Prognostic factors for survival in tandem vs single auto-SCT — a retrospective study

There remains a question mark over the survival benefit of single vs tandem autologous stem cell transplantation (auto-SCT) in patients with newly diagnosed multiple myeloma (NDMM). While tandem auto-SCT is used routinely in the MM setting, it has not been verified which patients benefit from the additional SCT and which do not.

A study by Joanna Blocka and colleagues evaluated which patient response groups, following the first auto-SCT, would benefit from a second (tandem) auto-SCT. It was hypothesized that improvement of response status rather than depth of response following initial auto-SCT would impact patient survival in tandem vs single auto-SCT.

Study design

• Patients with NDMM treated with high-dose chemotherapy and single or tandem auto-SCT between June 1996 and May 2016 (N = 978) were included in the analysis
  • Trial patients were from the German Speaking Myeloma Multicenter Group (GMMG) HD4 trial and received transplant in Heidelberg University Hospital or other GMMG centers (n = 323)
  • Non-trial patients (NTP) received transplant at Heidelberg University Hospital (n = 655)
• Primary endpoints: progression-free survival (PFS) and overall survival (OS)
• Response improvement was defined as reaching ≥ 1 response level deeper than that determined prior to auto-SCT
• Induction therapy regimens included
  • vincristine, doxorubicin, and dexamethasone (VAD)
  • bortezomib, doxorubicin, and dexamethasone (PAD)
  • thalidomide, doxorubicin, and dexamethasone (TAD)
  • bortezomib, cyclophosphamide, and dexamethasone (VCD)
  • novel agents, such as bortezomib, thalidomide, lenalidomide
• A multivariate analysis was conducted to determine the relationship between response improvement to auto-SCT and age, novel agents used during induction or maintenance therapy, elevated lactate dehydrogenase (LDH) level, International Staging System (ISS) score, and cytogenetic profile

Results

Patient characteristics

• Of the total patient cohort (Table 1), 53% received a single auto-HCT and 47% received tandem auto-HCT

Table 1. Baseline characteristics and treatment history of patients receiving single and tandem auto-SCT¹

auto-SCT, autologous stem cell transplantation; GMMG-HD4, German Speaking Myeloma Multicenter Group HD4 trial; ISS, International Staging System; NTP, non-trial patients *Induction with vincristine, doxorubicin, and dexamethasone plus thalidomide maintenance vs induction with bortezomib, doxorubicin, and dexamethasone plus bortezomib maintenance †≥ 1 of gain of 1q21, deletion of 17p13, translocation (4;14), or translocation (14;16)
Characteristic	Single auto-SCT (n = 518)	Tandem auto-SCT (n = 460)	p
Median age, years	61	56	< 0.0001
Cohort GMMG-HD4* NTP	11.2 88.8	57.6 42.4	< 0.0001
ISS stage I II III	47.3 26.5 26.2	41.3 35.1 23.6	0.06
Novel agents used in induction therapy Yes No	67.4 32.6	53.0 47.0	< 0.0001
Maintenance therapy Yes No	40.6 59.4	81.2 18.8	< 0.0001
Novel agents used in maintenance therapy Yes No	60.6 39.4	81.4 18.6	< 0.0001
High risk cytogenetics† Yes No	46.9 53.1	43.5 56.5	0.42

 Response to auto-SCT

• Table 2 presents patient responses prior to first auto-SCT (following induction therapy) and after first auto-SCT
• When comparing the characteristics of patients who improved response to first auto-SCT with those who did not
  • the number of patients receiving induction therapy with novel agents, such as thalidomide, bortezomib, and lenalidomide, compared to those who did not significantly differed (p = 0.02)
  • No significant differences were observed in relation to sex, age, ISS, LDH levels, cytogenetic profile, and number of induction therapy

Table 2. Outcomes before and after first auto-SCT for patients receiving single/tandem auto-SCT¹

auto-SCT, autologous stem cell transplantation; CR, complete remission; MR, minimal response; nCR, near complete remission; PD, progressive disease; PR, partial response; SD, stable disease; VGPR, very good partial response *Given as % of patients within the single or tandem groups
Response, % of patients*	Single auto-SCT		Tandem auto-SCT
	Before (n = 496)	After (n = 518)	Before (n = 444)	After (n = 460)
CR	2.4	5.4	4.7	11.3
nCR	24.4	38.0	3.6	6.3
VGPR	17.5	20.1	16.0	23.0
PR	36.3	28.6	44.4	45.2
MR	8.5	5.8	13.5	10.9
SD	4.8	2.1	11.0	3.3
PD	6.0	—	6.8	—

Tandem vs single auto-SCT

• By the median follow-up (86 months), 445 deaths and 765 PFS events were recorded
• PFS, but not OS, was significantly improved in patients that had undergone tandem vs single auto-SCT (PFS: hazard ratio [HR] 0.80 [95% CI, 0.67 –0.96], p = 0.01)
  • Interaction tests showed depth of response was not associated with significantly different survival benefit to tandem auto-SCT
• Multivariable analysis of all patients showed
  • patients achieving a very good partial response (VGPR) following first auto-SCT had improved PFS with tandem vs single auto-SCT (HR 0.50 [95% CI, 0.35–0.73], p < 0.001) but not OS
  • no PFS or OS benefit of tandem auto-SCT was observed for patients in CR, nCR, ≤ PR after first auto-SCT
  • PFS and OS after second auto-SCT was significantly superior in patients achieving an improved response to first auto-SCT vs those who did not
    • PFS: HR 0.64 (95 % CI, 0.48–0.85), p = 0.002
    • OS: HR 0.69 (95 % CI, 0.48–0.99), p = 0.04
    • In patients who did not improve their response after first auto-SCT, no benefit in terms of PFS or OS was observed
  • Cytogenetic characteristics did not significantly influence PFS or OS, irrespective of auto-SCT treatment in interaction tests

Prognostic factors for survival

• When considering the entire study cohort, a number of prognostic factors for OS and PFS were observed (Table 3)
• Multivariate analysis of patients achieving an improvement in response following first auto-SCT identified further factors that influenced PFS and OS (Table 4)
• Maintenance therapy did not have a significant impact on OS and PFS, irrespective of response following first auto-SCT

Table 3. Prognostic factors associated with PFS and OS¹

auto-SCT, autologous stem cell transplantation; CR, complete remission; ISS, International Staging System; LDH, lactate dehydrogenase; PFS, progression-free survival
	Prognostic factor		HR	95% CI	p
Positive prognostic factors	CR following 1st auto-SCT	PFS	0.65	0.49–0.87	0.004
Tandem auto-SCT	PFS	0.83	0.69–0.99	0.03
Negative prognostic factors	ISS II	PFS	1.46	1.20–1.78	0.0001
OS	1.90	1.45–2.51	< 0.0001
ISS III	PFS	1.36	1.11–1.67	0.003
OS	1.90	1.44–2.53	< 0.0001
High-risk cytogenetics	PFS	1.70	1.42–2.03	< 0.0001
OS	2.23	1.72–2.88	< 0.0001
Age	OS	1.02	1.01–1.04	0.001
Elevated LDH	OS	1.47	1.12–1.93	0.006

 Table 4. Multivariate analysis of prognostic factors for PFS and OS by patient response improvement to first auto-SCT¹

auto-SCT, autologous stem cell transplantation; ISS, International Staging System; LDH, lactate dehydrogenase; OS, overall survival; PFS, progression-free survival
Relationship	Prognostic factor		HR	95% CI	p
Patients achieving response improvement to first auto-SCT
Positive prognostic factors	Tandem auto-SCT	PFS	0.64	0.48–0.85	0.002
OS	0.69	0.48–0.99	0.04
Negative prognostic factors	ISS II	PFS	1.47	1.08–1.98	0.01
OS	1.99	1.29–3.07	0.002
ISS III	PFS	1.43	1.05–1.94	0.03
OS	2.22	1.45–3.40	0.0003
High-risk cytogenetics	PFS	1.80	1.35–2.39	< 0.0001
OS	2.28	1.56–3.33	< 0.0001
Patients not achieving response improvement to first auto-SCT
Negative prognostic factors	ISS II	PFS	1.36	1.04–1.77	0.02
OS	1.73	1.20–2.50	0.004
ISS III	OS	1.82	1.22–2.70	0.003
High-risk cytogenetics	PFS	1.53	1.21–1.94	0.0004
OS	2.16	1.54–3.03	< 0.0001
Age	OS	1.02	1.00–1.04	0.02
Elevated LDH	OS	1.45	1.01–2.07	0.04

Conclusion

Response improvement, but not depth of response, following initial auto-SCT is a significant prognostic factor for PFS following tandem vs auto-SCT. The authors note that this likely reflects a difference in patient susceptibility, at a cellular level, to high-dose chemotherapy.

Fewer patients in the population achieving an improved response following first auto-SCT had received novel agents as part of induction therapy. This may indicate that an improvement had already been achieved with these new regimens prior to first auto-SCT. While the use of novel agents in induction therapies make the results of this analysis less applicable to some clinical practices, economic limitations mean that the use of high-dose chemotherapy and auto-SCT are a core part of the treatment of patients with NDMM and therefore remain clinically relevant. By identifying a specific subset of patients who benefit from tandem auto-SCT, the results of this study may mean some patients can avoid the unfavorable side effects of a less effective therapy.