Local Therapy for Oligometastatic Disease—Cart Before the Horse?

Historically, metastatic cancers were treated with 1 approach: systemic therapies, usually in the form of cytotoxic chemotherapy combined with palliative interventions, the latter without any tumor-directed intent.1Ramalingam S Belani C. Systemic chemotherapy for advanced non-small cell lung cancer: Recent advances and future directions.Oncologist. 2008; 13: 5-13Crossref PubMed Scopus (296) Google Scholar In the event of any site of disease progression, there would be a switch from this first-line systemic therapy to a second-line systemic therapy, usually less effective. Local therapies were only prescribed for palliation of pain, bleeding, obstructive symptoms, neurologic compromise, and so on. Taking a step back, metastatic cancers were traditionally described at diagnosis as disease states that extended beyond the boundaries of en bloc surgical resection or radiation for locally advanced disease. There was inadequate thought to identifying subgroups of unique patients with unique anatomic and spatial distributions of metastatic disease with potentially better prognoses until surgical series described resections of limited sites of metastatic disease to improve both the survival and quality-of-life outcomes for patients with cancer in the early 20th century.2Palma DA Salama JK Lo SS et al.The oligometastatic state—Separating truth from wishful thinking.Nat Rev Clin Oncol. 2014; 11: 549-557Crossref PubMed Scopus (205) Google Scholar Then, in the 1990s, Hellman and Weichselbaum formalized the notion that certain cohorts of patients with metastatic cancer could experience a prolonged time interval with limited sites of metastatic disease, and hence, the oligometastatic disease (OMD) paradigm was established.3Hellman S Weichselbaum RR Oligometastases.J Clin Oncol. 1995; 13: 8-10Crossref PubMed Google Scholar Although our medical oncology colleagues had started to parse patients with metastatic cancer into more favorable categories based on mutational spectra and associated therapies (ie, biology), surgeons and then radiation oncologists started to identify subsets of patients with metastases who could benefit beyond palliation with local therapies, based on locations of disease spread and numbers of lesions. The first subset included patients with stage 4 cancers, from colorectal and sarcoma primaries, who either had disease that metastasized to limited organs (eg, lungs, liver, and adrenals) and/or in limited numbers.4Strong VE D'Angelica M Tang L et al.Laparoscopic adrenalectomy for isolated adrenal metastasis.Ann Surg Oncol. 2007; 14: 3392-3400Crossref PubMed Scopus (163) Google Scholar, 5Pastorino U Buyse M Friedel G et al.International Registry of Lung Metastases. Long-term results of lung metastasectomy: Prognostic analyses based on 5206 cases.J Thorac Cardiovasc Surg. 1997; 113: 37-49Abstract Full Text Full Text PDF PubMed Scopus (1291) Google Scholar, 6Fong Y Fortner J Sun RL Brennan MF Blumgart LH Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer: Analysis of 1001 consecutive cases.Ann Surg. 1999; 230: 309-318Crossref PubMed Scopus (3064) Google Scholar With the identification of tumor biomarkers and driver mutations (eg, HER2Neu, EGFR, and ALK) that could be targeted pharmacologically, subsets of metastatic solid tumors could start receiving directed systemic therapies beyond the 1-size-fits-all approach of cytotoxic chemotherapy.7Slamon DJ Leyland-Jones B Shak S et al.Use of chemotherapy plus a monoclonal antibody against HER2 for metastatic breast cancer that overexpresses HER2.N Engl J Med. 2001; 344: 783-792Crossref PubMed Scopus (9295) Google Scholar, 8Lynch TJ Bell DW Sordella R et al.Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib.N Engl J Med. 2004; 350: 2129-2139Crossref PubMed Scopus (9960) Google Scholar, 9Kwak EL Bang YJ Camidge DR et al.Anaplastic lymphoma kinase inhibition in non-small-cell lung cancer.N Engl J Med. 2010; 363: 1693-1703Crossref PubMed Scopus (3797) Google Scholar In more recent years, immuno-oncology (IO) use has also permitted longer disease-free survival intervals and overall survival in metastatic disease settings.10Peterson C Denlinger N Yang Y. Recent advances and challenges in cancer immunotherapy.Cancers (Basel). 2022; 14: 3972Crossref PubMed Scopus (1) Google Scholar However, despite leveraging IO to make fundamental strides in improving patient outcomes, this effort has yet to maximally personalize therapy, unless one thinks of quartiles or halves of tumor PD-L1 expression to be specific enough as a biomarker. As a reminder, IO is usually effective in one-third or so of patients with metastatic disease, but it is given to nearly all patients with stage 3 non-small cell lung cancer (NSCLC) adjuvantly.11Antonia SJ Villegas A Daniel D et al.PACIFIC InvestigatorsDurvalumab after chemoradiotherapy in stage III non-small-cell lung cancer.N Engl J Med. 2017; 377: 1919-1929Crossref PubMed Scopus (2390) Google Scholar The ALK experience in metastatic NSCLC is the one to follow, however. Clean and powerful preclinical data that left no mechanism ambiguous, followed by US Food and Drug Administration approval of crizotinib based on phase 1 studies with limited patient numbers, left little doubt that what was found preclinically would translate.9Kwak EL Bang YJ Camidge DR et al.Anaplastic lymphoma kinase inhibition in non-small-cell lung cancer.N Engl J Med. 2010; 363: 1693-1703Crossref PubMed Scopus (3797) Google Scholar,12Soda M Choi YL Enomoto M et al.Identification of the transforming EML4-ALK fusion gene in non-small-cell lung cancer.Nature. 2007; 448: 561-566Crossref PubMed Scopus (4263) Google Scholar That is altogether a less likely scenario in today's trial programs assessing local therapies for OMD, driven by animal studies using inadequate tumor models, inadequate understanding of host responses to therapy, early-phase trials without biologic import, and inappropriate metrics and readouts for the larger studies. More effective systemic therapies, in addition to novel noninvasive imaging efforts, have helped to make potential OMD more obvious and relevant. For radiation oncologists, the development of SBRT in the 2000s and its safe and feasible use for metastatic disease accelerated the use of local therapy for OMD. It has now taken nearly 20 years to generate enough data from single-institution experiences, small, single-arm prospective trials, real-world data experiences (prospective registries), and finally, small, randomized phase 2 trials to support the commencement of large, phase 3 randomized trials assessing local-therapy benefits in the management of OMD.13Katipally RR Pitroda SP Juloori A Chmura SJ Weichselbaum RR. The oligometastatic spectrum in the era of improved detection and modern systemic therapy.Nat Rev Clin Oncol. 2022; 19: 585-599Crossref PubMed Scopus (4) Google Scholar However, the only phase 3 trial to have shown a survival advantage with the use of local therapy for metastatic disease benefited from the biology of disease treatable with targeted therapy.14Wang XS Bai YF Verma V et al.Randomized trial of first-line tyrosine kinase inhibitor with or without radiotherapy for synchronous oligometastatic EGFR-mutated NSCLC.J Natl Cancer Inst. 2022; : djac015Crossref PubMed Google Scholar Another large randomized phase 3 (RPh3) study for oligometastatic breast cancer, NRG BR 002, closed early owing to futility in showing a survival benefit with the addition of local therapy in the experimental arm.15Chmura SJ, Winter KA, Woodward WA, et al. NRG-BR002: A phase IIR/III trial of standard of care systemic therapy with or without stereotactic body radiotherapy (SBRT) and/or surgical resection (SR) for newly oligometastatic breast cancer (NCT02364557). J Clin Oncol. 40;16(suppl):1007.Google Scholar The SARON study, a third RPh3 trial for OM NSCLC, was converted to an RPh2 study owing to slower-than-expected accrual (personal correspondence). This brings into question the feasibility of completing these larger RPh3 studies in a reasonable time frame. Even NRG LU 002, which, at 218 patients enrolled at interim analysis, is the largest randomized OMD trial to date, has taken 3 years to reach this goal, despite pragmatically incorporating all Food and Drug Administration approved systemic therapy regimens permitted for metastatic NSCLC. In 2017, Salama and colleagues raised the specter that radiation oncologists were nearly uniformly treating patients with oligometastases in consolidation or salvage (for oligoprogression) with stereotactic body radiation therapy (SBRT) off any clinical trial and generally with limited supportive data.16Lewis SL Porceddu S Nakamura N et al.Definitive stereotactic body radiotherapy (SBRT) for extracranial oligometastases: An international survey of >1000 radiation oncologists.Am J Clin Oncol. 2017; 40: 418-422Crossref PubMed Scopus (95) Google Scholar That pattern exists today, again despite the dearth of level 1 evidence supportive of these practice patterns altering survival with the most current systemic therapy regimens and despite multiple open clinical trials. Especially in light of newer systemic-therapy agents that have proinflammatory effects on host tissues that could lead to significant adverse events (ie, certain tyrosine kinase inhibitors and IO regimens), this approach of indiscriminately treating patients who have oligometastases with SBRT and other local therapy approaches is less than ideal, and it begs the question of whether most of these patients’ cases are being properly reviewed in multidisciplinary settings. Very few studies have described the patterns of failure in the setting of IO, making it difficult for us to understand whether local therapies would even be helpful or relevant. What have we learned from clinical trials in OMD? (1) The use of the number of metastases to enroll patients at diagnosis, consolidation, or oligoprogression is an exceedingly poor criterion— without the current existence of accurate predictive or prognostic models, it is unfortunately a snapshot in time (see Fig. 1), with disease burden staying the same or changing significantly. (2) Being agnostic to systemic therapy is good for including standard-of-care (SOC) treatments and generalizing an outcome but may miss the different synergies between IO regimens and radiation and/or surgery (with the practical implication that no single individual pharmaceutical company will support a trial agnostic to a systemic therapy regimen). (3) Contrarily focusing on 1 systemic therapy may miss effective alternative combinations. (4) Induced OMD versus disease that is oligopersistent from diagnosis represent different biology based on the primary cancer, suggesting that these cohorts of patients should be stratified or not included in the same trial. (5) Patients with targetable mutations should have their own trials (multi-institutional to speed accrual, in light of mutation frequency) and may need different sequencing in light of unique patterns of failure. (6) Metachronous versus synchronous disease again represent different biology or different time points in the evolution of the same disease, calling into question whether including both populations for inclusiveness and in support of trial enrollment could again compromise outcomes. (7) Doses and fractionation are other variables poorly understood biologically within the context of immune and other host-tissue responses; do trials need to permit use of high ablative doses, low ablative doses, or ablative doses at all? No one study can answer all of these questions. With better preclinical biology, validated with robust, early-stage, translation-heavy clinical trials, the outcomes of the larger phase 3 trials will be more likely to follow the preclinical findings. This means identifying metastatic tumor or host-tissue biomarkers predictive and prognostic of (1) durable responders to systemic therapies and (2) patients with true oligometastatic disease who will maximally benefit from local therapies synergistic with these systemic therapies. With all that being said, there is still good reason to be bullish about the future of local therapies as part of the management of OMD. As a field, we need to define when and under what conditions local therapy will add enough value while still offering a limited toxicity profile. However, as metastatic patients live longer, there will be greater appetite to continue to sequentially or concurrently apply local therapies, attempting to squeeze more radiation and/or surgery into less and less anatomic space without overlap. Most critically, we collectively need to establish with biology which patients will have sustained OMD or limited oligoprogressive disease over time. The following points highlight where we need to continue to refine our approach for treating OMD within the context of clinical trials and/or prospective evaluations.1.Sequencing of therapy and trial eligibility: For nontargetable disease, is consolidation the best time to enroll patients and use local therapy? It traditionally provides time for disease to declare itself and an assessment of response to systemic therapy. However, longitudinal biologic assessments, such as those conducted by our leukemia colleagues, should better inform the timing of local therapy administration as a function of true systemic disease burden and control. Rather than the number of lesions, we could follow circulating tumor DNA (ctDNA) levels to determine disease burden that can potentiate the development of new sites of disease; if that level is low, we may want to treat all visible metastases, no matter the number, if safe and obvious. If the ctDNA level is high, the disease being seen is the tip of the iceberg, and local therapy may be less relevant to disease outlook. Functional imaging with ctDNA measurements may provide the winning combination to identify earlier and more accurate disease events that predict which patients will optimally benefit from local therapy.2.Trial design: How about the use of sequential multiple-assignment randomized trial designs for early OMD studies?17Kidwell KM. SMART designs in cancer research: Past, present, and future.Clin Trials. 2014; 11: 445-456Crossref PubMed Scopus (35) Google Scholar The advantages are many: fewer patients needed to answer the relevant clinical and biologic questions, multiple questions answered in the assessment of multiple dynamic treatment regimens, and less time needed to answer the questions. Such an early-phase approach driven by biology may more likely ensure a successful RPh3 validation study. The time saved may not be so obvious with oncologic studies that are attempting to establish durable benefits of distinct therapies in the absence of robust predictive and prognostic tools that can act as early surrogates for outcomes. These predictive tools can be tumor intrinsic, tumor extrinsic and host intrinsic, or both. If ctDNA can be a surrogate, then it may become possible to generate early readouts of local treatment interventions.18Dang DK Park BH. Circulating tumor DNA: Current challenges for clinical utility.J Clin Invest. 2022; 132:e154941Crossref Scopus (1) Google Scholar Overall, early-phase, translational-heavy studies are a must. A phase 3 study of a single local-treatment approach for a single primary cancer in the setting of OMD could take 4 to 6 years to conduct. These days, in such a time frame, many systemic treatment regimens will be approved, leading the trial to be constantly amended if agnostic to systemic therapy. If it is not agnostic to systemic therapy, the systemic therapy and thus the question being asked may become obsolete. Hence, it is anticipated that umbrella and/or basket trials will need to be prioritized to increase the odds of identifying successful systemic-agent and local-therapy combinations for a given primary tumor. This brand of biomarker-enriched, early-phase studies in patients will be more valuable than parallel animal “clinical trials” in which host biology and responses contributing to treatment response and pharmacokinetics-pharmacodynamics do not match human responses, especially in more heavily treated patient populations.3.Real-world data: A UK consortium established that prospective registry studies could lead to significant insight into assessing the benefits of local therapy in OMD, enough to permit therapies to be considered SOC by the National Health System.19Chalkidou A Macmillan T Grzeda MT et al.Stereotactic ablative body radiotherapy in patients with oligometastatic cancers: A prospective, registry-based, single-arm, observational, evaluation study.Lancet Oncol. 2021; 22: 98-106Abstract Full Text Full Text PDF PubMed Scopus (73) Google Scholar The question is whether similar approaches would be accepted to change SOC in the US and elsewhere. The jury is out, but some combination of real-world data and biomarker-enriched clinical trials may be enough to justify use of new therapies in the OMD setting.4.Predictive and prognostic biomarkers for OMD: Very obviously, disease burden at diagnosis is a snapshot in time, as it is when there are limited sites of progressing disease or other time points currently dictated by computed tomography– or magnetic resonance–based imaging. Only after the fact (ie, months or years into a cancer diagnosis) can longitudinal information give us a sense of how that individual patient's disease biology evolved, interacted with host signaling, and responded to therapy (Fig. 1). That insight comes too late for many patients. As a consequence, we must make an effort to collect biologic specimens from relevant tissues—blood, tumor, and host tissues—as often as palatable and feasible on prospective clinical trials and/or registries to understand the complex dynamic and regulatory or counterregulatory signaling driving cancer behavior. With enough of these longitudinal studies, in which patients become their own controls (a way to account for the heterogeneity in tumor and host biology among individuals), we will start being able to develop predictive and prognostic biomarkers for subsets of patients with OMD. Biologic insight from serial tissue studies—novel pathways and not just 1-off molecules—need discovery biology, not just candidate biology, and may offer new targetable targets at the same time as biomarker development. Functional imaging and artificial intelligence-based interpretation of large data sets will become more mainstream but must incorporate the timing and extent of limited metastatic disease with recording of patterns of failure as additional data points. Ultimately, the genetics of the tumor and the host will dictate predictive and prognostic OMD models.5.Better understanding of the potential and limits of systemic therapy and its synergy with local therapy: Although our goal is to integrate local therapy into metastatic treatment paradigms, we must be cognizant and have understanding of the complex biologic strengths and limitations of novel systemic therapies, including a biologic appreciation of when and how synergy exists and when toxic effects with or without local therapy such as radiation manifest depending on different approaches (eg, conventional vs ablative or traditional sequencing vs pulsar). Micrometastatic disease response to systemic therapy induction may determine OMD status. Use of ctDNA after every cycle or course of systemic therapy to measure underlying response of what cannot be imaged or minimal residual disease may help us truly categorize patients as having oligometastatic disease, oligopersistent disease from diagnosis, or oligoprogressive disease, which could benefit from local therapies.6.Personalization of therapy: More than likely, OMD can be identified in a subset of all advanced primary cancer settings. For some patients, this may be recognized early in the disease process, and for others, later, all dependent on how their tumors respond to systemic therapy. The key is for our field, and not just medical oncology, to conduct the biologic studies necessary to understand why disease is not spreading to more distant sites. There may be something unique about that patient's immune recognition of disease, vascular limitations in host tissues, metabolic requirements, and so on that suggest a distinct pattern and timing of cancer spread. Knowing these parameters, 1 patient's tumor may need half as much radiation as another's to effectuate control of disease. All of these considerations are part of an effort to personalize OMD therapy once we can establish and characterize its presence in a patient. Early-phase, translation-heavy trials with biopsies of seen disease in the body and unseen disease in the blood repetitively will help us to personalize our knowledge of a patient's disease to personalize treatment. Is it extremely likely that a unique patient's host-tissue biology (not just tumor or serum immune repertoire but also liver, adipose, muscle, etc) will predict for tumor responses and benefits of local therapy in the setting of OMD, allowing us to personalize our approaches? In 30 years or so, our field has gone from theory to a potential shift in treatment paradigm in the management of patients with perceived OMD. This novel indication would represent one of the newest and most important advances and treatment indications in radiation oncology. However, a biologic basis for selecting the appropriate patients for local therapy will optimize the therapeutic ratio for this treatment paradigm.