Improving the process of ordering outside genomic testing for lung cancer FNA and small biopsy specimens – A multidisciplinary quality improvement project

INTRODUCTION

Lung cancer remains the most common cause of cancer death in the United States and worldwide. Molecular characterization of lung tumors has rapidly added new, increasingly effective and less toxic treatments to the lung cancer armamentarium, even for patients who are poor candidates for traditional chemotherapy.^[1] Thus, timely identification of genomic characteristics not only an urgent issue, but occasionally a mortal one.^[2]

At Mayo Clinic in Arizona, the molecular characterization of lung cancer specimens is performed by an outside vendor (Caris Life Sciences, Phoenix, AZ). Providers at every step in the process from lung cancer diagnosis to the interpretation of genomic data for treatment decisions have raised concerns about occasional gaps, delays, and miscommunication. For this reason, a quality improvement project was undertaken to identify the current state of the process, to identify gaps and opportunities for consolidation, clarification, and streamlining, and eventually, to implement an improved process.

Molecular characterization of lung cancer specimens is essential to determining prognosis and treatment. Timely identification of genomic characteristics of tumor cells is vital for rapid initiation of targeted therapy. Appropriate targeted therapies are more effective and less toxic than broad-spectrum chemotherapy and produce a progression-free survival advantage.^[2] The current process for ordering these crucial studies has often led to delays, confusion, and missed opportunities to initiate effective therapy, sometimes resulting in adverse patient outcomes. These delays create a substantial gap in quality of care provided to our patients. In particular, we believe that this gap in quality is a reflection of underuse. There is a failure to provide timely tissue testing results when it may have produced a favorable rapid and targeted treatment for a patient. Moreover, there is no consistent and standardized process to submit specimens for outside genomic testing.^[3]

The College of American Pathologists (CAP), International Association for the Study of Lung Cancer (IASLC), and Association for Molecular Pathology (AMP) joint consensus guidelines state that the average turnaround time (TAT) from biopsy to genomic reporting be no more than 10 days.^[4] Lung cancer genetic testing has a turn-around time of 6 weeks on average at Mayo Clinic in Arizona, representing a substantial, yet modifiable gap in quality of care.^[4]

Achieving a target TAT of 10 days is expected to yield an important impact to Mayo Clinic and its patients. Roughly 10% of patients with non-small cell lung cancer (NSCLC) have an actionable mutation identifiable through genomic testing, and potentially, by improving access to trial therapy, we would improve patient survival.

MATERIAL AND METHODS

To achieve the goal of reducing TAT, the team leveraged stakeholder input to fuse practice knowledge from a variety of specialties, assessment of the workflow from patient encounter to tissue acquisition to molecular target reporting.

The team identified stakeholders of the process and gathered their input to define the problem through survey, data collection, and observation of the current state. Feedback from the stakeholders revealed that there was uncertainty about ownership of responsibility for sending order requests, obtaining molecular testing results, as well as relaying those results once available.

The team addressed stakeholders concerns by following the Define, Measure, Analyze, Improve, and Control methodology to standardize the process for NSCLC tissue acquisition, genetic testing, Caris reporting, and TAT for patients seen at Mayo Clinic Arizona. Tissue samples included for this study were endobronchial ultrasound fine-needle aspirate sample with cell blocks, surgical resections, and navigational bronchoscopic small biopsy specimens. Direct lift-offs from cytology fine-needle aspiration slides were not accepted by Caris.

We convened a multidisciplinary team comprising stakeholders from the fields of oncology, pulmonology, radiology, pathology, thoracic surgery, and quality improvement. After providing perspectives on the sources of and possible solutions to the issues identified, one team member (KMM) interviewed physicians, advanced practitioners, fellows, residents, administrators, and laboratory professionals involved with every stage of the process and assembled a flow diagram outlining the current process and identifying potential gaps and opportunities for error. The team then reconvened to provide additional feedback about the flow diagram, which will now be used to guide plan, do, study, and act (PDSA) cycles targeted to each of the identified gaps.

RESULTS

Baseline measurements

Thus far, in collaborating with pathology stakeholders, the overall QNS rate is between 2% and 7% since implementing our streamlined process in March of 2018. The stability of the QNS rate is important given the increase in Caris cases.

The gap in quality of care is that the TAT for molecular testing is well beyond the standard of care set forth by CAP, IASLC, and AMP. The factors contributing to this gap include the manual paper ordering process for genomic testing, transportation of specimen and paper orders to another Mayo pathology location, and checkout process of specimen within pathology at Mayo Clinic in Arizona. Additional uncertainty exists as to where the responsibility lies both in ensuring that the molecular testing is performed and that the results are obtained and relayed in a timely fashion [Figure 1].

Close

Figure 1:

Process map for tissue turnaround time for lung cancer specimens.

Export to PPT

Process changes based on this feedback

1. Pathology workflow process standardized

2. Pathology work queue standardized operating procedure document and workflow generated

3. Standardized workflow for ordering outside molecular assays created and implemented.

Mayo Clinic in Arizona’s baseline measure of TAT from sample obtainment for NSCLC to molecular testing reporting was 6 weeks in October 2017 [Figure 2]. Once specimens were received by Caris, the external TAT was 17.5 days for a sample size of 29 patients, and this monthly sample size was consistent throughout the fourth quarter of 2017. The most recent sampling was updated to show sustained effect in June 2019 [Figure 3]. The data source for TAT was the pathology dashboard, a detailed process flow, as well as interviews with the stakeholders noted above. In late 2017, the number of cases sent to Caris steadily increased, but the Caris external TAT remained steady while the total TAT continued to decrease based on process improvements.

Close

Figure 2:

Baseline pathology dashboard assessment of turnaround time.

Export to PPT

Close

Figure 3:

Follow-up pathology dashboard assessment of turnaround time.

Export to PPT

Post-interventions have resulted in a reduction of total TAT for sample resulting from 6 weeks on average to 4 weeks. Moreover, we were able to develop and implement new Thoracic Oncology NSCLC Molecular Profiling Procedure as of May 2018. Reports from pathology note improved TAT already without increase in QNS. Thus, it is anticipated that we will achieve and exceed our goal of improving TAT without QNS in the coming months, leading way to further PDSA cycles and process improvement.

DISCUSSION

We reduced TAT by ~33% without increasing QNS. Through the identification and proactive engagement of multidisciplinary providers throughout entire project and the engagement and active participation of a project champion (key stakeholder and critic), we were able to achieve the goals noted above and expect this reduction in TAT to continue toward CAP standards. Importantly, manual processes are inherently time intensive, and obtaining agreement and engagement of all stakeholders can also be encumbering. The barriers to implementation that we experienced included: (1) Achieving timely consensus from stakeholders involved in the process, (2) needing multiple meetings to come to agreement on roles and workflow modifications within pulmonary and pathology departments, and (3) coordinating multiple provider schedules to achieve project consensus.

To continue our current trajectory of success, we have implemented a communication plan, including a detailed email of molecular testing workflow that was sent from our pathology manager to pathology department and pulmonary department with commentary as well as an updated workflow process. This new workflow has also been the subject of attention in pathology and pulmonology department staff meetings. Our transition plan for the future includes: (1) Pathology owns the standard operating procedure and associated forms. (2) The supervisor for pathology support, laboratory medicine, and pathology is the owner. (3) Point persons in pulmonology (RCC) and pathology (LC) have been assigned should any confusion in the workflow process arise. Items for the future study include: Implementation of clinical decision support, educational sessions including grand rounds dedicated to molecular testing standards and our updated process, as well as integration/standardization of molecular testing processes from an enterprise perspective.

Limitations: Since many interventions were implemented simultaneously, it is difficult to surmise which of these interventions had the greatest impact. However, it is clear that having defined roles within a team as well as standardized processes for coordinating tissue processing between teams and with outside laboratories is crucial for optimizing TAT. We suspect that the scenario presented here is common and that standardized approaches to tissue processing such as the ones presented here have the potential to make a meaningful impact on patient care within a fairly short period of time. Stakeholder assessments are also key to identify limitations unique to a given institution’s culture, workflow, and role perceptions. Thus, despite our study’s noted limitations, we believe these findings to be highly generalizable.

SUMMARY

Our project highlights process limitations for lung cancer testing that substantial impact patient care. Using a multidisciplinary team, lean principles, and key stakeholders, we were able to set forth a reproducible methodology for improving TAT in busy, tertiary referral centers working without outside testing facilities. We believe that our experience may be instructive to other institutions implementing large-scale quality improvement projects.