Intravenous (IV) taxanes for metastatic breast cancer (mBC) are associated with toxicities, such as chemotherapy-induced peripheral neuropathy (CIPN), which can detrimentally impact outcomes.

 To assess the impact of CIPN on clinical and economic outcomes in women with mBC, initiating IV paclitaxel/ nab-paclitaxel.

 Adult women in the MarketScan Commercial and Medicare Supplemental Database with a mBC diagnosis, initiating IV paclitaxel or IV nab-paclitaxel (index date = first administration) from November 1, 2013, to September 30, 2018, who had no prior neuropathy diagnoses, and continuous enrollment 12 months prior to and ≥ 3 months following index were selected. Propensity score–matched CIPN and non-CIPN cohorts were defined, based on postindex CIPN diagnosis. Clinical characteristics and all-cause and breast cancer (BC)-related health care utilization and costs per patient per month (PPPM) were compared between matched CIPN and non-CIPN cohorts during follow-up.

 Among the 5870 women with mBC initiating IV paclitaxel/nab-paclitaxel, 42.7% developed CIPN. The matched cohorts each included 1950 women. Patients with CIPN were more likely to have a dose reduction (46.1% vs 38.2%, P < .001) or develop depression, diabetes, insomnia, liver dysfunction, or arthritis compared with the non-CIPN cohort, P < .05. Patients with CIPN were more likely to have an inpatient admission (39.2% vs 34.9%, P < .01) or emergency department visit (46.7% vs 35.6%, P < .001), as well as all-cause and BC–related costs that were $1102 and $725 PPPM higher, respectively, than women without CIPN (P < .01).

 CIPN was common in women, following IV paclitaxel/nab-paclitaxel treatment and was associated with dose reductions, the development of comorbidities, and elevated health care costs. Therapies for mBC that offer increased tolerability are needed to help improve patient outcomes and control costs.

 2021;27(Spec 1). Published online December 23, 2020. 

Women with metastatic breast cancer (mBC) who develop chemotherapy-induced peripheral neuropathy (CIPN) after initiating intravenous (IV) paclitaxel or IV nab-paclitaxel, utilize more health care services and incur increased costs compared to women who do not develop CIPN. Patients with CIPN also experience an increased comorbidity burden and more treatment alterations in comparison to patients without CIPN. Treatments for mBC with reduced rates of CIPN are needed, to help improve long-term outcomes and reduce the burden on the health care system.

42.7% of women developed CIPN following initiation of IV paclitaxel/nab-paclitaxel

Incremental all-cause and breast cancer-related health care costs were higher for patients who developed CIPN by $1102 and $725 per patient per month (PPPM), respectively.

Despite advancements in therapy, metastatic breast cancer (mBC) remains incurable. Treatment is focused on palliative care, with goals of prolonging survival and improving quality of life (QOL).

 Intravenous (IV) taxanes are some of the most active chemotherapies for the treatment of mBC.

 However, taxanes are largely nonspecific and associated with treatment-related toxicities, the development of which can result in treatment alterations (eg, dose reductions and regimen changes) or delays.

Chemotherapy-induced peripheral neuropathy (CIPN), which has been reported in >60% of patients with mBC treated with IV paclitaxel, is a common taxane toxicity.

 It is characterized by sensory and/or motor manifestations driven by peripheral nerve damage. The condition is believed to be caused by the loss of neuronal microtubule function, although solvents used in IV taxanes have also been implicated.

 The development of CIPN is dose-dependent and schedule-dependent; effects can also be cumulative and limit future treatment by requiring dose reductions, delays, or discontinuations.

 Furthermore, CIPN can become a chronic condition, leaving patients with long-term effects, including numbness, disability, and neuropathic pain.

In addition to their impact on patient outcomes, treatment-related toxicities, like CIPN, can place an increased burden on the health care system. Studies have reported increased costs for patients with chemotherapy-related toxicities compared to those without.

These costs can be substantial, with increases of up to 70% in monthly all-cause expenses when toxicities occur.

Although studies have assessed treatment-related toxicities in mBC, few have been focused specifically on CIPN.

 Similarly, analyses of CIPN have not been focused specifically on mBC.

 This retrospective analysis examined the clinical and economic burden of CIPN in patients with mBC, initiating IV paclitaxel or IV nab-paclitaxel.

This retrospective cohort study utilized deidentified data from the US IBM MarketScan Commercial and Medicare Supplemental databases from November 1, 2012, to December 31, 2018. The commercial database contains the inpatient, outpatient, and outpatient pharmacy details of employees and their dependents, covered under a variety of health plans. The Medicare database contains the full health care details (medical and pharmacy) of retirees with Medicare supplemental insurance provided by employers, including both Medicare-covered and employer-paid portions.

Disability data were obtained from the MarketScan Health and Productivity Management Database from November 1, 2012, to December 31, 2017 (end of available data). This database contains short-term and long-term disability data for a subset of patients in the MarketScan Commercial and Medicare Supplemental Databases. All study data were obtained using codes from the International Classification of Diseases, Ninth and Tenth Revisions, Clinical Modification (ICD-9-CM and ICD-10-CM); Current Procedural Terminology Fourth Edition; Healthcare Common Procedure Coding System; and National Drug Codes.

Adult women with greater than 1 inpatient claim, or greater than or equal to 2 nondiagnostic outpatient claims for breast cancer (BC) (ICD-9-CM: 174. xx; ICD-10-CM: C50.xx) were identified within the MarketScan Commercial and Medicare Database from January 1, 2014 to September 30, 2018. Eligible women had greater than or equal to 1 nondiagnostic claim for metastasis occurring 60 days prior to the first BC claim, through the end of data (excluded codes for metastasis to the breast). The index date was that on which the first IV paclitaxel or nab-paclitaxel was administered (identified via the Healthcare Common Procedure Coding System [HCPCS]; hereafter referred to as IV paclitaxel/nab-paclitaxel). Continuous enrollment was required for 12 months prior to, and 3 or more months following the index date. Women who had undergone taxane treatment or had a neuropathy diagnosis (

) in the preperiod were excluded, as were patients with greater than 1 taxane administration on the index date.

Patients with CIPN had 1 or more nondiagnostic claims with a diagnosis of neuropathy, paresthesia, myalgia, or arthralgia during follow-up, and were continuously enrolled for 3 or more months following the first neuropathy claim (to assess post-CIPN costs). Patients without CIPN had no neuropathy diagnoses during the study period. The final CIPN and non-CIPN cohorts were propensity score–matched based on preperiod characteristics, to control for demographic and clinical differences between groups. Covariates included evidence of any inpatient admission, length of follow-up, National Cancer Institute Charlson Comorbidity Index (NCI-CCI), use of neurotoxic agents, number of toxicities, number of chemotherapy agents administered, and evidence of endocrine therapy, radiation therapy, or BC–related surgery. A 1:1 match was implemented, with the nearest neighbor technique without replacement (

The preperiod was the 12 months preceding the index date. Outcomes were assessed over the follow-up period, which extended from the study index date until the earliest of the following: (1) administration of docetaxel; (2) end of continuous enrollment; or (3) end of the study period (December 31, 2018). For patients with CIPN, follow-up included the duration both prior to, and following the diagnosis of CIPN.

