To assist in the SCT decision process

MS3T: The SCT Spectrum Timing Tool

 KROEGER:

Regarding timing of transplantation three major factors have to be taken into consideration:
(1) Disease-specific factor (How advanced is the disease? What is the life expectancy without transplantation?)
(2) Transplant-specific factor? (Do I have a fully matched donor?)
(3) Patient-specific factor? (What is my age? Do I have a co-morbidity?)
These factors should be always taken into consideration for balancing the risk and the benefit of transplantation.

A Proposed Practical Approach to Risk Stratification of Myelofibrosis Patients Qualified for Hematopoietic Stem Cell Transplantation.

— Zhenya Senyak

In recognition of the pain, suffering and loss of life caused by qualified MPN patients delaying HSCT, a working group of MPN specialists, transplant specialist, MPN SCT patients and patient advocates formed the MPN Stem Cell Transplantation Timing Taskforce (MS3T) in January, 2018. Building on the considerable work done in prognostic stratification of MPN survival risk, the two primary objectives of the MS3T were to produce a tool (the SCT spectrum tool) to graphically indicate SCT temporal constraints and implement a communications program in collaboration with organizations supporting HSCT to introduce that tool to patients and hematologists not specializing in MPNs. (While the focus of this work is on intermediate-2 and high risk MF patients, it is anticipated tool itself would be applicable to high-risk PV and untreated AML.)

Two factors driving delay in SCT quickly surfaced in the course of initial research. (1) One is a natural concomitant of a rare disease. The majority of MPN/MF patients are treated by hematologists and physicians not specializing in the myeloproliferative neoplasms. (2) Exacerbating the effect of that great medical divide, the primary prognostic scales on which we rely to stratify risk were developed by scientists and MPN specialists, nearly all of whom had access to facilities and funding available through clinical trial and academic institutions.

As a result some of the most important work adding to the prognostic value of risk assessment scales — such as the addition of karyotype anomalies and a greatly expanded landscape of mutational disease drivers — is not generally available to MPN patients or their physicians. For example, the largest provider of medical services in the United States, the Veterans’ Administration does not honor routine requests by its hematologists for simple mutational analyses. In general, insurers in the private sector will decline such requests without substantiating evidence of its pressing need and application. The availability of chromesomal status reports can rarely be found in the charts of MPN patients.

At the same time some of the more commonly available clinical and somatic prognostic indicators of a progressing MPN disease — fibrosis, inflammation, splenomegaly, symptom burden, etc. — are only partially or not all included in DIPSS+. These are visible and palpable progression markers readily available to hematologists in general oncology/hematology centers and clinics.

Missing the target: Ready. Fire. Aim!

The really solid and important work done in discovering progression markers and epigentic pathways and all the rest appears in scientific journals. Someday, as the price of genetic sequencing falls, it will all come together toward a cure or at least decisive disease management. For now, however, very little of advanced molecular research has entered MPN practice. In general, the same tests, the same meds that have been prescribed since 2011 — and in some cases since 1950 — still represent the standard of care.

By not having access to routine mutational analyses, the work done in the labs and clinical trials, the lists published by NCCN, the extraordinary achievement of Grinfeld et al in combining comprehensive MPN genetic and clinical data, have little relevance for the guy walking out of his hematologist’s office shaking his head.

Hematologists and transplanters rely on visible, measurable and historic data in assessing SCT timing. Beyond elemental molecular data, i.e, JAK2 v617f status (available for $480 with a 5 day turnaround) no basic mutational or genetic information, no karyotype data is generally available in the MPN patient’s chart. And, considering the limited therapeutic options and the insurance hassles involved, there’s not much reason for a general practice hem/onc to order the analysis.

