, mostly chronic ITP, with ORR of 60.5% and CR rates of 39.5%. Overall responses at 12 and 24 months were 61% and 45%, respectively.^[12] In another study, he compared LD rituximab with SD rituximab. Overall response and response at the last follow-up was 52% and 23% in the LD rituximab arm compared to 66% and 35% in the SD rituximab arm. Contrary to this, Gracie et al. found similar results in LD rituximab versus SD rituximab with ORR of 61% versus 64% at 6 months and 50% versus 40% at 3 years follow-up, respectively.^[13] In our study, 13 patients received LD (100 mg fixed weekly dose) rituximab with ORR-47.1% compared to 52.9% in patients receiving SD rituximab (P = 0.67). Recently, a meta-analysis by Li et al. showed that pooled ORR was 63% and pooled CR was 44%.^[14] This analysis included 329 patients from nine studies. Thirtyone patients had side effects, and 30 of them were mild to moderate, no deaths were reported. This clearly indicates that LD rituximab is an effective and safe option.

CONCLUSION

With our study results and available literature, it is clear that rituximab still has a very important role in the management of persistent and chronic ITP. Overall response is up to 80%, and long-term response ranges between 30% and 40%. Early response at 2 and 4 weeks, early use of rituximab (<1 year), and age more than 20 years correlate with significantly better outcomes. The dose of rituximab (SD vs. LD) and prior lines of treatment do not affect the outcome. Side effects associated with rituximab are mainly infusion-related reactions, and these are easily manageable. Due to the definite duration of treatment and lower cost of therapy, rituximab will remain a feasible post-first-line treatment option in the near future.