The index line of therapy (LOT) included chemotherapies administered in the 37 days following index. Non-taxane chemotherapies included capecitabine, carboplatin, cisplatin, cyclophosphamide, doxorubicin, epirubicin, eribulin, fluorouracil, gemcitabine, ixabepilone, methotrexate, and vinorelbine. The 37-day period allowed the inclusion of concomitant medications administered on a different cycle. The end of the Index LOT was defined by a change in the regimen (switch, augmentation, or drop) or the end of follow-up. Discontinuation of a regimen component was defined as a treatment gap of 45 days or more.

The IV paclitaxel/nab-paclitaxel dose changes were assessed during the Index LOT. Dose was estimated by dividing the paid amount on IV paclitaxel/nab-paclitaxel claims by the wholesale acquisition cost. Capitated claims, claims with a JW modifier, indicating medication waste, or a $0 paid amount were excluded. To assess dose changes over time, only patients with 3 or more non-$0 claims for IV paclitaxel/nab-paclitaxel were included. The magnitude of a dose change was defined as greater than or equal to 10% of the index dose. Changes were examined in relation to the prior dose.

Patient Characteristics and Comorbidities

Demographics were measured on the index date. During the preperiod, the NCI-CCI was calculated.

 Prevalent comorbid conditions were assessed during the preperiod and follow-up. During follow-up, incident comorbidities were examined among patients who did not experience the specific comorbidity during the preperiod.

Health Care Resource Utilization and Costs

All-cause and BC–related health care resource utilization and costs per patient per month (PPPM), including inpatient admissions, outpatient medical encounters (emergency department [ED] visits, outpatient office visits, and other outpatient services), and outpatient prescriptions were reported during follow-up. The BC–related health care resource utilization and costs were defined using inpatient admissions with a primary diagnosis of BC; outpatient services with a diagnosis of BC in any position; medical or pharmacy claims for BC treatments (chemotherapy, endocrine therapy, human epidermal growth factor receptor 2 [HER2] therapy, or radiation); and pharmacy claims for neuropathy medications (antidepressants, anticonvulsants, opioids and narcotics, systemic corticosteroids, nonopioid and nonsteroidal pain relief, topical local anesthetics, mexiletine, tocainide, and pentoxifylline). Costs were calculated using paid amounts on adjudicated claims. Costs for capitated plans were estimated via payment proxy. Costs were inflated to 2019 US$ using the medical care component of the Consumer Price Index.

The MarketScan Health and Productivity Management (HPM) Database was used to assess short-term and long-term disability and estimate indirect costs from November 1, 2012, to December 31, 2017. Because of the limited sample of patients with HPM data, the unmatched CIPN and non-CIPN cohorts were used for the work disability analysis. All eligible patients with available HPM data were included. The duration of the postindex period was based on the availability of HPM data that overlapped with the postindex period for the larger study. All outcomes were reported as PPPM. Indirect costs were estimated by applying an estimated daily wage to short-term or long-term disability days. Wages were calculated using sex and region-matched data from the US Bureau of Labor Statistics in 2017. Short-term and long-term disability was set as 70% and 60% of the wage rate per day, respectively.

Continuous variables were reported as mean and standard deviation. Categorical variables were reported as frequencies and percentages. All data analyses were conducted using the WPS version 4.1 (World Programming, United Kingdom).

Of the 5870 women with mBC, initiating treatment with IV paclitaxel/nab-paclitaxel, 2509 (42.7%) were diagnosed with CIPN during follow-up. The mean time from index to CIPN diagnosis was 198.9 (± 231.5) days. After matching, the CIPN and non-CIPN cohorts each comprised 1950 women, and the cohorts were well matched (

). The average follow-up was similar at 21.0 (± 12.1) months in the CIPN cohort, and 20.6 (± 12.9) months in the non-CIPN cohort. The time to first CIPN diagnosis was 169.4 (± 194.1) days, and 46.7% of the participants were diagnosed with CIPN during the Index LOT.

Most patients initiated IV paclitaxel (Table 1). Only 8.7% of the CIPN cohort and 11.9% of the non-CIPN cohort initiated IV nab-paclitaxel. The majority of patients with CIPN (82.9%) and those without CIPN (84.3%) received IV paclitaxel/nab-paclitaxel monotherapy. The duration of the Index LOT was also similar between cohorts at 96 (± 64) days in the CIPN cohort and 99 (± 63) days in the non-CIPN cohort. Patients with CIPN were more likely to end the Index LOT by dropping a medication (84.4% vs 81.2%, P = .010), whereas patients without CIPN were more likely to end the Index LOT because follow-up had come to an end (5.7% vs 2.3%, P < .001).

IV Paclitaxel/nab-Paclitaxel Dose Changes

Over 75% of the study participants qualified for the dose change analysis (CIPN, 77.4%; non-CIPN, 82.2%). Patients with CIPN were significantly more likely to undergo a dose reduction ≥ 10% during the Index LOT (46.1% vs 38.2%, P < .001, 

Among patients with a dose reduction, no significant differences were observed in the total number of, time to, or magnitude of dose reductions between cohorts (Table 2). However, patients with CIPN were more likely to receive a medication for neuropathy in +/- 10 days of a dose reduction, compared to patients without CIPN (40.8% vs 34.6%, P < .05).

Preperiod comorbidity burdens were similar between cohorts (Table 1). During follow-up, patients with CIPN had a significantly higher prevalence of depression, liver dysfunction, nonrheumatoid arthritis, insomnia, rheumatoid arthritis, falls, and fractures, compared to patients without CIPN (P < .05, 

). The CIPN cohort also had a significantly higher proportion of patients with incident depression, diabetes, insomnia, falls, fractures, liver dysfunction, and nonrheumatoid arthritis (P < .05, Table 3).

All-Cause Health Care Utilization and Costs

Patients with CIPN had higher health care service utilization and direct all-cause costs PPPM over the follow-up. They were also more likely to have 1 or more inpatient admissions (39.2% vs 34.9%, P < .01) or ED visit (46.7% vs 35.6%, P < .001) compared to patients without CIPN (Table 4). Patients with CIPN also showed a significantly greater number of monthly ED visits, office visits, other outpatient services, and outpatient pharmacy refills over follow-up (Table 4).

