Uniformity in counting prior anti-cancer lines of therapy (LoT) in patients with solid tumors

Published in Cancer
Uniformity in counting prior anti-cancer lines of therapy (LoT) in patients with solid tumors
Like

Kamal S. Saini1,2, Chris Twelves3

1 Covance Inc., Princeton, NJ, USA; 2 East Suffolk and North Essex NHS Foundation Trust, Ipswich, UK; and 3 University of Leeds and Leeds Teaching Hospitals Trust, Leeds, UK

Correspondence: Kamal S. Saini (Kamalveer.saini@covance.com, @KSainiMD), or Chris Twelves (C.J.Twelves@leeds.ac.uk)

Whenever I prescribe systemic anti-cancer therapy (SACT) on my hospital electronic prescribing system, I have to click on “Line of therapy” and choose a number; while this may sound like a simple question, I’m never sure what the right answer is! I treat people with breast cancer, so in routine care should I count hormonal treatments as lines of therapy (LoT) or just chemotherapies? In the neo-adjuvant setting, if we make a planned change of regimen is that still a single LoT? Do I include treatments patients received as adjuvant therapy when I prescribe for someone with metastatic disease? If a patient with metastatic disease is responding to treatment has break off chemotherapy, then re-starts that same chemotherapy some months later, is that part of the same LoT?

This question of what constitutes a LoT also comes up when deciding whether a patient is eligible for a clinical trial, most of which specify how many LoTs a patient can have received to be eligible. This is made more difficult because the “rules” as to what constitutes a LoT vary from protocol to protocol and between different types of cancer.

This question of defining a LoT causes frustration and doubt not only in members of the clinical teams, but also in research nurses, junior staff, medical monitors, auditors, regulators and sponsors. A key reason for this is a lack of guidelines and uniform definitions to help us determine when one LoT ends and another begins. This makes the counting of LoT time-consuming, resource-intensive and error-prone, hampering both clinical research and patient care.

Having spent countless hours performing this thankless task of determining LoT in various capacities, as clinicians and researchers, the authors came to appreciate the full extent of the problem and the multiple nuances of defining a LoT. Further, we also discovered that this was a problem that no one had yet attempted to address comprehensively.

In the absence of any peer-reviewed guidance on this topic, we set out to devise a comprehensive solution of our own, focusing on SACT.

Our guiding principles were the following:

  1. Universal
    • Like RECIST, ECOG PS, and TNM staging, our solution should be applicable to all solid cancers
    • Specifically, we must avoid having separate guidelines for separate anatomical cancers
  2. Minimal data and burden
    • Our solution should impose as little administrative burden on treating team as possible
    • Amount of additional data needed to accurately determine LoT should be minimum
  3. Deal with ambiguity
    • It should be applicable to the majority of the wide variety and diversity of clinical scenarios
    • Applicable to both clinical research and day-to-day patient care

As we set out to design our framework, we encountered several fundamental complexities, which conceptually could be categorized into two broad themes: The “What” problems and the “How” problems.

There are no easy or obvious answers to the question “What prior therapies should be assigned a LoT.

  • Extent of disease: Everyone would agree that systemic therapy for metastatic disease should be assigned a LoT. But what about systemic therapy in the (neo)adjuvant setting or given for locally advanced cancer? What about chemotherapy for local recurrence? Where do we draw the line between treatments? Is it always possible to draw a line between treatments?
  • Route of administration: Similarly, it is clear that systemic (oral and intravenous) anti-cancer agents should be assigned a LoT. But what about anti-cancer agents that are given locally or regionally but clearly have systemic effects (e.g., intra-muscular fulvestrant) or the potential to have systemic exposure and/or effect (e.g., intra-peritoneal chemotherapy or intra-tumoral immunotherapy)? Or regional anti-cancer therapy that is given to achieve local disease control with the express intention of avoiding systemic exposure (e.g., intra-arterial administration to treat liver metastases, or isolated limb perfusion for some sarcomas)? Should we differentiate depending on the intent of systemic reach and effect, versus the intent of loco-regional reach and effect?
  • Anti-cancer modality: It seems clear that systemic therapy with agents such as chemotherapy, targeted agents, endocrine therapy, and immunotherapy should be assigned a LoT. But what about “systemic radiotherapy”, using radiopharmaceutical agents administered intravenously (e.g., radiolabeled antibodies), orally (e.g., radioiodine), or intra-arterially (e.g., as 90Y microspheres)? We decided to presently assign LoT only to SACT; thus radiation (or other local treatments including surgery), would not constitute a LoT. But what about chemo-radiotherapy where a SACT is given specifically as a radio-sensitizer i.e. for a local effect?

Once we determined and defined what treatments needed to be assigned a LoT, there still remained the question of “How to best determine and express LoTs, keeping in view the enormous complexity and ambiguity. We had to wrestle with several thorny issues while developing the overall framework of our solution, trying to iron out the inherent imprecision of the current paradigm, yet respecting deeply entrenched and familiar conventions as far as possible.

