Quality of life assessment in testicular non-seminomatous germ cell tumour survivors

Purpose: Patients with Germ cell tumours (GCT) are at risk of long-term toxicities due to multimodality therapy. It is debatable whether there is an impact on the quality of life(QoL) of GCT survivors. Methods: A case-control study was conducted at a tertiary care centre in India, using the EORTC QLQ C30 questionnaire, to compare the QoL between GCT survivors(disease free> 2 years) and healthy matched controls. A multivariate regression model was used to identify factors affecting QoL. Results: A total of 55 cases and 100 controls were recruited. Cases had a median age of 32 years (interquartile range, IQR 28-40 years), ECOG PS of 0-1(75%), advanced stage III (58%), chemotherapy (94%) and 66% were >5 years from diagnosis. The median age of controls: 35 years (IQR 28-43 years). Statistically signi�cant difference was seen for t emotional (85.8 14.2 vs 91.7 10.4, p 0.005), social(83.0 22.0 vs 95.2 9.6, p <0.001) and global scales (80.4 21.1 vs 91.3 9.7, p <0.001). Cases had more nausea and vomiting(3.3 7.4 vs 1.0 3.9, p 0.015), pain(13.913.9 vs 4.89.8, p<0.001), dyspnea(7.9 + 14.3 vs 2.7 9.1, p 0.007), and appetite loss(6.7 14.9 vs 1.9 7.9, p 0.016) and greater �nancial toxicity(31.5 32.3 vs 9.0 16.3, p <0.001). Adjusting for age, performance status, BMI, stage, chemotherapy, RPLND, recurrent disease, and time since diagnosis, no predictive variables were signi�cant. Conclusion: There is a detrimental impact of history of GCT in long term survivors of GCT.


Introduction
Germ cell tumour(GCT) commonly affects men between the ages of 20-40 years, contributing to 1-3% of all cancers occurring in men [1].Advances in the management of testicular cancer in the last thirty years have resulted in excellent long-term survival [2,3].The management of GCT and nonseminomatous germ cell tumour (NSGCT) in particular, involves multimodality therapy [4].Multimodality treatment involving different combinations of chemotherapy, surgery, and radiation act as a double-edged sword.Their high e cacy brings with it a host of undesirable adverse effects.
High survival rates thus lead to survivors of testicular cancer living with some degree of sequelae of the treatment.This includes the impact on various aspects of their life and late manifestations of underlying disease biology, predisposing them to other health risks [5][6][7][8].
GCT survivors are known to be at a higher risk of developing neurotoxicity, nephrotoxicity, ototoxicity, pulmonary toxicity, Raynaud's phenomenon, and hypogonadism [5,9].Some complications, such as the increased risk of cardiovascular disease and metabolic syndrome, are complex in origin and are believed to be due to interactions of macrovascular endothelial damage secondary to cisplatin and infradiaphragmatic radiation and testicular dysgenesis syndrome, linked with testicular tumour pathophysiology [10].Second malignancies, especially after pelvic and abdominal radiation and chemotherapy such as cisplatin and etoposide, are less common but may impact the survival of such patients [11,12].
GCT survivors have been reported to suffer from high rates of depression.[13] Furthermore, all three treatment modalities and the disease process related to testicular dysgenesis syndrome can negatively affect fertility [14,15].An adverse effect on self-perception of masculinity has also been reported in GCT survivors [16].In addition, they are also at risk of low sexual performance [17], psychosocial issues [8] and nancial toxicity [7].Anxiety and pressure in coping with the burden of cancer may lead to unhealthy lifestyle habits, which can further compound health risks to survivors [18].
Whether these factors ultimately compromise the quality of life(QoL) of GCT survivors is a matter of speculation [19][20][21].The most appropriate method of evaluation, timing, and how best to integrate measures of QoL into the holistic care of survivors are areas of ongoing investigation [22][23][24].Data from low-middle-income countries (LMIC) in patient-reported outcome measures and long-term survivorship are scarce for GCT survivors.
We present a case-control study on the quality of life of NSGCT survivors vs healthy controls attending the urogenital clinic, focusing on the late effects (> 2 years since diagnosis).

Methodology
We conducted a case-control study at the urogenital malignancy clinic at the All India Institute of Medical Sciences, New Delhi, a tertiary care centre in the capital of India, which registers about 10,000 new cancer cases per year.Survivors of NSGCT above 18 years who attended the clinic from January 2015 to July 2019 who gave informed consent for participation were included as cases.Patients had to be disease free for at least two years to be eligible.Controls were identi ed as age-matched healthy adult males (≥ 18 years), who accompanied patients in the clinic, in the ratio of 2:1 controls per case.These individuals were related to the cases in question, living in the same house/neighbourhood, had to be free of any comorbidities, not undergoing treatment for any illness or surgery in the last 4 weeks at the time of administration of the questionnaire, with no history of any malignancy currently or in the past, and capable of giving informed consent.Baseline data were collected through medical case record les.Patients were staged as per American Joint Committee on Cancer(AJCC) 8th edition TNM guidelines [25].Their performance status was ascertained using the Eastern Cooperative Oncology Group(ECOG) score [26].Using the Indian cut-offs of body mass index(BMI) grouping, they were divided into lean( BMI of less than 23kg/m 2 ) and overweight/obese(BMI ≥ 23kg/m 2 ) [27].
QoL was assessed as per the European Organisation for Research and Treatment of Cancer (EORTC) QLQ-C30 (version 3) questionnaire for both groups.
[28] The QLQ-C30 questionnaire comprises multi-item scales and single-item measures.There are ve functional scales (physical, role, cognitive, emotional, and social functioning), three symptom scales (fatigue, pain, and nausea and vomiting), six single items (appetite loss, diarrhoea, dyspnea, constipation, insomnia, nancial impact) and one global QoL measure.All questions use a four-point response pattern (1= "not at all," 2= "a little," 3= "quite a bit," and 4= "very much"), except the global QoL scale, which has a seven-point response format.The raw scores were linearly transformed into a score between 0 and 100 [29].Higher scores mean better outcomes for the functioning and the global QoL scales on the one hand, while on the other, they indicate a higher symptom burden for the symptom scales.The calculation of scores was done as per the EORTC scoring manual.Permission to use this questionnaire (in two languages, English and Hindi) was obtained before the conduct of this study.(Online resource 1) The study was approved by the institute's ethics committee (29.09.2018;IECPG-445/27.09.2018).

Statistics
Descriptive statistics were used to present the baseline characteristics of patients and controls.The transformed EORTC QLQ30 scores were compared for each category across both groups using a t-test for means and a rank sum test for the median.Each score was analyzed by an unconditional logistic regression model controlling for variables including age, performance status, body mass index, stage at diagnosis, chemotherapy exposure, etc. Multivariate logistic regression analysis was utilized to assess the impact of covariates on the statistical signi cance of all component score differences.
Since GCT is relatively rare, for adequate power, a 2:1 control-to-case ratio was taken [30,31].A p-value of < 0.05 was considered signi cant and all values were two-sided.Data entry was done using Microsoft Excel, and analysis was done with STATA 14 (StataCorp™, College Station, TX, USA).

Baseline characteristics
A total of 55 NSGCT survivors and 100 controls were included in the study.The median age of patients was 32 years (interquartile range, IQR 28-40 years), and of healthy controls, 35 years (IQR 28-43 years).The majority of patients had an ECOG PS of 0-1(75%), advanced stage III (58%), and received chemotherapy (94%).About 2/3rd of all patients were more than ve years from diagnosis, and 11% had a history of recurrence.The baseline characteristics of the patients are summarized in Table 1.

Quality of life scores
On comparing functional scores, we found lower scores in all domains for patients as compared to those for controls.This difference was statistically signi cant for the emotional (85.8 ± 14.2 vs 91.7 ± 10.4, p 0.005),social(83.0±22.0 vs 95.2 ± 9.6, p < 0.001) and global (80.4 ± 21.1 vs 91.3 ± 9.7, p < 0.001) areas.At the same time, it was not found to be different for the physical, role and cognitive domains (Table 2).As can be seen from Table 3, NSGCT survivors had a higher symptom load for all symptoms in comparison to healthy controls.This difference in symptom burden was statistically signi cant for nausea and vomiting(3.3 ± 7.4 vs 1.0 ±3.9, p 0.015), pain(13.9±13.9vs 4.8±9.8,p < 0.001), dyspnea(7.9± 14.3 vs 2.7 ± 9.1, p 0.007), and appetite loss(6.7 ± 14.9 vs 1.9 ± 7.9, p 0.016).The difference in fatigue, diarrhoea and constipation, although numerically greater in the survivors, did not reach the prespeci ed parameters of statistical signi cance.

Appetite loss Median
Mean + SD 0 (0-0) 6.7 + 14.9 0 (0-0) 1.9 + 7.9  1A show a left skew with maximum variation in social and emotional scores as compared to physical, role and cognitive scores.The median value of these parameters approaches 100, indicating high functional ability.In Fig. 1B, the symptom scores demonstrate a right skew for all measures.The symptom burden for nausea/vomiting, dyspnea and appetite loss seems to be minimal for most patients, barring a few outliers.The maximum variation and highest scores were seen for nancial di culty with a widespread range of values.Constipation and insomnia appear to be persistent concerns in this population, judging from the symptom burden, and the persistence of fatigue and pain is also apparent, NSGCT survivors had statistically greater nancial toxicity (Table 3) as compared to their matched controls (31.5 ± 32.3 vs 9.0 ± 16.3, p < 0.001) On multivariate analysis, when adjusting for age, performance status, BMI, stage, chemotherapy, RPLND, recurrent disease, and time since diagnosis, no predictive variables were signi cant (Table 4).

Discussion
Our study reveals a detrimental impact of cancer diagnosis in lives of NSGCT survivors as compared to normal population, which is not tempered with the increasing passage of time.The consequence of GCT treatment affects emotional, social and nancial domains signi cantly.We found a considerable symptom burden of dyspnea, pain, loss of appetite and nausea/vomiting that persisted even though 60% of our patients were more than ve years from their initial diagnosis.All these factors likely contributed to the signi cant numerical and statistical difference seen in the global QoL between the cases and healthy controls.
QoL impairment has been a subject of debate among GCT survivors.Mykletun et al. demonstrated that on a long-term follow-up of 11 years, GCT survivors achieved comparable QoL as that of the general population [19].They found that the treatment modality did not determine QoL.However, self-reported stress and adverse effects could help identify low QoL groups.On the other hand, Kerns et al [20] found the cumulative burden score, a composite of a questionnaire and physical examination, to be high in 20% of GCT patients.Speci cally, receiving platinum-based chemotherapy for four cycles was associated with receiving a high cumulative burden score, whereas there was no impact of stage on multivariate analysis.Most of the patients in our study (94%) were exposed to chemotherapy, which could explain our results aligning with those of Kerns et al. [20].Socioeconomic factors [32] and nancial toxicity [7] have been shown to interact and in uence QoL in complex ways.Anxiety and depression, both more common in GCT [5], may impact QoL [13,33].

Multivariate regression analysis
The method of assessment of QoL could also be an important factor that can alter results.We utilized the EORTC QLQ C30 [28], a standard questionnaire used across cancers and available in many languages.It has been utilized in several studies, alone or in combination with other questionnaires [7,34].It also had the advantage of being validated in Hindi, which was relevant in a north Indian setting [35].
Others have used a GCT-speci c questionnaire, the EORTC -TC26 [36], that focuses on the issues of sexuality and masculinity.Some studies have also used other broad questionnaires, such as SF 36 [19].
The best among them and how to compare and contrast results across studies using different methods is still being evaluated [22].
The strengths of this study are the enrollment of GCT cases with a long follow up, comparison with healthy controls, and the generation of QoL data for the rst time from an LMIC country.Limitations include its small sample size, single-point analysis, lack of use of a dedicated GCT questionnaire, relatively shorter follow-up, and lack of details of socioeconomic and health economic perspective.
To conclude, our study demonstrates that QoL is signi cantly affected for GCT survivors as compared to controls in a tertiary care centre in India.Decreased QoL is seen in several domains, and this is unaffected by differences in stage, treatment modality or the general condition of the patient.The difference in symptom burden for variables such as nausea/vomiting, dyspnea and appetite loss may have been driven by a few patients rather than the whole group.Financial toxicity remains a pertinent concern.Further studies are needed to explore the best method to measure different aspects of QoL and to help identify predictors that could identify which patient subgroups may have the long-term burden of impaired QoL.

Declarations
Con ict of interest: Funding: The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.
Financial and/or competing interests: The authors have no relevant nancial or non-nancial interests to disclose.
-seminomatous germ cell tumour; IQR: Interquartile range; SD: Standard deviation Figure1depicts the box and whisker plot of the functional scores(1A) and symptom scores(1B) of the NSGCT survivors in the study.The functional scores in Fig.

Figures Figure 1 EORTC
Figures

Table 1
Baseline characteristics of patients with testicular non-seminomatous germ cell tumour (n=55) a Categorized as lean and overweight/obese as less than 23 kg/m 2 and > 23 kg/m 2 , respectively.ECOG PS: Eastern Cooperative Oncology Group Performance Status; IQR interquartile range; RPLND: Retroperitoneal Lymph node dissection.

Table 3 :
Comparison of the EORTC QLQ-C30 symptom scores between testicular NSGCT survivors and controls

Table 4 :
Relationship between the baseline and treatment characteristics, and the global score in EORTC QLQ-C30 [23]d that the difference in global QoL did not change when adjusted for age, performance status, stage, chemotherapy, RPLND and recurrence.With the limitation of a small sample size in the subgroups, this implies that global QoL is driven by factors beyond treatment modality and disease status.The impact of time since diagnosis (with a cut-off of 5 years) on global QoL had a trend toward statistical signi cance on multivariate analysis.The correlation of time since diagnosis on QoL in NSGCT survivors has been previously reported by Jovanovski et al.[23].They used the short form 12 (SF 12) questionnaire, a general questionnaire and a shorter version of the SF 36 questionnaire and found that ve years after diagnosis, the physical component score of the scale slightly improved in NSGCT survivors while the mental component score declined.With the caveat of limitations of crossstudy comparisons, use of different questionnaires, and small sample size, these results indicate the need for further studies to explore the temporal relationship of QoL in GCT survivors.