Greater utilization of health care services translated to increased all-cause health care costs PPP for patients with CIPN compared to patients without CIPN. During follow-up, patients with CIPN incurred total all-cause health care costs that were $1012 PPPM higher, compared to patients without CIPN ($11,181 ± $10,240 vs $10,169 ± 10,029, P < .01). The higher costs for patients with CIPN were derived from a significantly higher number of office visits, outpatient services, and outpatient pharmacy costs (

BC–Related Health Care Utilization and Costs

The CIPN cohort had significantly higher utilization of BC–related services and health care costs PPPM over follow-up. A significantly higher proportion of the CIPN cohort had 1 or more BC–related ED visits (19.3% vs 13.2%, P < .001); office visits (98.6% vs 97.3%, P < .01); or outpatient pharmacy refills for either a BC (56.0% vs 51.8%, P < .01) or neuropathy medication (93.3% vs 86.3%, P < .001) compared to the non-CIPN cohort. Patients with CIPN also had a significantly higher number of monthly office visits and outpatient pharmacy refills for BC or neuropathy medications (P < .01) (Table 4).

Higher utilization of BC–related outpatient medical and pharmacy services by the CIPN cohort led to an incremental cost for BC–related services of $725 PPPM, compared to the non-CIPN cohort. Increases in office and other outpatient services costs were the primary drivers of elevated costs (Table 4).

A total of 274 patients with CIPN and 314 patients without CIPN qualified for the short-term disability analysis. A greater proportion of patients with CIPN had a short-term disability claim during follow-up (58.0% vs. 42.0%, P < .001). No significant differences in lost workdays (CIPN, 5.1 ± 16.2; non-CIPN, 3.9 ± 11.0, P = .28); or indirect costs (CIPN: $702 ± $2185; non-CIPN: $539 ± $1657, P = .31) were observed between the cohorts. A total of 255 patients with CIPN and 288 patients without CIPN were included in the long-term disability analysis. No differences were observed between cohorts in the proportion of patients with a long-term disability claim (CIPN, 11.0%; non-CIPN, 12.0%, P = .567); the number of lost work days (CIPN, 1.3 ± 12.0; non-CIPN, 1.0 ± 3.8, P = .69); or indirect costs (CIPN, $169 ± $1586; non-CIPN, $122 ± $461, P = .63).

This claims-based study examined the burden of CIPN in patients with mBC, initiating IV paclitaxel/nab-paclitaxel. Among the 5870 women with mBC 42.7% developed CIPN over an average of 24.7 (± 13.7) months. The rate of CIPN identified in our study is higher than that of other claims-based studies, which report rates of peripheral neuropathy of less than 20%; there are several reasons for this difference

 First, our analysis allowed patients to have prior, non-taxane chemotherapy, which may be associated with an increased risk of CIPN compared to studies limiting analyses to the first or second LOT.

 Second, our follow-up extended beyond the line of IV paclitaxel/nab-paclitaxel treatment, leading to a longer follow-up than studies that assessed toxicities during a LOT. Thus, the current findings provide new insight on the longer-term impact of CIPN in mBC.

Fewer than half the patients with CIPN in our analysis were diagnosed with CIPN during the Index LOT. Therefore, the long-term burden of CIPN would have been underestimated if follow-up was limited to the Index LOT. Furthermore, other studies have reported that peripheral neuropathy is undercoded when assessed via ICD-9 codes.

 We used a broader CIPN definition, which included ICD-9/ICD-10 codes for peripheral neuropathy, paresthesia, myalgia, and arthralgia. These factors all likely resulted in higher overall rates of CIPN than previously reported in other claims-based analyses. However, the rate of CIPN identified in the current study is largely in line with, if not lower than, rates of CIPN reported in other retrospective, observational, or clinical trial analyses, which utilized electronic health records, patient surveys, or clinical scales to assess CIPN.

Similar to other studies, the current analysis showed that CIPN in mBC is associated with increased health care service utilization and costs. All-cause costs for patients with CIPN were higher than those for patients without CIPN by $1102 PPPM, or approximately $13,224 per patient annually. The BC–related costs for patients with CIPN were higher than those for patients without CIPN by $725 PPPM, or approximately $8700 per patient annually. Hansen et al.

 used claims to examine costs in patients with mBC treated with taxanes, based on toxicity development. The monthly costs for patients with toxicities were $1142 higher during the first LOT, and $1448 during the second LOT than similar costs for patients without toxicities.

 found incremental costs for patients with mBC and toxicities that ranged from $845 to $5320 per month depending on the treatment and LOT. Similarly, Pike et al.

 reported incremental annual costs of $16,940, or approximately $1412 per month, in patients with BC who developed CIPN, compared to those who did not. The slightly lower incremental costs observed in our analysis likely resulted from two main methodological differences. The current study defined cohorts based on the presence of CIPN alone. Therefore, our non-CIPN group likely had a higher rate of non-CIPN toxicities compared to the cohorts without evidence of any toxicity in other studies.

 This difference would likely lead to a lower incremental cost. Secondly, our analyses followed patients beyond their first LOT, resulting in a longer follow-up than most other studies.

The work disability analysis revealed limited differences between cohorts. The population comprised patients with mBC, who may be less likely to return to work owing to advanced disease, a point which cannot be overlooked when assessing disability. Similarly, Pike et al.

used employer disability data and found a nonsignificant trend for increased disability claims and disability days among patients with CIPN. Zanville et al.

 surveyed BC survivors regarding their ability to work; a greater proportion of patients reported being able to work at 1 year vs 1 month, leading the authors to conclude that patients developed coping mechanisms that facilitated a return to work with time.

An association between CIPN and QOL impairments has also been reported.

 used a patient survey and found that CIPN symptoms were significantly associated with worse functional performance, gait abnormalities, and an increased risk of falls. In the current analysis, no differences in the prevalence of fractures or falls were found between the CIPN and non-CIPN cohorts during the preperiod. Conversely, during follow-up, the CIPN cohort was significantly more likely to experience a fall or fracture compared with the non-CIPN cohort.

 also reported an increased rate of falls with more severe CIPN. Two studies reported an increased use of pain medication among patients with neuropathy following chemotherapy.

 Similarly, the current analysis showed that patients with CIPN also had increased use of neuropathy medication, many of which were also pain medications. Furthermore, Bao et al.

 reported increased rates of patient-reported anxiety, depression, and insomnia with increased CIPN severity. The current analysis mirrored this finding with an increased prevalence and incidence of depression and insomnia in patients with CIPN vs patients without CIPN during follow-up.

The study included patients with commercial or Medicare supplemental insurance. Thus, the results may not be generalizable to patients with other types of insurance or the uninsured. In addition, patients had 3 or more months of postindex enrollment, and patients with CIPN had 3 or more months of enrollment following a diagnosis of neuropathy. Thus, the results may not be applicable to patients who die shortly after the initiation of IV paclitaxel/nab-paclitaxel or a CIPN diagnosis.

Misclassification of clinical status, diagnoses, or other outcomes is possible, as administrative claims are collected for billing, and are subject to miscoding and undercoding. This may be particularly relevant when assessing CIPN as a previous study reported the undercoding of neuropathy.8 Furthermore, the lack of clinical data (eg, cancer stage, biomarkers, pain scales, etc.) makes it difficult to assess the severity of cancer or CIPN. Relative dose changes were estimated using paid amounts on claims. Changes in rebates, variability in generic agent prices, or site of administration could all affect these estimates. Conclusions in the work disability analyses were limited because of the low sample sizes. Not all patients could have been matched, as 22% of the CIPN cohort and 42% of the non-CIPN cohort were not matched 1:1.

CIPN is a common treatment-related toxicity in cancer. In this claims-based analysis, the long-term rate of CIPN in mBC was estimated at 42.7%. Development of CIPN following the administration of IV paclitaxel/nab-paclitaxel was associated with significantly higher utilization of all-cause and BC–related health care services and costs, an increased comorbidity burden, and more treatment alterations. These findings demonstrate the substantial long-term impact of chemotherapy and provide an estimate of both the patient burden and economic impact of CIPN. Considering the increasing survival of cancer populations with the availability of new treatments, the longer-term impact of toxicities, like CIPN, need to be considered throughout treatment. The development of alternative therapies with lower toxicity could provide substantial long-term benefits, not only for patients but also for the health care system. 

 IBM Watson Health, Cambridge, MA (BLB, ML, KW, HV, CLC); Mirador Global, Kennett Square, PA (KMF, CC); Athenex Inc., Buffalo, NY (JW, ID).

 Concept and design (BLB, KW, KMF, JW, CC, HV, ID); acquisition of data (BLB, KW, KMF, HV); analysis and interpretation of data (ML, KW, KMF, JW, CC, HV, CLC, ID); drafting of manuscript (BLB, ML); critical revision of manuscript for important intellectual content (KW, KMF, JW, CC, ID); statistical analysis (CLC); obtaining funding (JW, ID); administrative, technical, and logistical support (ML, KW, HV, CLC); and supervision (KMF, JW, CC, ID)

 This study was sponsored by Athenex, Inc.

 BLB, KW, and HV are employed by IBM Watson Health, which received funding from Athenex, Inc. to perform this study. ML and CLC were employed by IBM Watson Health at the time of this study. KMF and CC are employed by Mirador Global, which received research funding from Athenex, Inc. to conduct this study. JW and ID are employed by and own stock in Athenex. This research was presented in part at the 2020 Virtual Academy of Managed Care Pharmacy (AMCP) Nexus conference.

The data that support the findings of this study are available from IBM Watson Health. Restrictions apply to the availability of these data, which were used under license for this study.

1. Cardoso F, Harbeck N, Fallowfield L, Kyriakides S, Senkus E. Locally recurrent or metastatic breast cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. 

 2012;23(suppl7):vii11-19. doi: 10.1093/annonc/mds232

2. Ghersi D, Willson ML, Chan MM, Simes J, Donoghue E, Wilcken N. Taxane-containing regimens for metastatic breast cancer. Cochran Databse Syst Rev. 2015;2015(6):CD003366. doi: 10.1002/14651858.CD003366.pub3

3. Rashid N, Koh HA, Baca HC, et al. Clinical impact of chemotherapy-related adverse events in patients with metastatic breast cancer in an integrated health care system

. 2015;21(10):863-871. doi: 10.18553/jmcp.2015.21.10.863

4. Rivera E, Cianfrocca M. Overview of neuropathy associated with taxanes for the treatment of metastatic breast cancer. 

2015;75(4):659-670. doi: 10.1007/s00280-014-2607-5

5. Reyes-Gibby CC, Morrow PK, Buzdar A, Shete S. Chemotherapy-induced peripheral neuropathy as a predictor of neuropathic pain in breast cancer patients previously treated with paclitaxel. 

 2009;10(11):1146-1150. doi: 10.1016/j.jpain.2009.04.006

6. Seretny M, Currie GL, Sena ES, et al. Incidence, prevalence, and predictors of chemotherapy-induced peripheral neuropathy: a systematic review and meta-analysis. 

 2014;155(12):2461-2470. doi: 10.1016/j.pain.2014.09.020

7. Winters-Stone KM, Horak F, Jacobs PG, et al. Falls, functioning, and disability among women with persistent symptoms of chemotherapy-induced peripheral neuropathy

 2017;35(23):2604-2612. doi: 10.1200/JCO.2016.71.3552

8. Shah A, Hoffman EM, Mauermann ML, et al. Incidence and disease burden of chemotherapy-induced peripheral neuropathy in a population- based cohort. 

2018;89(6):636-641. doi: 10.1136/jnnp-2017-317215

9. Hurvitz S, Guerin A, Brammer M, et al. Investigation of adverse-event-related costs for patients with metastatic breast cancer in a real-world setting. 

. 2014;19(9):901-908. doi: 10.1634/theoncologist.2014-0059

10. Hansen RN, Ramsey SD, Lalla D, et al. Identification and cost of adverse events in metastatic breast cancer in taxane and capecitabine based regimens. 

. 2014;3:259. doi: 10.1186/2193-1801-3-259

11. Pike CT, Birnbaum HG, Muehlenbein CE, Pohl GM, Natale RB. Healthcare costs and workloss burden of patients with chemotherapy-associated peripheral neuropathy in breast, ovarian, head and neck, and non-small cell lung cancer. 

 2012;2012:913848.doi: 10.1155/2012/913848

12. Rashid N, Koh HA, Baca HC, Lin KJ, Malecha SE, Masaquel A. Economic burden related to chemotherapy-related adverse events in patients with metastatic breast cancer in an integrated health care system. 

(Dove Med Press). 2016;8:173-181. doi:10.2147/BCTT.S105618

13. Zanville NR, Nudelman KN, Smith DJ, et al. Evaluating the impact of chemotherapy-induced peripheral neuropathy symptoms (CIPN-sx) on perceived ability to work in breast cancer survivors during the first year post-treatment. Support Care Cancer. 2016;24(11):4779-4789. doi:10.1007/s00520-016-3329-5

14. SEER-Medicare: Calcuation of comorbidity weights. Accessed December 11, 2020. https://healthcaredelivery.cancer.gov/seermedicare/considerations/calculation.html

15. Cho H, Mariotto AB, Mann BS, Klabunde CN, Feuer EJ. Assessing non-cancer-related health status of US cancer patients: other-cause survival and comorbidity prevalence. 

 2013;178(3):339- 349. doi: 10.1093/aje/kws580

16. Consumer Price Index details report tables Accessed March 1, 2020. https://www.bls.gov/cpi/home.htm

17. Bao T, Basal C, Seluzicki C, Li SQ, Seidman AD, Mao JJ. Long-term chemotherapy-induced peripheral neuropathy among breast cancer survivors: prevalence, risk factors, and fall risk. 

. 2016;159(2):327-333. doi: 10.1007/s10549-016-3939-0

Peer-Reviewed

We examine the incidence and impact of chemotherapy induced peripheral neuropathy on clinical and economic outcomes in women with metastatic breast cancer initiating intravenous paclitaxel/nab-paclitaxel.

Chemotherapy-Induced Peripheral Neuropathy in Metastatic Breast Cancer Patients Initiating Intravenous Paclitaxel/nab-Paclitaxel

Supplemental materials for article examining the incidence and impact of chemotherapy induced peripheral neuropathy on clinical and economic outcomes in women with metastatic breast cancer initiating intravenous paclitaxel/nab-paclitaxel.

Supplemental Material: Chemotherapy-Induced Peripheral Neuropathy in Metastatic Breast Cancer Patients Initiating Intravenous Paclitaxel/nab-Paclitaxel

Intravenous (IV) taxane therapy for metastatic breast cancer (mBC) has been associated with toxicities and demanding dosing schedules, which can limit treatment
effectiveness.

 To assess treatment patterns, toxicities, and costs in women with mBC initiating IV paclitaxel or IV nab-paclitaxel.

Adult women diagnosed with BC from January 1, 2014, to September 30, 2018, were identified in the MarketScan Commercial and MarketScan Medicare Supplemental databases. Women had a metastatic disease diagnosis and newly initiated treatment with IV paclitaxel/nab-paclitaxel (first administration date was considered the index date), and continuous enrollment for at least 12 months prior to and at least 3 months following the index date. Treatment discontinuation, dose reductions, toxicities, and health care utilization and costs per patient per month (PPPM) were assessed over the full follow-up and the index line of IV paclitaxel/nab-paclitaxel therapy (Index LOT).

 The sample included 8890 women aged 54.6 (±10.9) years, followed for 18.9 (±13.5) months. Most (82.0%) initiated IV paclitaxel/nab-paclitaxel monotherapy; 83.1% had early discontinuation (<18 weeks of treatment) of the Index LOT. Among the 6943 women eligible for the dose-change analysis, 42.4% evidenced an IV paclitaxel/nab-paclitaxel dose reduction ≥10% during the Index LOT. The most common toxicities during the Index LOT were gastrointestinal upset (30.5%), myelotoxicity (27.0%), infection (26.2%), general symptoms (25.9%), and chemotherapy-induced peripheral neuropathy (22.7%). Over follow-up, 39.7% of women had an inpatient admission and 43.0% had an emergency department visit. The mean of all-cause total costs was $11,991 PPPM, while BC-related total costs were $5320 PPPM.

Many mBC patients initiating IV paclitaxel/nab-paclitaxel experienced dose reductions, toxicities, and/or early discontinuation of the Index LOT, which may limit treatment effectiveness. More tolerable treatments with reduced dosing complexity could improve mBC treatment and help contain costs.

metastatic breast cancer, chemotherapy treatment patterns, health care service utilization and costs.

Early discontinuation, dose reductions, and toxicities are common among women with mBC initiating IV paclitaxel or IV nab-paclitaxel, and they may impact the effectiveness of treatment and longterm patient outcomes. Further, the economic burden among patients treated with IV paclitaxel or nab-paclitaxel is substantial at $11,991 per patient per month, or approximately $143,892 annually. Treatments with improved tolerability stand to improve the clinical burden for patients and economic burden on the health care system.

83.1% of women with mBC discontinued IV paclitaxel/nab-paclitaxel early, and 42.1% had a dose reduction.

78.6% of patients experienced a toxicity requiring medical attention while on IV paclitaxel/nab-paclitaxel.

In 2020, an estimated 279,100 women will be newly diagnosed with breast cancer (BC) in the United States, representing 30.5% of all new cancer diagnoses in women.

 Among this group, about 6% have distant metastases at diagnosis (mBC)

; the remainder of women with mBC experience progression from earlier-stage BC.

 While the overall mortality rate for BC has decreased by 40% since 1989,3 it remains the leading cause of death for women aged 45 to 64 years, as estimated 5-year relative survival for mBC is only 27%.

Despite improvements in detection and treatment, mBC remains incurable; therapy aims to ameliorate symptoms, delay disease progression, and improve quality of life and survival.6 Intravenous (IV) taxanes (paclitaxel, nab-paclitaxel, docetaxel) have been foundational BC treatments for the last 3 decades.

Despite the efficacy of taxanes in extending survival, many patients experience dose-limiting adverse effects (AEs) including infusion reactions, chemotherapy-induced peripheral neuropathy (CIPN), and myeloid cytopenia.

Further, management of BC is associated with high health care costs.

 Costs vary by stage, with mBC patients incurring the highest costs of care, estimated at $134,682 per patient in the year following diagnosis; cumulative costs are estimated at $270,033 to $588,491 for patients who survive ≥3 years after mBC diagnosis.

; however, correlations have been reported between treatment-related toxicities and higher health care costs.

Although some data exist on the economic burden of mBC, limited data are available on the specific costs of care with IV paclitaxel and IV nab-paclitaxel treatment.

This observational, retrospective study used administrative claims data to examine treatment patterns, toxicities, health care resource utilization (HCRU), and health care costs among women with mBC initiating treatment with IV paclitaxel or IV nab-paclitaxel.


This observational retrospective analysis utilized the deidentified IBM MarketScan Commercial Claims and Encounters (commercial) and MarketScan Medicare Supplemental (Medicare) administrative claims databases for the period November 1, 2012, to December 31, 2018. The commercial database contains the inpatient, outpatient, and outpatient prescription drug experience of employees and their dependents, covered under a variety of health plans.

The Medicare database contains the full health care experience (both medical and pharmacy) of retirees with Medicare supplemental insurance provided by employers; Medicare and employer covered portions of payments are included. All study data were obtained using National Drug Codes and codes from International Classification of Diseases, Clinical Modification, 9th and 10th revisions (

); from Current Procedural Terminology, 4th edition; and from Healthcare Common Procedure Coding System.


Women with ≥ 1 inpatient or ≥ 2 nondiagnostic outpatient claims with a diagnosis of BC (

: C50.xx) from January 1, 2014, through September 30, 2018, were identified within the MarketScan databases. Women were also required to have ≥ 1 nondiagnostic claim with a diagnosis of metastasis occurring 60 days prior to the first BC claim through the end of data (excludes codes for metastasis to the breast) and ≥ 1 claim for IV paclitaxel or IV nab-paclitaxel (hereafter referred to as IV paclitaxel/nab-paclitaxel) in the outpatient medical setting following the diagnosis of metastasis. The index date was the date of first administration of IV paclitaxel/nab-paclitaxel.

Eligible women had to be ≥ 18 years on the index date and have continuous enrollment with medical and pharmacy benefits for ≥12 months prior (preperiod) to and ≥ 3 months following the index date. Patients who received any IV taxane in the 12-month preperiod or >1 taxane on the index date were excluded (to isolate impact of paclitaxel). The variable follow-up period extended from index until the earliest of: (1) docetaxel administration; (2) end of continuous enrollment; or (3) end of the study period (December 31, 2018).


