Daniel Satele

Background/Objectives. Decades of research have failed to uncover effective approaches to prevent chemotherapy-induced peripheral neuropathy (CIPN), a common side effect of neurotoxic chemotherapy. Increased interest in placebo response as a potentially under-recognized confounder in CIPN trials was recently sparked by the results of a multisite NCI-funded phase II-III CIPN prevention study of duloxetine, a promising serotonin–norepinephrine reuptake inhibitor that enhances pain-inhibitory mechanisms within the central nervous system. Study findings revealed high and nearly equivalent response rates in three randomized treatment groups—little to no CIPN was reported by 65.2%, 66.0%, and 68.0% of study participants who received duloxetine 30 mg, 60 mg, or placebo treatment, respectively. Methods. We performed a meta-analysis of placebo response rates from seven randomized, double-blinded, placebo-controlled trials conducted over the past 20 years and comprising 191 placebo participants that were specifically testing interventions for oxaliplatin- and paclitaxel-induced peripheral neuropathy and that serially collected patient responses on the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire for Chemotherapy-Induced Neuropathy questionnaire. Additionally, we sought to identify trial- and patient-specific factors that predicted higher placebo response rates from a participant-level pooled analysis. Results. The placebo response rate was 10.0% [95% CI: 5.8%, 16.6%] when response was defined more conservatively as patients reporting no neuropathy at all. When the placebo response was defined more broadly based on patients reporting no or a little neuropathy, the placebo response rate was higher (39.6% [95% CI: 27.4%, 53.2%]). Male participants, receipt of oxaliplatin, and a 2:1 randomization ratio favoring the intervention arm were individually associated with a higher placebo response. Conclusions. High placebo response rates can threaten scientific progress toward identifying effective treatments for cancer treatment-associated side effects, like CIPN. Careful attention to study design factors, participant eligibility, and patient and research staff expectations may help to minimize placebo response rates in future CIPN intervention studies.

While oxaliplatin-induced peripheral neuropathy (OIPN) is more common and severe in patients who receive the previous standard, 6-month oxaliplatin-based treatment, we hypothesized that OIPN was still pervasive in patients who received shorter, 3-month-treatment regimens. Using six EORTC QLQ-CIPN20 questions that quantify numbness (N), tingling (T) and shooting/burning pain (P) in upper/lower distal extremities, our aim is to quantify patient-reported responses over 3 months (6 cycles) of oxaliplatin regarding symptom-specific timing, location and severity. For each question, patients were asked how each of the sensory symptoms had affected them during the preceding week, with 1 = “Not at all”, 2 = “A little”, 3 = “Quite a bit” and 4 = “Very much”. The proportional odds model for the cumulative log odds of response that allowed symptom-specific patient heterogeneity to be obtained was applied to a pooled dataset from the placebo arms of two multisite OIPN prevention trials and fit separately to the upper/lower distal extremities. For each symptom, we report the cycle-specific marginal probabilities for each response. In 141 patients, substantial patient heterogeneity in the likelihood, at a given cycle, of a more severe response for a symptom was present. Distinct patterns in the probabilities for each response over time for N and T were observed between the upper/lower distal extremities, while the probabilities of a response >1 for P was largely negligible in both locations. Despite the decrease in exposure to oxaliplatin from 6 to 3 months, OIPN was still pervasive with patients experiencing considerable N and T in the fingers (or hands) and toes (or feet).

