FormalPara Key Summary Points

This was a qualitative research study with semi-structured, one-on-one interviews with patients with nuclear protein of the testis (NUT) carcinoma and informal caregivers.

NUT carcinoma is a rare and newly established cancer occurring in relatively young patients (median age 16–21.9 years) for which there are no existing standards for treatment; median overall survival is 6.7 months from diagnosis.

Published qualitative literature describing the patient experience of NUT carcinoma is crucial to increase understanding of the condition, particularly given the rapid growth of tumors and associated symptoms and impacts; literature on NUT carcinoma is scarce.

Conceptualizing the patient experience is challenging due to the heterogeneity of the condition; our conceptual framework for NUT carcinoma based on patient interviews describes location-specific symptoms, pain, and systemic effects (e.g., weakness, fatigue, and weight loss).

Locations of NUT carcinoma tumors were variable (lung, head and neck, and other) as were affected symptoms and impacts experienced.

The resulting conceptual framework provides an evidence base for selecting patient-reported outcome measures and designing patient-centric endpoints in future trials to evaluate the benefit of new therapies.

Introduction

NUT (nuclear protein of the testis) carcinoma (NUTca) is a genetically defined epithelial malignant neoplasm hallmarked by chromosomal rearrangement of the NUT gene. Also known as NUT midline carcinoma, NUTca can grow anywhere in the body, but it is mainly present in the midline supradiaphragmatic structures (head, neck, and lung). The sinonasal area is the most common midline tumor site. The true incidence and prevalence of NUTca is unknown, as the disease remains undiagnosed or misdiagnosed given its rarity. Only a minority of cases are diagnosed with NUTca at the beginning of the disease, while the most common incorrect diagnoses are “poorly differentiated carcinoma” or “poorly differentiated squamous carcinoma.” However, given the increased awareness and availability of easily applicable diagnostic tests, NUTca has been detected more frequently in the past decade, most specifically in adults [1,2,3]. In the small samples of patients studied, the median age at diagnosis ranges from 16 [2] to 21.9 years [3], but NUTca has been reported in patients aged 0.1–81.7 years [2,3,4].

NUTca runs a devastating clinical course. It usually presents with rapidly enlarging masses, characterized at advanced stages by early metastatic spread to either locoregional lymph nodes or less common, distant sites. Consequently, most patients present with mass-related symptoms (such as rhinorrhea, epistaxis, nasal obstruction, proptosis, diminished vision, dysphagia, persistent cough, shortness of breath, or pain), while nonspecific symptoms such as fever and weight loss have been seen only occasionally. In a cohort of 54 patients, the median overall survival was 6.7 months, with only 19% (CI 7–31%) of patients alive after 2 years [2].

Currently there is no standard of care for NUTca. Patients may undergo surgery, radiation therapy, and/or chemotherapy, although there is no evidence of the effectiveness of these treatments. These treatments have symptomatic adverse events of varying severity and impact on quality of life.

While increased survival is the primary goal, the symptomatic experience of patients and their functional status are also essential to inform our understanding of the benefit–risk profiles of therapies in development and address a high unmet need to better care for this vulnerable patient population. This study is, to our knowledge, the first attempt to conduct research directly with NUTca patients and their caregivers. No qualitative research manuscripts describing the experience of patients with NUTca were retrieved, and limited information was found on Facebook support groups or medical association websites. Additionally, there were no specific patient-reported outcome (PRO) measures identified in the literature. To provide this missing evidence, we conducted this qualitative research to describe the experience of patients with NUTca and propose a CF. The objective was to provide a first platform for an evidence base to inform the selection of outcome measures appropriate to assess the experience of patients and ensure that endpoint design in future trials is patient-centric, including methods employed in other rare disease areas as well [5].

Methods

Study Design and Procedure

As indicated by the guidance from the United States Food and Drug Administration (FDA), the selection or development of a PRO measure should be supported by direct experience from patients [6]. We conducted semi-structured interviews using open-ended questions that allowed participants to spontaneously report their experience and that promoted discussion rather than following a prescriptive script with participants. We did not define a purposive sample with specific recruitment targets covering a range of tumor locations due to the anticipated recruitment challenges in this rare and aggressive cancer. Instead, we adopted a convenience sampling approach (using a volunteer-led online support group) while monitoring the variety of tumor locations in the sample interviewed. Given the nature of patient-centered research in rare disease [7], we also decided to interview caregivers of patients who were too sick to participate or had died, to collect their observation of the patient experience.

To participate, patients had to be 12 years old or older with a diagnosis of NUTca (either currently or in remission); caregivers had to be 18 years old or older and had to be (or have been at some point previously) a caregiver to a patient (any age) with NUTca. Both patients and caregivers had to be able to participate in one or two 1-h phone interview(s); be able to speak, read, write, and understand English; and reside in the United States. Patients treated for a cancer other than NUTca and patients or caregivers having any visual, auditory, cognitive, or language impairment that would prevent reading, understanding, or answering interview questions were excluded. Our goal was to interview n = 10 patients and n = 20 caregivers.