The first line of IV paclitaxel/nab-paclitaxel therapy, starting on the index date, was defined. Index line of therapy (LOT) components included IV paclitaxel/nab-paclitaxel and all chemotherapy agents (capecitabine, carboplatin, cisplatin, cyclophosphamide, doxorubicin, epirubicin, eribulin, flourouracil, gemcitabine, ixabepilone, methotrexate, and vinorelbine) administered in the 37 days following the index date. As patients may have been receiving nontaxane chemotherapy prior to the index date, the 37-day period allowed identification of concomitant medications that were administered on a different cycle. The end of the Index LOT was defined by a regimen change or the end of follow-up; medication discontinuation was defined as a gap in treatment of ≥ 45 days. For patients with a subsequent LOT during follow-up, the components of the second LOT were identified.

Changes in IV paclitaxel/nab-paclitaxel dose were assessed during the Index LOT. Dose was estimated by dividing the paid amount on IV paclitaxel/nab-paclitaxel claims by the wholesale acquisition cost; capitated claims, claims with a JW modifier (indicating medication waste), or a $0 paid amount were excluded. To facilitate assessment of changes over time, only patients with ≥ 3 non-$0 claims for IV paclitaxel/nab-paclitaxel were included in the dose analysis. The magnitude of a dose change was defined as ≥ 10% of the index dose, and dose changes were examined in relation to the prior dose.


Demographics were measured on index. The National Cancer Institute (NCI) Comorbidity Index and treatment characteristics were examined during the preperiod.

 The prevalence and incidence of comorbidities were assessed during the Index LOT and full follow-up. Comorbidities included depression, diabetes, insomnia, falls, fractures, liver dysfunction, rheumatoid arthritis, and nonrheumatoid arthritis (osteoarthritis, psoriatic arthritis, and other polyarthropathies). Patients without a diagnosis of the comorbidity of interest in the preperiod were eligible for the incidence analysis.


Prevalent and incident toxicities were assessed over the Index LOT and the full follow-up; the incident toxicity analysis included only patients without a diagnosis for the toxicity of interest during the preperiod. During the Index LOT, dose changes in the 14 days following a toxicity, and neuropathy medication use in the +/– 10 days around a dose change, were both reported as proxies for toxicity-associated events. Over the full follow-up, hospitalizations or emergency department (ED) visits in the 14 days following an incident toxicity (temporal correlation) were assessed as a proxy for toxicity severity.

Toxicities included alopecia, CIPN, cardiovascular events (angina, atrial fibrillation, atrial flutter, atrioventricular block, cardiac arrest, cardiac tamponade, disseminated intravascular coagulation, heart failure, myocardial infarction, stroke, thromboembolism, and thrombosis), cardiovascular symptoms (bradycardia, chest pain, and tachycardia), general AEs (fever, headache, malaise/fatigue, edema, fluid retention, epistaxis, and dysgeusia/dysosmia), gastrointestinal (GI) events (ascities, colitis, esophagitis, GI bleeding, GI obstruction, GI perforation, pancreatitis, and peptic ulcer) , GI symptoms (constipation, dysphagia, and stomatitis), GI upset (abdominal pain, diarrhea, dyspepsia, and nausea/vomiting), hematological effects, infection, infusion-related/hypersensitivity effects, joint/muscle effects, musculoskeletal effects, myelotoxicity, nervous system effects (ataxia, dizziness, vertigo, syncope, and seizures), encephalopathy, neutropenia, paresthesias, pulmonary events, renal toxicity, and skin effects (full list in 


All-cause and BC-related HCRU and costs per patient per month (PPPM), including inpatient admissions, outpatient medical encounters (ED visits, outpatient office visits, and other outpatient services), and outpatient prescriptions were reported over follow-up. BC-related HCRU included inpatient admissions with a diagnosis of BC in the primary position; outpatient services with a diagnosis of BC in any position; medical claims for BC treatments (chemotherapy, endocrine therapy, HER2-directed therapy, or radiation); pharmacy claims for BC; and pharmacy claims for medications prescribed to relieve neuropathy (antidepressants, anticonvulsants, opioids and narcotics, systemic corticosteroids, nonopioid and nonsteroidal pain relief drugs, topical local anesthetics, mexiletine, tocainide, and pentoxifylline). Costs represented the paid amounts of fully adjudicated claims; costs for capitated plans were estimated via payment proxy. All costs were inflated to 2019 US dollars using the medical care component of the Consumer Price Index.


Mean and SD were reported for continuous variables. Frequencies and percentages were reported for categorical variables. All data analyses were conducted using WPS version 4.1 (World Programming, United Kingdom).

A total of 8890 women with mBC initiating treatment with IV paclitaxel/nab-paclitaxel qualified for the study (

Mean age was 54.6 (±10.9) years at index; mean NCI comorbidity score was 0.6 (±1.0), and patients were followed for a mean of 18.9 (±13.5) months (


The most common preperiod chemotherapies were doxorubicin (59.7%) and cyclophosphamide (58.9%; Table 1). Endocrine therapy was used by 18.6% of patients; 2.5% of patients utilized HER2-directed therapy. More than 80% of women received ≥ 1 systemic BC therapy during the preperiod.

Most patients (86.8%) initiated IV paclitaxel on the index date; the remaining 13.2% initiated IV nab-paclitaxel (Table 2). Overall, 82.0% of patients received IV paclitaxel/nab-paclitaxel monotherapy. Among the 1596 patients who received polytherapy (combination with IV paclitaxel/nab-paclitaxel), carboplatin (62.8%), cyclophosphamide and doxorubicin (13.5%), and gemcitabine (7.5%) were the most common other therapy components.


The duration of the Index LOT was 3.2 (±2.1) months; 77.2% of patients ended their Index LOT due to discontinuation of ≥ 1 regimen medication. Considering that 82.0% of the patients received monotherapy for the Index LOT, IV paclitaxel/nab-paclitaxel was frequently discontinued. Regimen augmentation/switch was observed in 14.1% of women; the remaining 8.7% ended their Index LOT due to the end of follow-up (

). Overall, 83.1% of women had evidence of early discontinuation of the Index LOT, defined as <18 weeks (6 cycles) of therapy.

Among the 6943 patients eligible for the dose-change analysis, 3497 (50.4%) evidenced ≥ 1 IV paclitaxel/nab-paclitaxel dose change ≥ 10% over the Index LOT (Table 2). Patients had a mean of 2.0 (±1.8) dose changes; time from index to the first dose change was a mean of 39 (±38) days. A total of 2945 patients (42.4%) had at least 1 dose reduction ≥10%, with a mean of 1.5 (±1.0) reductions per patient. Time to the first dose reduction was a mean of 46 (±42) days, and the mean reduced dose was 70% (±20%) of the prior dose. Of the 2945 patients with a dose reduction during the Index LOT, 1626 (55.2%) evidenced a dose reduction in the 14 days following a toxicity; the most common toxicities prior to dose reduction were GI upset (18.6%), myelotoxicity (14.7%), CIPN (12.6%), and neutropenia (12.1%; Table 2).


