Evaluating fertility preservation interventions for alignment with ASCO Guidelines for reproductive aged women undergoing cancer treatment: a systematic review

In this review, we define interventions to include those testing strategies to increase consideration of and access to fertility preservation. Further, only articles in which interventions were evaluated relative to a comparison group are included.

In conducting the review, we followed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRSMA) guidelines [18]. We completed systematic literature searches in four electronic databases (PubMed, Embase, Web of Science, Scopus) for intervention studies published between January 2006 (to align with publication of first ASCO guidelines related to fertility preservation [13] to January 2022. Keywords included: cancer, oncology, fertility preservation, oncofertility, intervention, healthcare utilization, and healthcare delivery. Full search terms can be seen in Appendix 1. We also hand searched any protocol papers to identify new articles subsequently published that were not captured in the initial search. We used Covidence to organize and manage the review database.

The intervention study inclusion criteria were: English language, published between 2006 and 2022, that described an intervention including women of reproductive age (ages 13–49) with a cancer diagnosis and/or the providers that care for them. Recent data show the average age for the onset of menarche is at 12.4 years old. Thus, we selected 13 years old as the lower age limit to identify women of “reproductive age” [19]. Additionally, given that fertility preservation options vary by sex and age, we focused on reproductive age women. Excluded intervention studies (henceforth, “interventions”) were those that only included males or pediatric patients, were not accessible in full text, conference abstracts, study protocols, and reviews.

Using these criteria, we conducted initial title searches to identify articles for initial abstract review. Keywords were clarified to ensure that all relevant articles appeared in the search. Next, one author (SP) reviewed all titles and abstracts to identify articles for full-text review. Finally, we completed full-text reviews to determine the interventions to be included. Two authors (SP, CMM) completed data abstraction on all full-text review articles. All discrepancies were resolved through discussion to reach consensus.

We applied the CONSORT [20] and TREND [21] statement checklists to assess intervention quality and the extent to which articles aligned with quality and reporting guidelines. If articles mentioned any of the concepts included on our checklist, we considered the criteria fulfilled. All articles were then scored for fulfilled criteria on a scale ranging from 0 (lowest quality)–11 (highest quality). Abstracted data were organized based on assigned quality score from highest to lowest.

We used the population, intervention, comparator, outcomes, timeframe, and study design (PICOTS) framework [22] to select which intervention characteristics were extracted. General intervention characteristics included: intervention study design, sample targeted, unit of analysis, intervention time period, in what country or geographic region the intervention took place, intervention setting (e.g., oncology clinic, fertility clinic), sample size, and general sample characteristics. In this review, healthcare providers included oncologists, reproductive specialists, surgeons, and nurses.

For each intervention, we determined the alignment of intervention targets and relevant outcomes with the ASCO guidelines for fertility preservation prior to cancer treatment. A “yes” was assigned for the following guideline elements were targeted as an intervention strategy or assessed as an outcome: (1) provider-led discussion of potential impairment to fertility, (2) provider-led discussion of fertility preservation approaches, (3) timing of discussions occurring prior to beginning treatment, (4) patient was referred to reproductive specialist for fertility preservation, (5) fertility-related discussion was documented, and (6) patient was referred to psychosocial services for additional support [13].

We coded interventions for their alignment with a parsimony-based operational framework based on complexity theory and pragmatic trials [23]. The framework considers four characteristics: socioecological levels of influence targeted, conceptual clarity, methodologic pragmatism, and sustainability. We noted the inter-level mechanisms implied at each level of influence that were targeted by the fertility preservation intervention and whether this was based on stakeholder input. For methodological pragmatism, we evaluated whether interventions took place in real world contexts and what background factors were considered in the evaluation of the intervention to enhance the generalizability of effectiveness findings. We also examined whether the interventions had any discussion or consideration of sustainability factors. Lastly, we noted whether the intervention outcome assessment showed significant benefit of the intervention for the proscribed outcomes.

Extracted data were organized by quality assessments for: general intervention characteristics, alignment of intervention strategies and outcomes with ASCO guidelines and consideration of multiple levels of influence. Within each category, we rated intervention quality from 0 to 11.

Among the nine included interventions, the highest quality score assigned was 9 (n=4 interventions) and the lowest was 5 (n=1). All interventions reported inclusion criteria, described group assignment, reported baseline and group comparisons, and reported measured outcomes. Most interventions detailed sampling methods (n=7) [24‐31], provided sufficient detail for replicability based on CONSORT/TREND requirements (n=7) [24‐31], and reported on subgroup analyses (n=5) [25, 26, 28‐30, 32, 33]. Few interventions explained power considerations (n=4) [25‐27, 29, 32, 33] and reasons for missingness in addition to how missing values were handled (n=2) [27, 28]. No intervention fulfilled all 11 criteria, and none reported a specific theory that guided the intervention. See Table 1.

Descriptions of the interventions are detailed in Table 2. Most interventions utilized a quasi-experimental design (n=7) [24, 27, 28, 30‐34] without randomization to a comparison group . Two interventions were randomized control trials [25, 26, 29]. The majority of the interventions focused on educational components to increase patient and/or provider knowledge regarding fertility preservation options and to assist in discussions and the decision-making process. This primarily targeted the individual and interpersonal levels of influence. Two interventions utilized decision aids to assist patients further in their ability to have discussions around fertility preservation options and to prevent decisional regret in the future [25‐27]. Two interventions also leveraged the support of a nurse/patient navigator [24, 31]. Three took place in the USA [24, 28, 29]. Other settings included: Austria, Australia, Canada, the Czech Republic, Germany, Poland, and Switzerland. The majority of interventions were implemented in specialty cancer centers or healthcare practices, with one in regional fertility centers [25, 26]. Two interventions included patients as young as 12 [24, 32, 33]. Three interventions included males in the target sample [28, 30, 32, 33], one included patients’ parents [32, 33], and one specifically targeted breast surgeons [34]. Interventions typically did not focus on a particular cancer type, but when they did, breast cancer was most commonly included (n=6) [24‐27, 29, 31, 34]. Five interventions did not report race/ethnicity of the sample [27, 30‐34]; among those that reported the majority of participants were White. Most participants were in a relationship, had a college education, and were childless or had one child.

We assessed the extent to which the intervention strategies and relevant outcomes aligned with ASCO guidelines’ six expectations of fertility preservation service provision (See Table 3).

Guideline Element 1: Health care providers should initiate discussion of the potential impairment to fertility. The majority of the interventions (n=8) included strategies to assist providers in initiating discussion of potential treatment effects on fertility and measured the occurrence of these discussions as an outcome [24, 27‐34]. To facilitate these discussions, one intervention included a 24-h fertility preservation hotline available to providers [24]. Additionally, the electronic medical record (EMR) was modified to prompt oncologists initiate discussions with their patients and required them to document discussion before the patient chart could be closed [24].

Guideline Element 2: Providers discuss fertility preservation with patients. Over half (n=5) of the interventions noted discussions regarding fertility preservation approaches [24‐29], with one targeting discussion as an intervention strategy without measuring it as an outcome [25, 26]. The two most common intervention strategies to guide these discussions was providing patients with educational materials on fertility preservation [25‐27] and providing patients with decision aids to review the options [24, 28, 29].

Guideline Element 3: Provide fertility preservation services as early as possible prior to the patient beginning treatment. Only four interventions specified that discussions occurred prior to treatment, and all four mentioned this as part of an intervention strategy but not as an outcome [25, 26, 28, 29, 31]. None of the interventions actually measured when the treatment discussions were occurring or provided description of what was done to promote early discussions.

Guideline Element 4: Discussions of fertility effects and preservation be documented. Four interventions noted that fertility-related discussions were documented [24, 29‐31], however one did not measure documentation as an outcome [24]. In one intervention, a nurse navigator was utilized to screen referrals to the cancer center, expedite tests and consultations, and provide ongoing support to patients. The navigator documented all activities and services in the nursing section of the EMR [31].

Guideline Element 5: Referral to fertility preservation services. In the majority of interventions, providers referred patients to reproductive specialists either to discuss fertility preservation options or to receive fertility preservation treatment (n=8) [24‐31, 34].

Guideline Element 6: Refer patients for psychosocial services. This was noted in only two interventions [29, 31] with only one including referral as a measurable outcome [29]. As secondary outcomes, Partridge AH, et al (2019) examined attention to psychosocial concerns including noting emotional health, distress, referral for psychosocial support, etc. [29].

All interventions except one [27] were multilevel, with three targeting two levels (individual and interpersonal) [25, 29, 32, 33] and five targeting three levels (individual, interpersonal, and organizational/system/regional) [24, 28, 30, 31, 34]. Despite targeting multiple levels, all intervention effectiveness testing was conducted at the individual level. Primary outcomes included decisional conflict [25‐27], fertility-related discussion and documentation of discussions [24, 29‐31, 34], patient knowledge [32, 33], and patient satisfaction with intervention materials [28]. Eight interventions reported a significant effect for the primary outcome of intervention [24‐28, 30‐34].

When considering inter-level mechanisms for conceptual clarity, approximately half of the interventions included stakeholders’ views on the assumed process chains being targeted (n=4) [27, 28, 30, 31, 34]. Examples of stakeholder inclusion were utilizing multidisciplinary expertise to develop and pilot survey instruments, conducting focus groups with the patient population prior to the intervention, pilot testing decision aids, conducting a provider needs assessment to develop education tools and intervention materials, and establishing an interdisciplinary steering committee to guide intervention development and implementation.

When considering methodologic pragmatism, all interventions were implemented in real world contexts as they took place in cancer centers, healthcare practices, and fertility clinics. Across the targeted levels of influence, seven interventions considered the sociodemographic characteristics of the patients as background factors [24‐29, 31, 34], one intervention considered literacy level differences by region [30], with only one intervention making no mention of background factors [32, 33].

Only three of the 11 interventions discussed sustainability [25‐28]. Examples of sustainability included easily updating and/or transitioning to online decision aid materials and considering strategies to increase physician use of resources. Two interventions explored ideas for how they could make the intervention more sustainable in the future [24, 31]. All components of the multilevel framework are outlined in Table 4.