E18122 Background: We previously reported that patient subjective perception of their oncology clinical trial experience was influenced by factors beyond their treatment outcome (Chauhan, ASCO, 2012). This study assesses the association between patient perception of clinical trial participation and survival. Methods: This analysis included data from patients enrolled in 31 Phase I and II clinical trials conducted by the North Central Cancer Treatment Group (NCCTG) who completed the Was It Worth It (WIWI) satisfaction assessment one month after trial registration. A landmark analysis was conducted where only patients who were alive at the time of assessment were included in the analysis of overall survival (OS) and only those alive without progressive disease were included in the analysis of progression-free survival (PFS). The time to event for both outcomes starts at the time of assessment. OS and PFS probabilities were estimated using the Kaplan-Meier estimator. The log-rank test was used to compare OS and PFS between patients who were satisfied with their participation vs. those who were not. Results: Among 544 patients who were alive and completed the WIWI assessment at one month after trial registration, 405 (74%) patients reported that their participation in the trial was worthwhile and 139 (26%) answered that participating in the trial was not worth it. There was no difference in OS between the two groups (median 12.5 vs. 11.7 months, p = 0.09). Among 521 patients who were alive without progressive disease and completed the assessment, 392 (75%) reported satisfaction while 129 (25%) reported not satisfied. PFS was more favorable in patients who reported satisfaction with their participation compared to those who did not (median 4.8 vs. 3.9 months, respectively, p = 0.01). Conclusions: Although patient satisfaction did not translate into an overall survival benefit, progression free survival was more favorable in patients who felt their participation was worthwhile. This research supports improving design of future oncology trials by incorporating empathetic and psychosocial support for oncology patients during clinical trial participation. Support: U10CA180821, U10CA180882, UG1CA189823.

Single-item assessments have been the most often-used measures in National Cancer Institute (NCI) cancer control clinical trials, but normative data are not available. Our objective was to examine the normative data and clinically significant effect sizes for single-item numerical linear analogue self-assessment (LASA) scale for overall quality of life (QOL). We analyzed baseline data from 36 clinical trials and 6 observational studies with various populations, including healthy volunteers, cancer trial patients (patients with advanced incurable cancer or patients receiving treatment with curative intent) and hospice patients as well as their caregivers. The overall QOL LASA was rated 0 (as bad as it can be) to 10 (as good as it can be). We calculated the summary statistics and the proportion of patients reporting a clinically meaningful deficit (CMD) of a score equal to 5 or less on the 0–10 scale. In total, for the collective sample of 9,295 individuals, the average overall QOL reported was 7.39 (SD = 2.11) with a markedly skewed distribution with roughly 17% reporting a score of 5 or below indicating a clinically significant deficit in overall QOL. Hospice patients report a much worse average score of 5.7 upon entry to hospice; hospice caregivers average 7.4. Cancer patients vary within these two extremes with most patients averaging in the 7’s on the 0–10 scale (range, 0 to 10 p-value < 0.0001). Men and women’s QOL distributions were virtually identical (with average of 7.6 vs. 7.5, p-value = 0.046). Overall QOL was weakly related to performance status with a Spearman correlation coefficient of −0.29 (p-value < 0.0001). Overall QOL was related to tumor response (p-value = 0.0094), i.e. patients with a full or partial response reported a CMD in 11.4% of cases compared to 14.4% among those with stable disease and 18.5% among those with disease progression. Data missingness was high for performance status and tumor response associations. This study provides the normative data for cancer patients and healthy volunteers for overall QOL using the LASA. These can serve as benchmarks for future studies and inform clinical practice decision-making.

Background: Distress is prevalent among residents and often attributed to rigors of training. Aims: To explore the prevalence of burnout and depression and measured mental quality of life (QOL) among graduating medical students shortly before they began residency. Method: Pooled analysis of data from 1428 fourth year medical students who responded to 1 of 3 multi-institutional studies. Students completed the Maslach Burnout Inventory, PRIME MD, and SF-8 to measure burnout, depression, and low mental QOL (defined as mean mental SF-8 scores ½ a standard deviation below the population norm) and answered demographic items. Results: Shortly before beginning residency, 49% of responding medical students had burnout, 38% endorsed depressive symptoms, and 34% had low mental QOL. While no differences in the prevalence of distress was observed by residency specialty area, there were subtle differences in the manifestation of burnout by specialty. Medical students entering surgical fields had lower mean emotional scores, students entering primary care fields had lower mean depersonalization scores, and students entering non-primary care/non-surgical fields reported the lowest mean personal accomplishment scores (all p ≤ 0.03). Conclusion: Our results indicate a high prevalence of distress among graduating medical students across all specialty disciplines before they even begin residency training.