The semi-structured interviews were recorded and transcribed verbatim, with each lasting approximately 60 min. Patients and caregivers were interviewed individually, even if patients were under 18 years of age, assuming a parent or guardian was able to provide consent. There were separate patient and caregiver interview guides; both asked about the diagnostic process, symptoms experienced, and impacts of NUTca on various aspects of the patient’s daily life. The patient interview guide specifically had additional probes for in-depth descriptions of the way symptoms were experienced and about specific impacts (e.g., whether NUTca affected the way the patient thinks or communicates or affected their participation in daily activities). The caregiver interview guide had additional prompts that related to the impacts of caregiving (not reported in this manuscript). Demographic and health information was collected, including tumor location and treatments received.

Study documents were approved by a central institutional review board (Copernicus IRB tracking 20200591), and all procedures were conducted in accordance with ethical guidelines. Consent for publication was obtained. All guidelines of the Declaration of Helsinki of 1964 and later amendments (relevant to this non-interventional, qualitative study) were followed. Patients were recruited through a NUTca Facebook support group in partnership with the moderator of the group. Interested participants completed an online screener and informed consent/consent for publication form and, if eligible, were called and scheduled for an interview. Participants received $85 for the interview.

All interviews were conducted over the phone. Interviewers were trained in Good Clinical Practice (GCP) guidelines and had substantial experience (2+ years) conducting qualitative interviews with patients and caregivers. Additionally, all interviewers attended a mock interview to become familiar with the interview guide.

Coding and Analysis

De-identified transcripts were inductively coded and thematically analyzed [8,9,10,10]. Three researchers (AC, AL, MC) independently coded the first transcript before meeting to harmonize and create a codebook and format used for the remaining transcripts. To determine the adequacy of the sample size of participants interviewed, we performed a saturation analysis using chronologically ordered groups of five transcripts; new concepts (defined as analytical categories comprised of related codes supported by quotations) were tabulated in columns per group. Theoretical saturation had been reached and the sample size for the interviews was adequate if no new concepts emerged in the last group of five (i.e., low likelihood of new patient experience information being collected with additional interviews).

To develop the conceptual framework, codes and quotations were inductively categorized into higher-order, overarching categories referred to as concepts, sub-domains, and domains reflecting their conceptual underpinning. This involved an iterative process of cross-referencing and comparison between the different analytical categories (concepts, sub-domains, and domains) [9, 10, 11].

Results

Description of the Patients

We met our intended recruitment quota for patients with n = 10 interviews (n = 4 patients were patient/caregiver dyads). On average patients were approximately 40 years old and half were female (n = 5/10); most patients interviewed were Caucasian (n = 9/10) and married (n = 7/10), with a range of education and work statuses. None of the patients interviewed lived alone. About half of the patients (n = 6) were diagnosed with NUTca over 2 years prior to their interview; most patients reported receiving chemotherapy (n = 9/10), radiation (n = 7/10), and surgery (n = 8/10) as treatments for their NUTca tumors. Tumor location was primarily in the head/neck (Table 1).

Table 1 Characteristics of patient sample—patients (n = 10)
Full size table

Description of the Caregivers

We interviewed n = 17 (/20 planned) caregivers (among them n = 4 patient/caregiver dyads).

On average, caregivers were 45 years old, mostly female (n = 15/17), and white, non-Hispanic (n = 16/17). About half of the patients of caregivers were diagnosed with NUTca more than 2 years prior to the interview (n = 8/17; includes if patients were deceased at time of interview). Like the patient sample, most patients of caregivers had received chemotherapy (n = 13/17), radiation (n = 12/17), and surgery (n = 9/17); discordantly, most patients of caregivers had lung tumors (n = 9/17) versus head/neck; there were a higher number of total tumors reported (Table 2).

Table 2 Characteristics of patient sample—caregivers (n = 17)
Full size table

Codebook

The final codebook resulted in 270 codes describing signs and symptoms of NUTca and 230 codes describing impacts of NUTca. Codes were granular and provided nuances in the experience of associated concepts. The codes were also linked directly to the patient’s words, with as little researcher inference as possible from quote to code. Researchers, for example, coded “I feel nauseous every time I eat” as “nausea: when eating” but other codes for nausea included “nausea: new onset” or “nausea: worsens during day.” During analysis, codes were categorized into higher-level concepts (in the example provided, all grouped as “nausea”).

Codes that did not directly contribute to informing the conceptual framework of patient disease symptoms and impacts on daily life have not been retained [e.g., aspects of diagnostic history (misdiagnosed, biopsies) or burden of caregiving]. For symptomatic adverse events, we did not assume a patient’s symptoms were treatment-related unless specifically mentioned by the patient. Given that caregivers were reporting as proxies for patient experience, we did not distinguish between caregiver and patient codes in the analysis.