  • Vague vocabulary. Words like neoadjuvant, adjuvant, induction therapy, second adjuvant, maintenance, regimen, presurgical chemotherapy, etc. are used subjectively and often based on individual interpretation. Additionally, an anatomy-specific terminology has emerged; for example, chemotherapy given to a patient with breast cancer and liver metastasis would certainly be considered metastatic, whereas chemotherapy given to a patient with a liver metastasis from a colonic primary may well be labeled as “neoadjuvant” in certain clinical scenarios. Our solution had to encompass these disparate realities.
  • Complex treatment regimens: Patients with some tumour types may continue the same treatment beyond progression (e.g., HER2-directed therapy for breast cancer), or receive the same drug multiple times over the course of their cancer journey despite previous progression on that therapy (e.g., platinum therapy for ovarian cancer). Other patients complete, and respond to, a line of SACT then receive “consolidation” therapy with a different drug (e.g., patients with breast cancer who respond to a course of chemotherapy and start endocrine therapy whilst still responding). Our solution had to accommodate these complex realities within a relatively small set of common-sense based rules and guidelines.
  • Defining disease progression: Most oncologists do not routinely perform RECIST for all their patients. They take into consideration clinical findings, imaging, and laboratory test results to generate their own “real-world” definitions of when a patient has had clinically significant progression of disease. We wanted to make our solution applicable to both routine cancer care and to clinical trials, hence had to develop a pragmatic definition that would apply to both situations.
  • Clinical trial eligibility: Clinical trials try to recruit well-defined patient populations with specific characteristics required to answer the scientific problem they are addressing. One key characteristic often used is to define the prior anti-cancer therapies (dis)allowed by cancer trial protocols. Protocols often use two related yet distinct parameters to describe prior (dis)allowed anti-cancer therapies. The first is the disease setting (i.e., early, locally advanced, and metastatic) in which the therapy was administered, and the second is the therapeutic intent (i.e., curative or palliative).
  • While developing our framework, we had animated discussions if we should name the LoTs based on the treatment intent (see column 3 of Table below), i.e., curative (CLoT) and non-curative (N-CLoT) or palliative (PLoT), or based on the clinical setting (see Column 2 of Table), i.e., early (ELoT), locally advanced (LALoT), and metastatic (MLoT).
  • In order to avoid too much complexity, we eventually chose to have a blended approach incorporating both intent and setting. We proposed reporting LoT in a novel and standardised format as LoT N (CLoT + PLoT), where CLoT is the number of systemic anti-cancer therapies (SACT) administered with curative intent and/or in the early setting, PLoT is the number of SACT given with palliative intent and/or in the advanced setting, and N is the sum of CLoT and PLoT. Early setting and curative intent overlap in most clinical scenarios. Palliative setting usually overlaps with distant metastases, and in many instances, may also include locally advanced setting. In our paper, the term “palliative” is used interchangeably with “non-curative/life-extending”.
  • This inherent flexibility of counting LoTs based on intent of therapy, or based on clinical setting, using the same minimum dataset, is likely to be a useful feature of our solution.
  • We believe that our proposed standard dataset will help with the semi-automated identification of patients for clinical trials based on prior therapies.

Our solution: In our paper entitled “Determining lines of therapy in patients with solid cancers: a proposed new systematic and comprehensive framework”, published in the British Journal of Cancer, available for open access at https://rdcu.be/ciB7b, we first defined key concepts and terms to enable a common vocabulary to emerge. Second, we outlined which therapies should be counted as prior LoTs, and why. Third, we proposed a novel standardized format for documenting LoT that captures multifaceted data in a standardized manner that should lead to greater uniformity and thus improved data quality. Fourth, we proposed a minimum dataset that needs to be collected in order to determine LoT (see Table). Finally, we outlined draft guidelines, and identified some key issues on this topic that require wider discussion within the oncology community.

Table: The proposed standard dataset (top row) comprises the minimum data to be collected regarding the current disease status (columns 1–4) and anti-cancer therapy (columns 5–9) that would help determine the line of therapy (LoT, column 10). (Taken from Reference 1, with permission)

 

According to Herbert Simon’s principle of ‘satisficing’, we can either find optimum solutions for a simplified world, or satisfactory solutions for a more realistic world. Our framework aims at the latter. We hope that future clinical trials will utilize some of our proposals for more precise definition of the previous therapies received by their patient population.

Reference

  1. Saini KS, Twelves C. Determining lines of therapy in patients with solid cancers: a proposed new systematic and comprehensive framework [published online ahead of print, 2021 Apr 13]. Br J Cancer. 2021. doi:10.1038/s41416-021-01319-8; Available open access at https://rdcu.be/ciB7b

Please sign in or register for FREE

If you are a registered user on Research Communities by Springer Nature, please sign in

Subscribe to the Topic

Cancer Biology
Life Sciences > Biological Sciences > Cancer Biology

Related Collections

With collections, you can get published faster and increase your visibility.

Digital Imaging

BJC’s Digital Imaging series is open to receiving submissions assessing:
  • State-of-the-art in digital imaging technology and computational output;
  • Radiomics;
  • Algorithms and approaches designed to predict therapeutic response and enable “personalized medicine”;
  • Interrogation of the immune microenvironment and implications for immunotherapy selection;
  • Challenges of implementation in clinical practice;
  • Controversial applications of digital imaging

Publishing Model: Hybrid

Deadline: Ongoing