Background

Peripheral neuropathy (PN) is associated with multiple myeloma (MM) and often exacerbated by anti-MM treatments. TIPN poses significant morbidity and may result in therapy modifications. The impact of TIPN on treatment patterns and outcomes is not well studied in “real world” settings. We examined the association between TIPN in NDMM with treatment changes and efficacy outcomes in community oncology practices.

Methods

This was a retrospective review of clinical practice data from 12 US practices. Included pts had NDMM between 1/1/2008 and 7/1/2010, received anti-MM therapy and were ≥ 18 years of age at initiation of front line therapy. Inclusion was stratified with approximately half of the sample having TIPN from initial therapy (i.e. before 1stdocumented disease progression), as confirmed through review of medical records. Front line therapies were classified as those that included: 1) Lenalidomide (L), 2) Thalidomide (T), 3) Bortezomib ± Thalidomide (B), 4) Lenalidomide + Bortezomib (LB), and 5) Other regimens (O). Multivariate Cox regression models were used to identify pt level risk factors for TIPN, including regimen. Cox models with time varying covariates were used to predict: a) time to dose reduction; b) time to delay, discontinuation or switch; c) time to progression (TTP), and d) overall survival (OS). TIPN was modeled as a binary, time varying predictor of each outcome (pt observation split by non-TIPN and TIPN intervals), controlling for demographic and clinical characteristics, and treatment regimen.

Results

Mean age of sampled patients (n=229) was 66.6 years (range 40-91), 49.3% were female, 58.5% Caucasian, and 34.9% African American. Of 229 pts, 112 experienced TIPN during front line therapy. Of the 117 without TIPN, 23 (19.7%) experienced PN during front line that was not attributed to treatment. Relevant comorbidities were higher in the TIPN than non-TIPN pts but not significantly different; diabetes (26.8% v 21.4%, p =0.360) and renal disease (26.8% v 20.5%, p=0.28). TIPN severity during front line treatment was reported as severe (22.3%), moderate (25.9%), mild (25.0%), and unknown (26.8%). Front line treatment included B: 110 (48.0%), L: 36 (15.7%), O: 31 (13.5%), T: 30 (13.1%) and LB: 22 (9.6%), with B uniformly administered IV rather than SC. TIPN occurred in 69.6% of B containing vs. 30.4% of non-B containing regimens.

Median duration of front line treatment ranged from 5.1 (O) to 6.74 months (R). Overall, 94 pts (41.0%) had dose reductions, and 117 pts (51.1%) had a delay, discontinuation or switch in therapy. Of the 117, 62 continued with front line therapy after the delay, discontinuation or switch. Overall median TTP was 27.7 months, and median OS was 47.9 months. Median time to TIPN in TIPN patients was 4.52 months. Regimen was a significant predictor of time to TIPN (p = .0056), with L (HR = .323, p = .0009) and O (HR = .455, p = .037) at lower risk than B. Table 1 shows that time varying TIPN was associated with dose reduction and treatment delay, discontinuation, or switch. Each analysis controlled for gender, age, presence of any relevant comorbidity, performance status, stage of disease, and regimen.

Table 1

Cox Regression Results for Analysis of Time Varying TIPN

Endpointp-valueHR for Time Varying TIPNHR 95% CI
Dose Reduction .001 2.359 1.414-3.936 
Treatment Delay, Discontinuation or Switch .006 1.824 1.186-2.806 
Time to Progression .341 1.201 0.823-1.753 
Overall Survival .176 0.734 0.469-1.149 
Endpointp-valueHR for Time Varying TIPNHR 95% CI
Dose Reduction .001 2.359 1.414-3.936 
Treatment Delay, Discontinuation or Switch .006 1.824 1.186-2.806 
Time to Progression .341 1.201 0.823-1.753 
Overall Survival .176 0.734 0.469-1.149 

For analysis of dose reduction, the only other significant pt level predictor was comorbid disease (HR = 1.903, p = .006). Other pt level covariates were not associated with predicting delay, discontinuation or switch. Gender was a significant predictor of TTP (HR = .643, p = .017), indicating lower risk for males. For analysis of OS, comorbid disease and age were significant (HR = 1.7, p = .02; 1 year HR = 1.035, p = .002, respectively), as was stage (p = .0038), with more advanced stage tending to be associated with higher risk.

Conclusions

This retrospective study of TIPN in MM showed that occurrence of TIPN was associated with increased risk of dose reduction, and of treatment delay, discontinuation or switch. TIPN was not significantly associated with TTP or OS, likely because most treatment modifications (including dose reduction) did not involve complete treatment discontinuation. Regimen combinations with a lower neurotoxicity burden should be considered in NDMM patients.

Disclosures:

Panjabi:Onyx Pharmaceuticals, Inc.: Employment, Equity Ownership.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution