Musician's Experience After Breast Cancer Treatment: Defining Musical Toxicity and its Frequency

Abstract

PURPOSE

Over 50% of households in the United States have at least one musician—many musicians are also breast cancer survivors. This group has not been well studied, and given the level of fine sensory-motor skill required for musicianship, we hypothesized that musicians experience unique manifestations of breast cancer treatment toxicities.

METHODS

A nine-item Musical Toxicity Questionnaire (MTQ) was distributed to patients who had consented to participate in the Mayo Clinic Breast Cancer Registry. The MTQ screened participants by asking if they played a musical instrument or sang in the last 10 years: questions populated for those who answered yes. Respondents were asked if they noticed difficulty with their musical endeavor during or after breast cancer treatment, defined as acute musical toxicity (AMT). The questionnaire asked which side effect and cancer-directed therapy most influenced musical ability, what musical attributes were affected, and the timeline of resolution. Multivariable and classification tree analyses assessed relationships between AMT and treatment characteristics.

RESULTS

Of 1,871 survey respondents, 29% (535/1,871) self-identified as musicians. Over a quarter (27%, 144/535) reported AMT, and for 57% (82/144), AMT had not resolved at the time of survey. Of the treatments each participant received, chemotherapy was most often reported as most negatively impactful (63/89 who received chemotherapy, 71%). Decreased endurance was the most common musical difficulty (64% of those with AMT, 92/144), followed by decreased accuracy, trouble playing/singing quickly, and difficulty using proper technique. Multivariable and classification tree analyses revealed that receipt of chemotherapy was most strongly correlated with AMT.

CONCLUSION

These results will help oncology care teams counsel musicians, answer questions about impacts on musicality, and provide a timeline for resolution of musical symptoms.

BACKGROUND

Many breast cancer survivors are also musicians, as one in eight women will be diagnosed with breast cancer during their lifetimes and over 50% of American households have at least one musician.¹ More than 46 million Americans participate in a professional or amateur choir,² one in five older adults actively plays a musical instrument,³ and numbers are higher across all ages for the popular piano and guitar.^4,5 Musical endeavors require a high level of fine sensorimotor control, and musicians likely experience unique manifestations of breast cancer treatment toxicities. Chemotherapy-induced peripheral neuropathy (CIPN),⁶ upper-extremity dysfunction after surgery and radiation,^7,8 and joint pain associated with aromatase inhibitors⁹ are just a few toxicities that could be devastating for professional and amateur musicians alike.

CONTEXT

• Key Objective

• How do musicians experience breast cancer treatment, and how do treatment side effects affect musical ability?

• Knowledge Generated

• Over a quarter of musicians in this study experienced difficulty with their musical endeavor during or after cancer treatment. Chemotherapy contributed more than other treatments to musical toxicity, and for most musicians, musical abilities never returned to baseline.

• Relevance

• Oncology care teams should counsel musicians that fatigue and pain or discomfort could limit their ability to participate in musical endeavors during and after treatment, and that their musical ability could be permanently affected by breast cancer treatment.

Occupational outcomes are understudied in oncology, and whether a patient is a musician, surgeon, craftsman, or seamstress, common toxicities of breast cancer treatments may cause serious and long-lasting impairments of fine motor control. Work contributes to self-identity and facilitates social interaction, and failure to return to work after treatment can be difficult,^10-12 especially for patients who strongly associate their sense of self with their work—as is the case for professional musicians.^12,13 Amateur musicians enjoy a sense of fulfillment and community from their endeavors as well.^14-16

Similarly, there is very little in the medical literature concerning musicians diagnosed with cancer.^13,17,18 Investigating musical difficulty because of cancer treatments, which we call musical toxicity (akin to renal toxicity or skin toxicity), should lead to improved communication between care teams and patients, and create opportunities for early therapy interventions. Within emerging de-escalation treatment paradigms in breast cancer, sharing the risk of experiencing musical toxicity, the timeline over which it might resolve, and which treatments are most likely to affect musical ability should improve shared decision making. In the current study, musician breast cancer survivors were surveyed to assess the prevalence of acute musical toxicity (AMT), difficulty with musical endeavors during or immediately after breast cancer treatments, and to determine which medical symptoms and musical attributes were most associated with experiencing AMT.

METHODS

Survey Development

Comprehensive methods and the Musical Toxicity Questionnaire (MTQ) can be found in the Data Supplement (online only). The survey was approved by the Mayo Clinic institutional review board.

Experts in medical oncology, psychology, radiation oncology, surgery, and music education designed the MTQ. Participants were first asked if they had played music or sung in the last 10 years. If no, the survey ended, and if yes, the response was categorized as a musician. Musicians then selected their musical endeavor(s) and if they experienced difficulty with music during or after cancer treatments. If yes, the response was classified as acute musical toxicity, and endeavor-specific questions populated.

Participants who reported AMT were asked what symptom bothered them most and which treatment(s) were perceived to have most negatively affected musical ability: more than one response could be selected. Difficulty with specific aspects of musicality were queried, and participants were encouraged to select all that applied. Two instrument-specific items were omitted for vocalists. Respondents answered whether their musical abilities had returned to baseline, and over what time course. Musical ability today was rated on a scale of 0-10.

Survey Distribution

The MTQ was distributed via Research Electronic Data Capture to a subset of patients from the Mayo Clinic Breast Cancer Registry (MCBCR) who had email addresses on file and who had not received a survey in the previous 6 months. The MCBCR is a clinic-based registry that enrolls and follows patients with newly diagnosed breast cancer. Participants provide permission to review their medical records and consent to receive surveys. Patient-Reported Outcomes Measurement Information System (PROMIS-10) Global Physical and Mental Health scores from 1 year after diagnosis were extracted for respondents. Chart review was conducted to record the treatments received by each participant.

Analysis

Analysis included all musician responses. Descriptive statistics were conducted, and univariate analyses were performed to assess correlations between treatment variables. Univariate and multivariable logistic regressions as well as chi-squared tests were analyzed to identify relationships between AMT and treatment variables, demographics, and PROMIS scores. To understand the interaction of treatment variables on AMT, a classification tree analysis (CTA) was conducted. The full tree was pruned to reduce overfitting, and a complexity parameter minimized both branching points and error. For analysis, the overall cohort was broken down into three groups: singers, nonwind instrumentalists, and wind instrumentalists.

RESULTS

Musicians in the MCBCR

The survey was sent to 4,075 participants in the MCBCR, and 1,871 responded (46% response rate). In total, 557 responders self-identified as musicians, and after removing nonmusicians and blank responses, 535 respondents (29%) reported 802 unique musical endeavors, as many participants played multiple instruments or both sang and played an instrument.

The most commonly reported musical endeavors were singing (407/802 or 51% of musical endeavors), keyboard instruments (247/802, 31%), and plucked string instruments (62/802, 8%). Most respondents had stage I breast cancer (211/535 participants, 40%) or ductal carcinoma in situ (81/535, 15%), and 32% (172/535) had node-positive disease. The median time from diagnosis to survey was 5.3 years. Nearly all respondents (524/535, 98%) underwent surgery, of which 55% (292/535) underwent mastectomy and 44% (228/535) underwent only lumpectomy. Twenty-one percent (105/520) of patients who had a documented breast surgery underwent axillary lymph node dissection (ALND). Sixty-nine percent or 359/520 patients had sentinel lymph node surgery (SLNS, a term analogous to sentinel lymph node biopsy). Additional treatment and demographic characteristics can be seen in Table 1.

TABLE 1.

Demographic and Treatment Characteristics

Characteristic	No.	Proportion, %
		By Participant (n = 535)	By Musical Endeavor (n = 802)
Musical endeavora
Singing	407	76.1	50.7
Keyboard instruments	247	46.2	30.8
Plucked strings	62	11.6	7.7
Woodwinds	32	6.0	4.0
Bowed strings	23	4.3	2.9
Percussion	19	3.6	2.4
Brass	6	1.1	0.7
Other	6	1.1	0.7
Age at diagnosis, years
Median	65
Range	32-88
<40	41	7.7
40-49	117	21.9
50-59	152	28.4
60-69	142	26.5
70-79	80	15.0
≥80	3	0.6
Sex
Female	535	100
Diagnosis to survey completion, years
Median	5.3
Range	0.2-22
Stage
0 (DCIS)	81	15.2
I	211	39.5
II	169	31.6
III	64	12.0
IV	9	1.7
Unknown	1	0.2
Treatments
Breast surgery (n = 535)b
Lumpectomy	228	42.6
Mastectomy	292	54.6
No surgery	12	2.2
Unknown	3	0.6
Axillary surgery (n = 520)
SLNS	359	69.0
ALND	105	20.2
No nodes removed	52	10.0
Unknown	4	0.8
Radiation	347	64.9
Chemotherapy	231	43.2
Endocrine therapy	384	71.8

Abbreviations: ALND, axillary lymph node dissection; DCIS, ductal carcinoma in situ; SLNS, sentinel lymph node surgery.

^a Keyboard instruments: piano, organ, harpsichord, etc. Plucked strings: guitar, harp, electric bass, mandolin, ukulele, etc. Woodwinds: clarinet family instruments, double reed instruments, saxophones, flute family instruments. Bowed strings: violin, viola, cello, bass. Percussion: including drum set. Brass: trumpet, trombone, French horn, tuba, etc.

^b Breast surgery indicates the most aggressive breast surgery that a participant had throughout their breast cancer treatments. If a participant underwent both mastectomy and lumpectomy, this was classified as mastectomy.

AMT and Associated Symptoms

Overall, 27% of respondents (144/535) reported that cancer treatment affected at least one musical endeavor, and 109/144 (76%) of these respondents sang or played less than they would have normally because of this difficulty. Bowed strings players reported the highest proportion of AMT (7/23, 30%), while percussion players reported the lowest (3/19, 16%). Difficulty singing or playing for long periods of time was reported by 64% (92/144) of those with AMT, followed by trouble with accuracy (64/144, 44%), singing or playing quickly (52/144, 36%), and singing or playing with proper technique (52/144, 36%). In the wind instrument subgroup, playing long notes was the most commonly reported musical difficulty (8/9, 89% of this subgroup). Chi-squared analysis revealed that younger patients reported more AMT (P = .005), with 34% (14/41) of those younger than 40 years at diagnosis and 36% (42/117) of those age 40-49 years reporting AMT. Patients in older decades of life reported less AMT (30% of those age 50-59 years, 22% age 60-69 years, and 14% age 70-79 years).

Twenty-two percent of all musical endeavors (173/802) were affected by cancer treatment. Fatigue (105/173, 61%) and pain or discomfort (77/173, 45%) were most frequently reported as the most bothersome medical symptom. Within the nonwind instrument subgroup, pain or discomfort was most bothersome (53/77, 69%) while fatigue was less commonly reported (42/77, 55%). The opposite trend was seen for singers and wind instrumentalists, where fatigue was most common (55/87, 63% for singers, and 8/9, 89% for wind instruments). Musicians in these two cohorts also wrote in upper respiratory problems as a significant barrier, with 24% (21/87) of singers and 78% (7/9) of wind instrumentalists making note of this. See Table 2 for details.

TABLE 2.

The Incidence of Acute Musical Toxicity by Musical Endeavor. Respondent-Perceived Most Impactful Symptom and Treatment As Well As Specific Musical Toxicities Are Reported on an Overall Participant Level As Well As Instrument Sub-Group Level

AMT	No.	%
By participant with AMT (n = 144)	144/535	26.9
By musical endeavor with AMT (n = 173)	173/802	21.6
Singing	87/407	21.4
Keyboard instruments	46/247	18.6
Plucked strings	18/62	29.0
Woodwinds	8/32	25.0
Bowed strings	7/23	30.4
Percussion	3/19	15.8
Brass	1/6	16.7
Other	3/6	50.0

By Participant (n = 144)	No.	%	By Grouped Musical Endeavor
			Singers (n = 87)	Nonwind Instruments (n = 77)	Wind Instruments (n = 9)
Most impactful symptoma
Fatigue	68	47.2	63.2	54.5	88.9
Pain or discomfort	49	34.0	21.8	68.8	55.6
Numbness or tingling	29	20.1	3.4	36.4	0.0
Arm/hand weakness (instrumentalists only)	26	18.1	NA	37.7	0.0
Voice changes (singers only)	28	19.4	32.2	NA	NA
Upper respiratory problem (wind instrumentalists and singers only)	19	13.2	24.1	NA	77.8
Arm/hand swelling or heaviness (instrumentalists)	17	11.8	NA	22.1	11.1
Sensing where body is in space (instrumentalists)	4	2.8	NA	5.2	0.0
Other	24	16.7	10.3	16.9	22.2

By Participant (n = 144)	No.	%	By Grouped Musical Endeavor
			Singers (n = 87)	Nonwind Instruments (n = 77)	Wind Instruments (n = 9)
Most impactful treatmentb
Chemotherapy	63/89	70.8	67.3	71.4	40.0
Radiation	35/104	33.7	34.4	31.5	50.0
Surgery	39/140	27.9	14.0	46.7	37.5
Endocrine therapy	27/109	24.8	20.0	26.9	14.3

By Participant (n = 144)	No.	%	By Grouped Musical Endeavor
			Singers (n = 87)	Nonwind Instruments (n = 77)	Wind Instruments (n = 9)
Specific musical toxicityc
Endurance	92	63.9	64.4	58.4	77.8
Accuracy	64	44.4	54.0	22.1	22.2
Singing/playing quickly	52	36.1	21.8	51.9	11.1
Singing/playing with proper technique	52	36.1	32.2	37.7	33.3
Singing/playing long notes	43	29.9	42.5	2.6	88.9
Singing/playing with dynamic variation	30	20.8	18.4	14.3	33.3
Sensing how much pressure	16	11.1	NA	24.7	0.0
Hearing oneself and blending	15	10.4	13.8	3.9	0.0
Sensing distance between fingers or up and down the instrument	10	6.9	NA	11.7	22.2
Other	7	4.9	5.7	2.6	0.0

Abbreviations: AMT, acute musical toxicity; NA, not applicable.

^a Most impactful symptom.

^b Most impactful treatment: respondents were given the option of selecting multiple options if they felt that multiple symptoms or treatments most affected them. Total percentages are >100% because of the ability to select more than one option, and the combination of more than one musical endeavor per participant. Most impactful treatment: few responses of not sure, none, and other are not shown here.

^c Specific musical toxicity: respondents were encouraged to select as many musical attributes as were affected by breast cancer treatments.

Treatment Association With AMT

Of the participants with AMT who received chemotherapy, 71% (63/89) felt that it had the most negative impact of all treatments received. A smaller proportion of patients who underwent radiation therapy (35/104, 34%), surgery (39/140, 28%), and endocrine therapy (27/109, 25%) felt that those treatments were the most impactful. Participants could select more than one most impactful treatment. Although numbers were small, bowed string players, plucked string players, and percussionists felt that surgery was most impactful. See Table 2 for additional data.

Duration of Musical Toxicity

Fifty-two percent (90/173) of musical endeavors affected by AMT had not returned to baseline by the time of survey. Most respondents with AMT (73%, 105/144) reported that their musical ability for at least one musical endeavor did not recover or took >1 year to recover. Mean ability on a scale of 0-10 was 6.2 (standard deviation [SD], 2.7) for the entire cohort. This was lowest for bowed string players (mean, 5.2; SD, 3.0), wind players (mean, 5.9; SD, 3.9), and singers (mean, 5.9; SD, 2.6). See Table 3 for details.

TABLE 3.

The Timeline Over Which Musical Toxicity Resolved, and Respondent-Reported Ability at the Time of Survey Compared to Before Starting Breast Cancer Treatment

By Musical Endeavor With AMT (n = 173)	Count	%
When did AMT resolve?
During treatment	1	0.7
<3 months after finishing treatment	9	6.1
3-6 months after treatment	16	10.9
>6 months but <1 year	14	9.5
Between 1 and 2 years	12	8.2
>2 years after treatment	5	3.4
Never: I still have difficulty	90	61.2

By Musical Endeavor With AMT (n = 173)	Mean	Median
Ability today on a scale of 0-10a
Singing (n = 87)	5.9	6.0
Keyboard instruments (n = 46)	7.0	7.0
Plucked strings (n = 18)	6.0	5.5
Woodwinds (n = 8)	5.9	7.5
Bowed strings (n = 7)	5.3	5.0
Percussion (n = 3)	6.0	7.0
Brass (n = 1)	10.0	10.0
Other (n = 3)	8.3	10.0

Chronic Musical Toxicity	Count	%
All instrument groups by participant (n = 144)	105	72.9
For whom AMT took >1 year to resolve, or never resolved
Singing (n = 87)	67	77.0
Nonwind instruments (n = 67)	42	62.7
Wind instruments (n = 9)	6	66.7

Abbreviation: AMT, acute musical toxicity.

^a Ability today on a scale of 0-10, where 0 represents no longer able to play at all and 10 represents playing just as well, if not better, than before treatments began.

Multivariable Modeling and Classification Tree Analysis (CTA)

For the subset of patients who underwent surgery (n = 520), a series of univariate logistic regressions identified that multiple treatments were associated with AMT: chemotherapy, volume of radiation, mastectomy, type of axillary surgery, and reconstruction. In a multivariable logistic regression model, only receipt of chemotherapy (P < .001) and receipt of comprehensive regional nodal irradiation (RNI) as opposed to no radiation (P = .023) were significantly associated with AMT. Stage and nodal status were associated with AMT in univariate analysis, but because the objective of the analysis was to identify modifiable treatment factors, and because disease characteristics have a causal relationship to treatment factors, these items were excluded from multivariable analysis. See Table 4.

TABLE 4.

The Timeline Over Which Musical Toxicity Resolved, and Respondent-Reported Ability at the Time of Survey Compared to Before Starting Breast Cancer Treatment

Univariate Regressions		Full Logistic Regression
Variable	P	Variable	P
Node positive	<.0001	Disease variables removed from multivariable analysis
Stage	<.0001
Chemotherapy	<.0001	Chemotherapy	<.0001
Volume of radiation	<.0001	Volume of radiation	.0365
Whole/partial breast v none	.6849	Whole/partial breast v none	.0853
Comprehensive nodal v none	<.0001	Comprehensive nodal v none	.023
Comprehensive v whole/partial	.0003	Comprehensive v whole/partial	.8769
Mastectomy	.0127	Mastectomy	.5305
Type of axillary surgery	.0005	Type of axillary surgery	.3017
SLNS v none	.0183	SLNS v none	.1818
ALND v none	.0006	ALND v none	.4559
ALND v SLNS	.0138	ALND v SLNS	.5278
Reconstruction	.0450	Reconstruction	.6564
Endocrine therapy	.2948	Endocrine therapy	.1362

Abbreviations: ALND, axillary lymph node dissection; SLNS, sentinel lymph node surgery.

Because of the strong correlation between treatment variables (Fig 1 and Table 4), a CTA was used to determine what combination of variables best classified patients by AMT. In the cohort of 520 patients, the CTA selected chemotherapy as the treatment that best categorized respondents with AMT. RNI was selected as the next best variable. For patients not receiving chemotherapy, the CTA selected reconstruction as associated with AMT. For patients receiving chemotherapy and any volume of radiation, ALND did not seem to be associated with AMT, while this was an important factor for patients not receiving radiation. Patients who did not require chemotherapy and who received either partial-breast, whole-breast, or no radiation were least likely to report AMT. Figure 2 shows the full classification tree.

FIG 1.

The correlation between various treatment variables is represented in the heat map, with the strongest positive correlations represented by dark blue and the strongest negative correlations represented by dark red. ALND, axillary lymph node dissection; AMT, acute musical toxicity; RT, radiation therapy; SLNS, sentinel lymph node surgery.

FIG 2.

Because of the high correlation between treatment variables as demonstrated in Figure 1, a CTA was performed to evaluate what combination of variables best classified patients by experiencing AMT. In the overall cohort of patients, 27% reported AMT. 38% of those who received chemotherapy had AMT while only 18% of those who did not receive chemotherapy reported AMT. The tree continues to branch, identifying variables that select for AMT at each level. ALND, axillary lymph node dissection; AMT, acute musical toxicity; CTA, classification tree analysis; RT, radiation therapy; SLNS, sentinel lymph node surgery.

Finally, of the patients who underwent surgery (n = 520), univariate regressions comparing PROMIS-10 scores (at 1 year after diagnosis) to AMT were analyzed. The odds of having AMT decreased 8.6% for each additional point in the composite PROMIS mental health score and by 14% for each additional point in the composite physical health score. The probability that a patient experienced AMT with zero lymph nodes removed was 22%, which increased by 3.5% with each additional lymph node removed (P = .001). Neither cumulative chemotherapy dose nor surgery laterality was associated with AMT.

DISCUSSION

To our knowledge, this study is one of the first to describe the impact of cancer treatments on musician's ability. About one third of respondents were musicians, and over a quarter of musicians experienced difficulty with music during and/or after cancer treatment. For most of these patients, AMT never resolved, with musical ability remaining lower than before treatment.

The distribution of musical endeavors in this survey (singing, followed by keyboard instruments [piano] then plucked string instruments [guitar]), mirrors the popularity of musical activities among the general public.¹⁹ Although neither musical background nor education were queried, one half of all reported musical endeavors were singing, indicating that our cohort likely represents amateur musicians who primarily sing in community choirs.

Upper-extremity pain and dysfunction are prevalent among breast cancer survivors.^20,21 In one study, 44% of patients who underwent ALND, RNI, and chemotherapy sacrificed activities they enjoyed because of upper-extremity dysfunction,²⁰ and up to 45% of patients undergoing ALND and RNI will experience lymphedema.^8,22 Asymmetry of the shoulder and trunk musculature may develop after surgery, which can result in postural changes,²³ poor upper-body stabilization, and difficulty performing fine motor skills.²⁴ This dysfunction would make playing an instrument or standing/sitting to sing with proper technique more challenging, and may be especially impactful for musicians whose instruments require higher levels of upper-extremity utilization, such as bowed string instruments, percussion instruments, or holding a wind instrument aloft while playing. Indeed, in this study, comparatively more instrumentalists reported that surgery was a most negatively impactful treatment when compared with vocalists.

Interestingly, the CTA showed that for patients who received chemotherapy and radiation, SLNS or no axillary surgery appeared to better categorize patients with AMT compared with ALND. This is surprising, given evidence that ALND contributes most to the risk of musculoskeletal problems and lymphedema.^7,25-29 One explanation for this could be that our clinical practice guidelines include early physical therapy assessment for patients who are planned for ALND. This higher level of support may help mitigate musical toxicity, as multiple studies have shown lower rates of lymphedema and upper-extremity dysfunction with therapy intervention.^30,31 The opposite classification was selected for patients who received chemotherapy but no radiation: after mastectomy, ALND was selected to better classify AMT for this group of patients.

Many patients who receive taxane-based chemotherapy experience long-lasting peripheral neuropathy.^6,32 The prevalence of chronic neuropathy ranges anywhere from 11% to 80%,³³ and CIPN has been associated with fall risk, insomnia, anxiety, depression,³⁴ and lower overall quality of life.³⁵ Nonwind instrumentalists were most likely to report that numbness or tingling was a most negatively impactful symptom, and chose pain or discomfort as the overall most negatively impactful symptom. We suspect that peripheral neuropathy played a role in these experiences. Additionally, younger patient's musical ability was more affected than older patient's ability—perhaps because older patients are less likely to receive chemotherapy.

There was concordance between patient-reported perception of most negatively impactful treatment, multivariable analysis, and the CTA: all three methods showed that chemotherapy, followed by RNI, was most associated with reporting AMT. Reduced ability to sing or play for long periods of time and difficulty playing or singing quickly were the two musical attributes most affected by treatment. Similarly, fatigue and pain or discomfort were the two most limiting medical symptoms. A causal relationship between chemotherapy or RNI and AMT cannot be definitively asserted, but it is reasonable to postulate that the fatigue that comes along with both treatments strongly contributed to difficulties with not being able to sing or play for long periods of time. As a complete course of chemotherapy often takes several months and toxicity is cumulative, it is possible that the length of treatment alone affected participant's perception of a link between chemotherapy and fatigue.

The reported pain and discomfort could be associated with CIPN, skin or musculoskeletal discomfort after surgery or radiation, or joint pain associated with aromatase inhibitors. Difficulty playing or singing quickly may be a result of fatigue as well as CIPN, heaviness or pain associated with shoulder dysfunction, or lymphedema. These side effects would also affect coordination and the ability to use proper instrument or singing techniques.

The true impact of breast cancer treatments on musicians' abilities and quality of life requires more in-depth investigation, and likely varies with each musician's experience level and specific musical endeavor. Given that music is a factor in self-identity, maintenance of well-being, and fewer feelings of isolation and loneliness,¹⁶ we postulate that the impact of lost musical ability would be substantial for many patients. During the COVID-19 pandemic, many musical groups were forced to meet virtually or not at all, to the detriment of the well-being of the musicians in those groups.³⁶ One study reported that the ability to meet virtually maintained a sense of community, well-being, and social identity among choir members.³⁶ Singing or playing less than one would have otherwise was reported by the majority of patients experiencing AMT in our study, which may have contributed to overall poorer well-being, as reflected in PROMIS scores.

There is a large body of research investigating the use of music in the cancer care setting, including music therapy.^37,38 The use of music has consistently been shown to decrease anxiety and pain, and aid in emotional processing and connecting with the preillness self,^38-40 even if used virtually.⁴¹ Although not queried in our survey study, we suspect that musicians use their talents as a physical and psychological method of coping and recovery, either intentionally or subconsciously.

Despite comprising a large proportion of the population, musicians remain largely underrepresented in the medical literature. The toxicities of cancer treatment affect musicians differently than nonmusicians, but further research is needed to fully define this variation in experience. Mixed-methods investigations combining quantitative data (surveys, and neurocognitive or dexterity testing), qualitative interview data, and objective longitudinal musical data by tracking one's ability through treatments would help understand the impacts of treatment on ability as well as the effects of playing music on global well-being. This approach could similarly be applied to specialized cohorts such as athletes, surgeons, or other occupations that require a high level of fine motor dexterity.

Although the current study adds to a relatively unexplored area of the literature, it is limited in several ways. Breast cancer survivors have experience with a wide variety of treatments, but determining which treatment and symptom contributed to what musical side effect is limited to hypothesis. All surveys are subject to recall bias, and the median time from diagnosis to survey was just over 5 years. During this time, participant's musical abilities may have declined because of lack of practice, injuries, or natural aging, and this was not accounted for. We did not query musical proficiency level, so it is difficult to say how these results may generalize to individual musicians or musicians with other types of cancer. Few patients in this study received immunotherapy. The survey did not include items specific to neurocognitive function, and multiple respondents noted difficulty with processing music or sight reading in free response areas of the survey. We hope to address these shortcomings in a future mixed-methods study by including musicians who are <5 years removed from treatment, posing interview questions about cognitive function and mood, and targeting the connection between fatigue, pain, neuropathy, and the timing of these side effects with multimodal cancer treatments. We also hope to elucidate helpful interventions and therapies to make meaningful recommendations for future musicians.

In conclusion, over a quarter of musicians who underwent breast cancer treatment experienced AMT, and for the majority of these patients, abilities did not return to baseline after completing treatment. Chemotherapy and RNI were most associated with AMT, by patient-reported perception, multivariable analysis, and CTA. Oncology care teams should counsel musicians that fatigue and pain or discomfort may impede musical ability, which may never make a full recovery. Medical, surgical, and radiation oncologists should have a risk/benefit discussion with musical patients who highly value their musical ability.

AUTHOR CONTRIBUTIONS

Conception and design: Jessica F. Burlile, Janet E. Olson, Valerie Cangie, Shawna Ehlers, Kathryn J. Ruddy, Dean A. Shumway, Charles L. Loprinzi, Elizabeth J. Cathcart-Rake

Administrative support: Fergus J. Couch

Provision of study materials or patients: Fergus J. Couch, Janet E. Olson

Collection and assembly of data: Jessica F. Burlile, Nicole L. Larson, Jennifer L. Bradt, Fergus J. Couch, Janet E. Olson, Charles L. Loprinzi

Data analysis and interpretation: Jessica F. Burlile, Joshua D. Cameron, Heather J. Gunn, Nicole L. Larson, Judy C. Boughey, Mary M. Mrdutt, Valerie Cangie, Shawna Ehlers, Yasamin Sharifzadeh, Dean A. Shumway, Charles L. Loprinzi

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Musician's Experience After Breast Cancer Treatment: Defining Musical Toxicity and its Frequency

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/op/authors/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).