Symptoms Reported

Systemic (fatigue, weakness) and local symptoms were reported, dependent on tumor location: lung, head/neck, and other (included spine/back, shoulder/scapula, ankle, femur, pelvis, hip, lymph nodes, liver, arm, testicle). Severe pain was reported for all locations. Digestive symptoms were reported as well without clear association to a specific tumor location and possibly due to symptomatic adverse events (AEs) from treatment based on clinician opinion. Since digestive symptoms were not mentioned by patients as resulting from treatment, we retained them for completeness (Table 3).

Table 3 Symptom domains, concepts, and exemplary quotations
Full size table

Impacts Reported

Patients reported a range of impacts on different aspects of their life: daily activities, sleep, roles, social life, and finances. Patients reported that their experience and satisfaction with and access to care were important when considering their overall experience of NUTca (not included in the framework) (Table 4).

Table 4 Impact domains, concepts, and exemplary quotations
Full size table

Saturation of Concepts

Most of the concepts arose in the first 20 transcripts; in the fifth group of five transcripts, the only concepts reported related to hearing problems, forgetfulness, slow processing, and difficulty urinating. No new concepts emerged in the last two transcripts. These results indicated that conceptual saturation was acceptable in the sample interviewed and that the sample size was sufficient despite the heterogeneous nature of the condition.

Conceptual Framework

The CF was organized into four symptom domains (systemic, location-specific, pain, and digestive) and six impact domains (daily activities, emotions, sleep, social, role, and financial). Some social life impact concepts (isolation, not seeing friends, and limited social gatherings) may be related to AEs of treatment (Figs. 1, 2).

Fig. 1
figure 1

Symptom conceptual framework

Full size image
Fig. 2
figure 2

Impact conceptual framework

Full size image

Discussion

Most patients interviewed had tumors in the head, neck, or lungs, which based on the epidemiology of NUTca are the most frequently reported tumor sites [12]. Importantly, the few concepts on the experience of patients retrieved from the review of the limited available literature and clinical sources were expanded by the patient interviews. The comprehensiveness of the findings was difficult to evaluate since NUTca has many possible tumor site locations, presents differently depending on tumor location, and spans across age groups; given the rarity of NUTca, purposive sampling methods to account for a full range of experiences was difficult. However, given the acceptable saturation of concepts in the sample interviewed, the sample size was deemed sufficient to draw conclusions on the experience of patients with the most frequent tumor locations (head/neck and lung).

Our choice to use caregivers as proxies for patients was justified by anticipated recruitment challenges and feasibility of interviewing patients with hard-to-reach, rapidly progressing, rare carcinoma; this methodology serves as an example of the needed pragmatism and practical considerations of conducting research in similar populations. We found that the caregivers were extremely involved in the lives of patients (or previously were if patients were deceased) and provided useful information on the experience of patients. We did not analyze patients and caregivers separately due to the small sample size but acknowledge the potential differences between caregiver reporting and patient reporting. Over half (6/10) of the patients interviewed can be considered long-term survivors in comparison to the medium short life expectancy in NUTca. Due to the small sample size, no analyses comparing subgroups of patients according to the time since diagnosis were performed. This may call into question the generalizability of our findings, but in rare conditions, research design and analysis must be pragmatic in order to accumulate evidence to serve the research objectives [13].

One of the biggest challenges was creating a CF for a disease with such heterogeneity of reported concepts given the varying tumor locations; our goal was to create a CF that captured the patient experience of NUTca in a comprehensive way; this will allow researchers designing trials for NUTca to ensure that outcome measures are reflective of the experience of patients and that treatment benefit can be meaningfully interpreted in the context of this experience. We chose to separate the location-specific symptoms and impacts in order to create a CF that could potentially be updated with new tumor locations as add-ons, without questioning its structure. This CF can serve as a foundation and potentially maximize efficiency for future clinical research as further information is gathered on the experience of NUTca. As more clinical and patient experience data become available, the typicality of some concepts included in the current framework for comprehensiveness can be reviewed.

Pain emerged as a unique domain because of its importance in the experience of patients. Pain was reported for many tumor locations, but also as “radiating” to other parts of the body, making it sometimes difficult to pinpoint. To reflect the experience of patients and ensure its proper assessment, pain was proposed as a separate domain relevant across locations rather than being integrated with location-specific symptom domains. Further information on the patient experience of pain is warranted.

Conclusions

This qualitative research generated valuable information on the symptoms and impacts on daily life experienced by patients with the most frequent locations of NUTca. The resulting CF provides an evidence base for selecting patient-reported outcome measures to evaluate the benefit of new therapies and for designing meaningful endpoints for future clinical trials. This research illustrates how in rapidly progressing disorders such as NUTCa with diverse locations and manifestations, a patient-centric approach can be used to inform decisions [7]. We structured the framework to account for location-specific symptoms and to allow for the addition of concepts relevant to the patient experience of NUTca that were not captured based on the patient sample interviewed in this research.