Over the full follow-up period, the most common incident comorbidities were depression (12.3%), liver dysfunction (10.5%), nonrheumatoid arthritis (8.9%), and insomnia (7.4%); incident diabetes, falls, fractures, and rheumatoid arthritis each were observed in less than 5% of the eligible women. Toxicities were common; 95.4% of women had ≥1 prevalent toxicity, with a mean of 5.2 (±3.2) toxicities over the full follow-up (Supplemental Table 1). The most common toxicities were infection (62.3%), general symptoms (53.8%), CIPN (51.6%), GI upset (50.0%), and myelotoxicity (43.3%). During the Index LOT, 78.6% of patients had claims for ≥ 1 prevalent toxicity, with a mean of 2.3 (±2.1) toxicities. The most common toxicities during the Index LOT included GI upset (30.5%), myelotoxicity (27.0%), infection (26.2%), general symptoms (25.9%), and CIPN (22.7%).

Incident toxicities were assessed among the subset of women with no evidence of the toxicity of interest in the preperiod (

). Over the full follow-up, the incidence rate of toxicities ranged from 0.2 cases/100 person-years for alopecia to 56.5 cases/100 person-years for infections. Other common incident toxicities included CIPN, general symptoms, and GI upset with 49.7, 45.5, and 33.4 cases/100 person-years, respectively. Using the presence of hospitalizations or ED visits in the 14 days following the toxicity as a proxy for severity, the most severe toxicities were neurological, renal, and pulmonary events, with 91.4%, 68.2%, and 59.9% of patients, respectively, with a hospitalization/ED visit in the 14 days following the incident toxicity. During the Index LOT, the incidence of toxicities ranged from 0.4 cases/100 person-years for alopecia to 84.7 cases/100 person-years for general symptoms. Common toxicities included CIPN, infection, and GI upset, with 79.4, 79.6, and 77.1 cases/100 person-years, respectively. Neurological, renal, and pulmonary toxicities remained the most severe with 90.2%, 65.7%, and 59.0% of patients, respectively, with a hospitalization/ED visit in the 14 days following the toxicity.


During the full follow-up, 39.7% of patients had an inpatient admission and 43.0% had an ED visit (Table 4). Nearly all patients (99.7%) had ≥ 1 outpatient office visit with a mean of 2.0 (±1.1) visits PPPM. All patients had ≥ 1 other outpatient service, which included claims for chemotherapy infusion, with a mean of 6.4 (±3.8) services PPPM. In addition, 98.6% of patients used the outpatient pharmacy, with a mean of 2.9 (±2.2) prescriptions PPPM.

Mean total all-cause costs among patients with mBC treated with IV paclitaxel/nab-paclitaxel were $11,991 (±$11,681) PPPM during the full follow-up period (

). Inpatient costs were $1775 (±4875) PPPM, and all-cause medical costs were $11,274 (±$11,443) PPPM. Other outpatient services were the primary driver of all-cause total costs at $9067(±$9304) PPPM.


BC-related HCRU over the full follow-up period was substantial with 7.9%, 18.1%, and 97.2% of patients with ≥ 1 inpatient admission, ED visit, and office visit, respectively (Table 4). While 100% of patients received physician-administered BC-related medications with a mean of 1.0 (±0.8) administrations per month, 98.5% had additional BC-related other outpatient services with a mean of 4.3 (±3.1) services per month. Approximately half of women (49.9%) filled an outpatient prescription for a BC-related medication, and 89.3% of patients filled an outpatient prescription for a medication to treat neuropathy.

Total BC-related costs among patients treated with IV paclitaxel/nab-paclitaxel were $8173 (± $9140) PPPM during the full follow-up (Table 4). BC-related medical costs were $8031 (±$9085) PPPM, with $2445 (±$5768) PPPM from physician-administered BC medications, and another $5077 (±$5536) PPPM representing costs of other outpatient services.


In this study of 8890 women with mBC initiating IV paclitaxel or IV nab-paclitaxel, early discontinuation, dose reductions, and toxicities were common over the Index LOT. Early discontinuation occurred in 83.1% of women, and 42.4% of women in the dose-change analysis had ≥ 1 IV paclitaxel/nab-paclitaxel dose reduction ≥ 10%. Toxicities sufficiently serious for the patients to seek medical care (in order for them to be reported in claims data) were observed in 78.6% of women during the Index LOT. Assessment of HCRU and costs over the follow-up revealed that inpatient admissions and hospitalizations were common, and costs were substantial at $11,991 (±$11,681) PPPM, 68.2% of which were BC-related.

Over the Index LOT, 78.6% had evidence of a prevalent toxicity; this rose to 95.4% over the full follow-up. The rate of toxicities observed in this study are in line with previously published reports citing toxicity rates of 68% to 95% in taxane users.

We also identified high rates of incident toxicity both over the Index LOT and full follow-up; the most common were CIPN, general symptoms, and infections, each occurring in approximately 50% of women over the full follow-up. Although a causal relationship between toxicities and treatment changes cannot be established in claims data, the finding that 55.2% of women with a dose reduction during the Index LOT had a dose change in the 14 days following a toxicity diagnosis suggests that toxicities impacted treatment. In a chart review to examine the relationship between toxicities and dose changes, Rashid et al found that 20% of patients with a toxicity had a change to their chemotherapy regimen, which is similar to the 23.4% of women with dose reductions in this study

Toxicities do not only impact chemotherapy but can also drive increased HCRU. Multiple studies have reported increased health care costs in patients who experience toxicities.

 This study did not specifically address relative health care costs in patients with and without toxicities; however, the presence of ED visits or inpatient admissions was assessed in the 14 days following incident toxicities. Of the 21 toxicities assessed here, ED visits or inpatient admission rates of ≥ 30% were observed for 13 toxicities. Rates of ED visits or inpatient admissions surpassed 50% for incident cardiovascular disease events and symptoms; pulmonary events; and hematological, renal, and encephalopathy.

During the full follow-up, mBC patients treated with IV paclitaxel/nab-paclitaxel in this study incurred mean health care costs of $11,991 PPPM, or approximately $143,892 per patient annually. BC-related costs accounted for more than two-thirds (68.2%) of the monthly all-cause costs at $8173 PPPM or $98,076 per patient per year. For both all-cause and BC-related costs, other outpatient services and physician administration of BC-related medications were the primary cost drivers. Similarly, Hurvitz et al and Hansen et al examined total health care costs for mBC patients during paclitaxel treatment; they found costs ranging from approximately $7500 to $13,000 PPPM depending on LOT and toxicity burden.

 Other studies have also assessed the burden of mBC in broader populations (eg, not only taxane treatment) and found costs ranging from approximately $9000 to $11,000 PPPM, which are largely in line with this analysis.

In aggregate, the treatment changes, discontinuation, dose reductions, toxicities, incident comorbid conditions, and high HCRU and costs incurred by patients with mBC receiving IV paclitaxel/nab-paclitaxeldemonstrate a large burden for patients and the health care system.


This descriptive study is subject to several limitations. This analysis was limited to patients with commercial or Medicare supplemental insurance and ≥3 months of postindex enrollment; therefore, results may not generalize to the uninsured, those with other insurance, or patients who die shortly after initiating treatment. Second, there is the potential for clinical misclassification, as patients were identified through administrative claims data, which are subject to miscoding and undercoding. This may be particularly relevant to assessment of toxicities, as less severe conditions that may not require medical intervention (eg, headache) or conditions less likely to be coded (eg, nausea) are likely underreported. It is also extremely difficult to discern cancer severity due to the lack of staging and biomarker information in claims. Third, all dose change analyses were estimated using paid amounts on claims. Due to this methodology, only relative changes in dose were assessed; however, changes in factors such as rebates, variability in price of generic agents, or site of medication administration could impact estimates. Finally, this study cannot identify causal relationships between IV paclitaxel/nab-paclitaxel and toxicity development.

Alterations in chemotherapy, including early discontinuation and dose reductions, were common among women with mBC initiating IV paclitaxel/nab-paclitaxel in this study. High rates of toxicities were also observed, and proxy measures examining the timing between toxicities and dose reductions found that 55.2% of women with a dose reduction experienced a dose reduction shortly after a toxicity diagnosis. Similarly, high rates of hospitalizations or ED visits following severe incident toxicities (eg, encephalopathy) were observed. Treatments with increased tolerability have the potential to reduce toxicity rates, improve adherence to treatment, and improve patient outcomes. Additionally, the majority of health care costs ($11,991 PPPM or approximately $143,892 annually per patient) in this study were derived from other outpatient services including chemotherapy administration. These findings suggest that chemotherapy and toxicities exert a substantial burden on patients and payers; medications with improved dosing schedules and tolerability could therefore have the potential to reduce the clinical and economic burden of mBC. 

IBM Watson Health, Cambridge, MA (BLB, ML, KW, HV, CLC); Mirador Global, Kennett Square, PA (KMF, CC); Athenex Inc, Buffalo, NY (JW, ID).

 Concept and design (BLB, KW, KMF, JW, CC, HV, ID); acquisition of data (KW, KMF, HV); analysis and interpretation of data (BLB, ML, KW, KMF, JW, CC, HV, CLC, ID): drafting of manuscript (BLB, ML); critical revision of manuscript for important intellectual content (KW, KMF, JW, CC, ID); statistical analysis (CLC); obtaining funding (JW, ID); administrative, technical, and logistical support (ML, KW, HV, CLC); and supervision (KMF, JW, CC, ID).

 This study was sponsored by Athenex Inc.

Brenna L. Brady, PhD; brenna.brady@ibm.com

 BLB, KW, and HV are employed by IBM Watson Health, which received funding from Athenex Inc to perform this study. ML and CLC were employed by IBM Watson Health at the time f this study. KMF and CC are employed by Mirador Global, which received research funding from Athenex Inc to conduct this study. JW and ID are employed by and own stock in Athenex.

This research was presented in part at the 2020 Virtual AMCP Nexus conference.


Medical writing services were provided by Jessamine Winer-Jones of IBM Watson Health.


The data that support the findings of this study are available from IBM Watson Health. Restrictions apply to the availability of these data, which were used under license for this study.

1. Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. 

. 2020;70(1):7-30. doi:10.3322/caac.21590

2. Cancer stat facts: female breast cancer. National Cancer Institute / Surveillance, Epidemiology, and End Results Program. Accessed June 2020. https://seer.cancer.gov/statfacts/html/breast.html 

3. DeSantis CE, Ma J, Gaudet MM, et al. Breast cancer statistics, 2019. 

 2019;69(6):438-451. doi:10.3322/caac.215834

4. Mariotto AB, Etzioni R, Hurlbert M, Penberthy L, Mayer M. Estimation of the number of women living with metastatic breast cancer in the United States. 

 2017;26(6):809-815. doi:10.1158/1055-9965.EPI-16-0889

5. Curtin SC. Trends in cancer and heart disease death rates among adults aged 45-64: United States, 1999-2017. Natl Vital Stat Rep. 2019;68(5):1-9.

6. Waks AG, Winer EP. Breast cancer treatment: a review. 

 2019;321(3):288-300. doi:10.1001/jama.2018.19323

7. Gradishar WJ. Taxanes for the treatment of metastatic breast cancer. Breast Cancer (Auckl). 2012;6:159-171. doi:10.4137/BCBCR.S8205

8. NCCN Clinical Practice Guidelines in Oncology. National Comprehensive Cancer Network (NCCN). Breast cancer, version 4.2020.  https://www.nccn.org/professionals/physician_gls/

9. Rowinsky EK, Chaudhry V, Cornblath DR, Donehower RC. Neurotoxicity of taxol. 

10. Blumen H, Fitch K, Polkus V. Comparison of treatment costs for breast cancer, by tumor stage and type of service. 

11. Fu AZ, Jhaveri M. Healthcare cost attributable to recently-diagnosed breast cancer in a privately-insured population in the United States. 

. 2012;15(4):688-694. doi:10.3111/13696998.2012.673524

12. Ray S, Bonthapally V, McMorrow D, Bonafede M, Landsman-Blumberg P. Patterns of treatment, healthcare utilization and costs by lines of therapy in metastatic breast cancer in a large insured US population. 

13. Kruse GB, Amonkar MM, Smith G, Skonieczny DC, Stavrakas S. Analysis of costs associated with administration of intravenous single-drug therapies in metastatic breast cancer in a U.S. population. 

 2008;14(9):844-857. doi:10.18553/jmcp.2008.14.9.844

14. Hansen RN, Ramsey SD, Lalla D, et al. Identification and cost of adverse events in metastatic breast cancer in taxane and capecitabine based regimens. 

. 2014;3:259. doi:10.1186/2193-1801-3-259

15. Hurvitz S, Guerin A, Brammer M, et al. Investigation of adverse-event–related costs for patients with metastatic breast cancer in a real-world setting.

This article describes treatment patterns, toxicities, and health care service utilization and costs of women with metastatic breast cancer (mBC) initiating treatment with intravenous (IV) paclitaxel or IV nab-paclitaxel.

January 2021

Chemotherapy-Induced Peripheral Neuropathy in Metastatic Breast Cancer Patients Initiating Intravenous Paclitaxel/nab-Paclitaxel

Supplemental Material: Chemotherapy-Induced Peripheral Neuropathy in Metastatic Breast Cancer Patients Initiating Intravenous Paclitaxel/nab-Paclitaxel

Clinical and Economic Burden of Intravenous Paclitaxel or nab-Paclitaxel for Metastatic Breast Cancer

Supplemental Material: Clinical and Economic Burden of Intravenous Paclitaxel or nab-Paclitaxel for Metastatic Breast Cancer