To be or not to be

In any case, the lack of easy access to molecular data may be irrelevant to the phenomenon of delayed SCT. The stem cell transplantation decision is personal. From a practical perspective, to the patient, no amount of mutational, polymorphic, epigenetic or karyotypic data will make a difference

For the patient who will ultimately chose, decline or delay stem cell transplant, choice is not primarily a clinical matter, a therapeutic decision based on risk/benefit analyses. Medical elements play a role, of course, but not nearly the dominant role we seek to encapsulate by converting biometrics into a compelling prognostic device to aid SCT patient hematologist or transplant specialist.

Transplantation is fundamentally an existential issue, informed by patient-specific concerns along with transplant- and medical-specific elements. Among concerns are long-term risk, isolation, failure, fear of outcome, affordability, availability of support, GVHD, relapse and current personal and psychosocial matters unique to each of us.

A fuller description of patient SCT avoidance strategies was presented by Jeanne Palmer et al from Mayo Clinic Scottsdale at ASH 2017 and in her paper “Views on Bone Marrow Transplant: a Survey of Patients with Myelofibrosis.” 

“The key problem,” writes Jerry Avorn in NEJM “is medicine’s ongoing assumption that clinicians and patients are, in general, rational decision makers. In reality, we are all influenced by seemingly irrational preferences in making choices about reward, risk, time, and trade-offs that are quite different from what would be predicted by bloodless if precise quantitative calculations.”

There appear to be three major influences determining the timing of SCT: The symptomatic impact of disease progression, physician influence, patient choice. And over it all, the specter of time.

SCT Decision Drivers
Red: MPN/ Green: MD/  Blue: Patient

It’s about time.

The root of delay in timely entry to SCT lies not in the lack of good clinical indicators but elsewhere. There is above all the uncertainty of how much time is available in which to make the SCT decision. The consulting hematologist who is usually not an MPN specialist is too often unfamiliar with polycythemia vera and the nature of myelofibrosis and its transformation to acute myeloid leukemia. He or she is unlikely to take aggressive early action to move the patient into SCT and its attendant life changing risks.

And then there is the patient living his life, doing his laundry, dreaming his dreams, engaged in the complex calculus required to stake all on an uncertain, risky procedure that promises both new life and the threat of death or the sharply reduced enjoyment of waning years. What’s the rush?

To reduce the delay-associated mortality of qualified SCT patients a simple, timing tool can help tip the balance by influencing the patient’s own sense of imminence. When coupled with intensive outreach to the medical, transplant and hematologic communities the qualified patient’s wavering commitment to SCT can be strengthened.

The Age Factor.

My own cause of death will never be delayed SCT. I am intermediate-2 MF, 81 years old with modest splenomegaly, cachexia and several serious co-morbidities. Even if I had a twin brother, an identical match, a strong domestic support team and unlimited funds,  SCT is off the table. This is an easy decision. In fact it’s hardly a medical decision at all. I have a good QOL, only a limited time to death and to squander any of that on the rigors of induction, the uncertainties of recovery and near certainty of somewhat diminished capabilities for the remaining weeks or month of post-SCT life would be foolhardy.

I had to personally age before I could gain the perspective of a Host, the object and not the observer of molecular events unrolling in my genome tightly wrapped in its nuclear cellular vault. To feel the response of a dynamic genome as it reacts to the multiple insults to which flesh is heir is to realize the constraints aging places on our ability to adapt. It appears likely we should weigh aging more heavily in stratifying SCT risk assessment components.

Maybe not. Maybe conditioning and the state of our overall health can modify the effects of time on our outcome.

Accumulated over time the multiple billions of natural mitotic events with their opportunities for mutations, translocation, misssteps, coupled with exposure to the pathogens, radiation, and biological stresses surely impact all humans who have lived on the planet seven or more decades. For MPN patients, even with the generous extrapolation of an extended lifespan, the toll of the hematologic chaos — disease progression, therapeutic impacts, compromised immune system — that brought >65 year olds to consideration of SCT may severely limit the anticipated return on investment as measured by good QOL years remaining post transplant.

So why not, the thinking goes, live with the devil we know and hope for the magic silver bullet lodged in the armory of clinical trial and genetic engineering labs…or, at the very least, enjoy life now and anticipate good palliative care at the end. With the onset of MF occurring during later stages of life, focus on molecular and genetic elements of myelofibrosis, on karyotype and mutational landscape will have little impact on the SCT decision process. The compelling forces will be symptoms, pain, and awareness that time has about run out.

To the younger MF patient with a different view of the future, what weighs more heavily in the decision process are anticipated outcomes.

Ultimately, again, it’s about time.

For the elderly, too little, for the younger too much. For the elderly patient sacrificing remaining time for the mortality-shadowed option of a reduced symptom QOL is a bad bet, a zero sum game with lousy odds. For the younger patient, one critical problem with current prognostic scoring systems is the seeming availability of too much time, nearly two years or so for the highest risk DIPSS+ patient.

Too much time. Ian, Harvey and Bob were three of the most knowledgeable MPN patients, all married with families and active personal and professional lives, strong support systems, excellent MPN specialist hematologists. They were all at MF high risk and living in that expanded bubble of paroled time. The sudden appearance and expansion of blasts reduced all their options to a rarely successful salvage SCT.

The Spectrum Tool

To reduce the delay-associated mortality of qualified SCT patients a simple, calibrated timing tool coupled with intensive outreach to the medical and hematologic communities can help tip the balance by influencing the patient’s own sense of imminence and shoring up support of physicians  who are only rarely exposed to myeloproliferative neoplasms.

The issue of age and how to stratify its impact is only one of the thorny problems to be resolved in programming the Spectrum tool. In most prognostic systems, age is given the same weight as hemogloblin and leukocyte levels. And yet we differ so widely that biologically we defy our chronological age.

Building on the widely used, carefully curated prognostic scales, a patient oriented SCT spectrum timing tool needs to incorporate commonly available datasets, ready accessibility to physician, transplanter and patient alike, and a clear and rapid assessment of the multiple inputs required to make a timely SCT decision. Where possible, the stratification elements to be weighed need to have been derived from proven data. Once the inputs are determined, a web based computer app capable of historic input, dynamic record updating, weighting of multiple contributing elements, assessment and generation of a clear timing signal might be the ideal medium.

MIPSS, DIPSS+ and unfavorable karyotype

We do include the five IPSS and DIPSS risk survival factors in our Spectrum Tool inputs (age>65, Hg <10,. Leukocyte >25×109, blast>2%, and constitutional symptoms) but don’t spend much time discussing MIPSS or even DIPSS Plus for the same reason we largely ignore unfavorable karyotype. In the felicitous phrase of Nassema Gangat, (In JCO 2016)  molecular profiling is described as promising but “not ubiquitously available.”  Laura Michaelis in her 2017 Haematologica comment on the MD Anderson “Simple Model” (presented below) — and by implication all the surfacing MPN molecular findings — commends the  Simple scale for relying on “easy to obtain, objective and reproducible data,” and not requiring next generation sequencing “a test which has highly variable reimbursement patterns and is financially out of reach for many patients.”

Spectrum – Contributing elements.

A survey of the MS3T yielded a list of items that could be constructively added to existing prognostic scales in developing a SCT spectrum timing tool: Rec0mmendations include: Updated weighting of components in the DIPSS plus, inclusion of fibrosis, splenomegaly and inflammation as prognostic indicators, fitness assessment, clonal evolution, incorporation of new molecular drivers, QOL, broader symptom evaluation, and personal issues like support and finance. (See complete provisional list, below.)

Reference to age as an independent risk factor and the need to update our exception criteria was a common theme.

Recently, the cut-off age for HSCT was ≤65. With the advent of improved technique and Reduced Intensity Conditioning, that has extended to age 70 and beyond. Work is needed to further study the impact of aging on overall survival of SCT but enough may already be known of OS to increase the weighting of the age factor in risk assessment.

Saeed et al. last year examined the adoption of the new comorbidity index (HCT-CI/Age) and compared the pre-transplant assessment of mortality of 114 consecutive patients undergoing HSCT. Their conclusion: “Despite our small sample population, HCT-CI/Age was more discriminative to identify patients with poor outcome that might benefit from intensified management strategies or other therapeutic approaches rather than allogeneic HCT.”

There is a precedent to add weight to age in assessing HCT risk. The Charlson Co-morbidity Index evaluated multiple co-morbidities to assess pre-transplant risk and added 1 point for each decade over age 40. Sorror, et al in 2014 in a Journal of Clinical Oncology paper walked back his prior endorsement of HCT-CI/Age in favor of combining a more sensitive instrument incorporating age with co-morbidities.

From Sorror, et al. “Age has been a controversial predictor in transplant outcome; a study from the US showed poor prediction of allogeneic HCT mortality when elderly patients were stratified by age. The European experience reported similar outcome of patients with age above 50. Sorror et al studied the effect of addition of age to HCT-CI and was able to show that age ≥40 years carried worse prognosis and deserved to be amended to HCT-CI scoring system to facilitate patients’ stratification prior to transplant .”

“Age has long been used as a major factor for assessing suitability for allogeneic hematopoietic cell transplantation (HCT). The HCT-comorbidity index (HCT-CI) was developed as a measure of health status to predict mortality risk after HCT. Whether age, comorbidities, or both should guide decision making for HCT is unknown. Data from 3,033 consecutive recipients of HLA-matched grafts from five institutions contributed to this analysis. .

Sorror concluded: “Age is a poor prognostic factor. The proposed composite measure allows integration of both comorbidities and age into clinical decision making and comparative-effectiveness research of HCT.”
A current study (Bhatt et al., March, 2018, Bone Marrow Transplant) explored the SCT experience of older patients using the National Cancer Database of older patients (61-75 yers) with intermediate o high risk wi acute myeloid leukemia and found much lower hematopoietic SCT use than the general population. “Only 5.5% of older patients (n-17,555) underwent HCT Factors listed for these underutilization were: “cared for in a non-academic hospital, race other than white, older age, Charlson comorbidity score, uninsured status, Medicaid or Medicare insurance and lower educational status.”
Simplicity — toward designing the Spectrum Tool

In Haematologica, Rozovski. Verstovsk et al. “An accurate simple prognostic model consisting of age, JAK2, CALR and ML mutation status for patients with Primary myelofibrosis set out to examine a simple, limited risk-assessment tool employing an age- and mutation status-based scoring system that looked at a limited menu of mutations and came up with some striking results.

“The applicability of our prognostic scale depends on screening for mutations in CALR and MPL and quantification of the JAK2V617F allele burden. Recently, the WHO added CALR and MPL mutations to the PMF diagnostic criteria and, as a result, most diagnostic laboratories perform these tests. Moreover, most diagnostic laboratories assess the presence of JAK2 mutations by using quantitative PCR. Although the JAK2V617F allele burden is readily available, it is not routinely reported, although various assays yield similar quantification results.

“Patients with a favorable mutation status (high Janus kinase 2V617F allele burden/myeloproliferative leukemia/calreticulin mutation) and aged 65 years or under had a median survival of 126 months. Patients with one risk factor (low Janus kinase 2V617F allele burden/triple-negative or age >65 years) had an intermediate survival duration, and patients aged over 65 years with an adverse mutation status (low Janus kinase 2V617Fallele burden or triple-negative) had a median survival of only 35 months. Our simple and easily applied age- and mutation status-based scoring system accurately predicted the survival of patients with primary myelofibrosis.

“Since the initial publication of the IPSS prognostic score, several refinements have been proposed, most of which attempt to incorporate recurrent gene mutations that have been identified in patients with PMF. Some mutations, such as those in DNTM3 or TET2, have not been shown to correlate with survival outcome. Conversely, mutations in ASXL1, SRSF2, and EZH2 predicted short survival in a large cohort of patients, and only the ASXL1 mutation remained statistically significant when added to the IPSS prognostic score. A report by Tefferi et al. points to the CALR/ASXL1+ profile as the most detrimental mutation profile in PMF.

“The applicability of our prognostic scale depends on screening for mutations in CALR and MPL and quantification of the JAK2V617F allele burden. Recently, the WHO added CALR and MPL mutations to the PMF diagnostic criteria and, as a result, most diagnostic laboratories perform these tests. … By using only 2 variables, we developed a simple, easily applied model with excellent discrimination power for survival outcome of patients with newly diagnosed PMF.”

The sources of risk in transplant assessment

Two members of the Taskforce, Nick Kroger and Ruben Mesa, collaborated on a 2008 paper in Leukemia. Choosing between stem cell therapy and drugs in myelofibrois,”  spells out the risks and factors that need to be considered in constructing a SCT timing tool. Although the title poses the alternative of choosing between SCT and drugs in MF, in 2008 there’s wasn’t much to be said about alternative drugs. Perhaps not so much now, either,

“Patients presenting with myelofibrosis range from being asymptomatic (discovered on investigation for occult causes of leukocytosis or splenomegaly) to being severely debilitated. The median age at diagnosis is typically in the seventh decade of life (median 67 years),although patients may clearly present in the third to fifth decades of life. Symptoms related to PMF can be broken down into one of three main categories: myeloproliferative, cytopenias and constitutional.

“The natural history of myelofibrosis is quite variable as patients may experience morbidity or mortality directly from the clinical consequences of myelofibrosis or from transformation to acute myeloid leukemia (AML) (that is, PMF or post-ET/PV blast phase). Death, however, may well arise from the complications of myelofibrosis (that is, cytopenias) exacerbating underlying comorbidities such as coronary artery disease, or given the median age of the population afflicted, death from an unrelated comorbidity is always possible.”

A flexible system, an uncertain result

To illustrate how uncertain predictions of overall post transplant survival are, consider a Nakaya et al.Japanese study published in Biol Blood Marrow Transplant. “Does the hematopoietic cell transplantation specific comorbidity index (HCT-CI) predict transplantation outcomes?

Researchers from Seattle developed the hematopoietic cell transplantation-specific comorbidity index (HCT-CI), to determine the risk of mortality in several retrospective studies. However, its clinical utility has not been documented in prospective studies. The researchers Nakaya et al: “The aim of the present study was to evaluate the utility of the HCT-CI prospectively in a multicenter setting… We found that the HCT-CI in its original scale failed to predict OS and NRM in this set of patients.”

So Nakaya et al. devised a “flexible” HCT-CI system restratifying scores. This flexible system did seem to do better at predicting outcomes in the beginning but did not hold up in the multivariate analyses two years later. One predictor that did was chronological age.

Constitutional Symptoms — the first and final frontier

Symptoms are disease made manifest, the direct and palpable evidence of phenotypic activity working through its human host,  The constitutional result might be pain and early satiety but the clearly visible symptoms are splenic swelling and weight loss.  The accessibility of symptomatic  data coupled with the patient’s own subjective reporting makes it an ideal input for the Spectrum Tool.  It combines the patient’s self-knowledge and volition with readily accessible physical phenomena

The work on Patient Reported Outcomes and MPN fatigue burden, QOL and symptom assessment done in Scottsdale by Ruben Mesa, Robyn Scherber and the Mayo Clinic group is already central to MF diagnostics and research. As that work continues, we have the MPN-SAF as a benchmark tool to convert analog MF phenomena into a workable digital and prognostic format.

MPNSAF QOL study predicts OS

 

 

Basic guide for construction of the Spectrum Tool algorithm

“Patients candidacy for allo-SCT in PMF was mostly based upon the evidence that median overall survival (OS) for DIPSS intermediate-2 and high-risk patients after allo-SCT was superior to that after non-transplant management. At variance, the median OS of low-risk patients was shorter after transplantation than with non-transplant management. “Indication and management of allogenic stem cell transplantation in primarh myhelofibroisis: a consensus process by an EBMT/ELN international working group.” Leukemia 2015, Kroger et al/

Note on proposed absence of  relevant prognostic scoring system components.

As the SCT Spectrum Timing Tool is designed for use by non-MPN specializing hematologists in private and group practice and MPN patients there is little point in including unfavorable karyotype, complex or single abnormalities as (1) those data are likely unavailable and (2) acquiring those data would not, based on currently available meds, alter the therapeutic course or impact the SCT timing decision. For those same reasons of unavailability and lack of affect in the event of discovery the several driving and ancillary mutations discovered since creation of DIPSS+ eight years ago — notably ASXL1, SRSF2, IDH1/2 and EZH2 associated with poor survival — might be omitted from the Tool”s resident database as well.

Incorporation of the myelofibrois grading system of the WHO 2016 revisions (Arber, Orazi et al.bloodjournal)

Suggested prognostic components of the SCT Spectrum Timing App ( Note to the MS3T: Each element requires elucidation of stages, ranking in importance and consideration for acceptance/rejection.)

There are 13 broad elements in the prognostic LIST for consideration and stratification in the Spectrum tool

1. Three mutations: JAK2, CALR and MPL

2. AGE , levels <40 and each decade>40

3. Fibrosis, WHO grading,

4. Splenomegaly three or more levels,

5. Constitutional symptoms (from the MPNSAF )

6. Blasts in PB,

7. Transfusion dependence,

8..Hg<10

9.. leuk0cytes >>25×109

10. Unintended Weight loss,

11. Platelet count,

12. Modified co-morbidity index (HCI),

13. Derived psycho-social scale .

The form of the Spectrum tool

Considering the need for secure data storage, repeated updating of history and test results, and easy portable access by patients and physicians, the idea format for the SCT Timing Spectrum tool would be a computer app. On approval of the decision algorithm associated with the prognostic elements, design and programming of the app would be relatively inexpensive.

The  Grinfeld app:  https://jg738.shinyapps.io/mpn_app/

Personalized Prognostic Predictions for Patients with Myeloproliferative Neoplasms through Integration of Comprehensive Genomic and Clinical Information http://www.bloodjournal.org/content/130/Suppl_1/491   Grinfeld:  “We then developed a unifying predictive model for all MPN patients. In order to take into account the striking degree of heterogeneity in genetic events, clinical characteristics and potential clinical outcomes, we developed a multi-state random effects Cox proportional hazards model. This allowed integration of a total of 63 clinical and genomic variables in order to generate individualised patient predictions for survival and disease transformation for all MPN patients.” http://www.bloodjournal.org/content/130/Suppl_1/491

The DIPSS+ score calculator: https://qxmd.com/calculate/calculator_315/dipss-plus-score-for-prognosis-in-myelofibrosis

Novartis, Know your score: http://www.mpn10app.com/

 

 

 

Gangat et al.
DIPSS plus: a refined dynamic international prognostic scoring system for primary myelofibrosis that incorporates prognostic information from karyotype, platelet count, and transfusion status.
JCO 2010 29:392-397

Passamonti et al.
A dynamic prognostic model to predict survival in primary myelofibrosis: a study by the IWG-MRT (International Working Group for Myeloproliferative Neoplasms Research and Treatment).
Blood 2010:115:1703-1708.

 

 

 

Advertisements

Comments on: "MS3T: The SCT Spectrum Timing Tool" (1)

  1. Ruben Mesa said:

    Ruben MEsa – order of priority (most important first) and comments (2, (next should be other high risk somatic mutations – i.e. asxl1 (not listed currently), 12, 6, 7,1,9,10,8, 11 (specifically marked non drug related thrombocytpenia 100 x 10()/L or less), 5, 4 3, 13 (however if severe psycho-social or mood disorders would be higher)

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Tag Cloud

%d bloggers like this: