2  MPN-SAF Total Symptom Score (TSS)

Scoring Guide and Clinical Interpretation

Author

Ashwin Kishtagari

Published

August 19, 2026

Show code 
library(tidyverse)
library(gt)

2.1 Overview

The Myeloproliferative Neoplasm Symptom Assessment Form Total Symptom Score (MPN-SAF TSS), also known as the MPN-10, is a validated patient-reported outcome (PRO) measure used to assess symptom burden in patients with myeloproliferative neoplasms (MPNs).

The instrument was developed and validated by Emanuel et al. (2012) in a prospective international study of 1,408 MPN patients.

2.2 The 10 Symptom Items

The MPN-SAF TSS consists of 10 items, each scored on a 0-10 scale where 0 = absent and 10 = worst imaginable.

Show code 
items_df <- tibble(
  Item = 1:10,
  Symptom = c(
    "Fatigue",
    "Concentration problems",
    "Early satiety",
    "Inactivity",
    "Night sweats",
    "Itching (pruritus)",
    "Bone pain",
    "Abdominal discomfort",
    "Unintentional weight loss",
    "Fevers"
  ),
  Description = c(
    "Tiredness, weariness, lack of energy",
    "Difficulty focusing or thinking clearly",
    "Feeling full quickly after eating small amounts",
    "Problems with being active or moving around",
    "Excessive sweating during sleep",
    "Skin itching without apparent cause",
    "Pain in bones or joints",
    "Pain or discomfort in the abdomen",
    "Loss of weight without trying",
    "Body temperature above normal"
  ),
  `Common in` = c(
    "All MPNs",
    "All MPNs",
    "MF, PV",
    "MF",
    "MF, PV",
    "PV, MF",
    "MF",
    "MF (splenomegaly)",
    "MF",
    "MF"
  )
)

items_df |>
  gt() |>
  tab_header(
    title = "MPN-SAF TSS Items",
    subtitle = "10-item symptom assessment"
  ) |>
  cols_align(align = "center", columns = Item) |>
  cols_width(
    Item ~ px(60),
    Symptom ~ px(180),
    Description ~ px(280),
    `Common in` ~ px(100)
  ) |>
  tab_style(
    style = cell_fill(color = "#E8F4F8"),
    locations = cells_body(rows = seq(1, 10, 2))
  ) |>
  tab_style(
    style = cell_text(weight = "bold"),
    locations = cells_column_labels()
  ) |>
  tab_footnote(
    footnote = "MF = Myelofibrosis; PV = Polycythemia Vera; ET = Essential Thrombocythemia",
    locations = cells_column_labels(columns = `Common in`)
  )
Table 2.1: MPN-SAF TSS Symptom Items
MPN-SAF TSS Items
10-item symptom assessment
Item Symptom Description Common in1
1 Fatigue Tiredness, weariness, lack of energy All MPNs
2 Concentration problems Difficulty focusing or thinking clearly All MPNs
3 Early satiety Feeling full quickly after eating small amounts MF, PV
4 Inactivity Problems with being active or moving around MF
5 Night sweats Excessive sweating during sleep MF, PV
6 Itching (pruritus) Skin itching without apparent cause PV, MF
7 Bone pain Pain in bones or joints MF
8 Abdominal discomfort Pain or discomfort in the abdomen MF (splenomegaly)
9 Unintentional weight loss Loss of weight without trying MF
10 Fevers Body temperature above normal MF
1 MF = Myelofibrosis; PV = Polycythemia Vera; ET = Essential Thrombocythemia

2.3 Scoring Methodology

2.3.1 Total Symptom Score Calculation

The TSS is calculated as the sum of all 10 individual symptom scores.

\[\text{TSS} = \sum_{i=1}^{10} \text{Item}_i\]

Score Range: 0 to 100 (higher = greater symptom burden)

Show code 
#' Calculate MPN-SAF Total Symptom Score
#'
#' @param fatigue Numeric 0-10
#' @param concentration Numeric 0-10
#' @param early_satiety Numeric 0-10
#' @param inactivity Numeric 0-10
#' @param night_sweats Numeric 0-10
#' @param itching Numeric 0-10
#' @param bone_pain Numeric 0-10
#' @param abdominal_discomfort Numeric 0-10
#' @param weight_loss Numeric 0-10
#' @param fevers Numeric 0-10
#' @return Numeric TSS (0-100)
calculate_tss <- function(fatigue, concentration, early_satiety, inactivity,
                          night_sweats, itching, bone_pain, 
                          abdominal_discomfort, weight_loss, fevers) {
  
  scores <- c(fatigue, concentration, early_satiety, inactivity,
              night_sweats, itching, bone_pain, abdominal_discomfort,
              weight_loss, fevers)
  
  # Validate inputs
  if (any(scores < 0 | scores > 10, na.rm = TRUE)) {
    warning("Scores should be between 0 and 10")
  }
  
  sum(scores, na.rm = TRUE)
}

# Example calculation
example_tss <- calculate_tss(
  fatigue = 7,
  concentration = 5,
  early_satiety = 6,
  inactivity = 4,
  night_sweats = 8,
  itching = 3,
  bone_pain = 5,
  abdominal_discomfort = 6,
  weight_loss = 2,
  fevers = 4
)

2.3.2 Example Calculation

Show code 
example_df <- tibble(
  Symptom = items_df$Symptom,
  Score = c(7, 5, 6, 4, 8, 3, 5, 6, 2, 4)
)

example_df |>
  gt() |>
  tab_header(
    title = "Example Patient Scores",
    subtitle = glue::glue("Total Symptom Score = {sum(example_df$Score)}")
  ) |>
  cols_align(align = "center", columns = Score) |>
  tab_style(
    style = cell_text(weight = "bold", color = "#0066CC"),
    locations = cells_body(columns = Score)
  ) |>
  grand_summary_rows(
    columns = Score,
    fns = list(TSS = ~sum(.)),
    fmt = ~fmt_number(., decimals = 0)
  ) |>
  tab_style(
    style = list(
      cell_fill(color = "#FFE4B5"),
      cell_text(weight = "bold", size = px(14))
    ),
    locations = cells_grand_summary()
  )
Table 2.2: Example TSS Calculation
Example Patient Scores
Total Symptom Score = 50
Symptom Score
Fatigue 7
Concentration problems 5
Early satiety 6
Inactivity 4
Night sweats 8
Itching (pruritus) 3
Bone pain 5
Abdominal discomfort 6
Unintentional weight loss 2
Fevers 4
TSS — 50

2.4 Reference Values and Interpretation

2.4.1 Population Norms

Based on the validation study of 1,408 MPN patients (Emanuel et al., 2012):

Show code 
norms_df <- tibble(
  `Disease Subtype` = c(
    "Essential Thrombocythemia (ET)",
    "Polycythemia Vera (PV)", 
    "Myelofibrosis (MF)",
    "All MPNs Combined"
  ),
  N = c(NA, NA, NA, 1408),
  `Mean TSS` = c(18.7, 21.8, 25.3, 21.2),
  `SD` = c(15.3, 16.3, 17.2, 16.3),
  `Interpretation` = c(
    "Lowest symptom burden",
    "Intermediate",
    "Highest symptom burden",
    "Overall reference"
  )
)

norms_df |>
  gt() |>
  tab_header(
    title = "TSS Reference Values",
    subtitle = "From validation cohort (n = 1,408)"
  ) |>
  fmt_number(columns = c(`Mean TSS`, SD), decimals = 1) |>
  sub_missing(missing_text = "—") |>
  cols_align(align = "center", columns = c(N, `Mean TSS`, SD)) |>
  tab_style(
    style = cell_fill(color = "#D4EDDA"),
    locations = cells_body(rows = 1)
  ) |>
  tab_style(
    style = cell_fill(color = "#FFF3CD"),
    locations = cells_body(rows = 2)
  ) |>
  tab_style(
    style = cell_fill(color = "#F8D7DA"),
    locations = cells_body(rows = 3)
  ) |>
  tab_style(
    style = cell_fill(color = "#E2E3E5"),
    locations = cells_body(rows = 4)
  ) |>
  tab_source_note(
    source_note = "Source: Emanuel et al. J Clin Oncol 2012;30:4098-103"
  )
Table 2.3: MPN-SAF TSS Reference Values by Disease Subtype
TSS Reference Values
From validation cohort (n = 1,408)
Disease Subtype N Mean TSS SD Interpretation
Essential Thrombocythemia (ET) — 18.7 15.3 Lowest symptom burden
Polycythemia Vera (PV) — 21.8 16.3 Intermediate
Myelofibrosis (MF) — 25.3 17.2 Highest symptom burden
All MPNs Combined 1408 21.2 16.3 Overall reference
Source: Emanuel et al. J Clin Oncol 2012;30:4098-103

2.4.2 Severity Classification

Show code 
severity_df <- tibble(
  Category = c("Low", "Moderate", "High", "Very High"),
  `TSS Range` = c("0-10", "11-25", "26-44", "≥45"),
  `Percentile (approx.)` = c("<25th", "25th-50th", "50th-85th", ">85th"),
  `Clinical Implication` = c(
    "Minimal symptoms; monitor",
    "Moderate burden; consider symptom-directed therapy",
    "Significant burden; treatment indicated",
    "Severe burden; urgent intervention needed"
  )
)

severity_df |>
  gt() |>
  tab_header(
    title = "TSS Severity Categories",
    subtitle = "Clinical interpretation guide"
  ) |>
  cols_align(align = "center", columns = c(`TSS Range`, `Percentile (approx.)`)) |>
  tab_style(
    style = list(cell_fill(color = "#28A745"), cell_text(color = "white")),
    locations = cells_body(rows = 1, columns = Category)
  ) |>
  tab_style(
    style = list(cell_fill(color = "#FFC107")),
    locations = cells_body(rows = 2, columns = Category)
  ) |>
  tab_style(
    style = list(cell_fill(color = "#FD7E14"), cell_text(color = "white")),
    locations = cells_body(rows = 3, columns = Category)
  ) |>
  tab_style(
    style = list(cell_fill(color = "#DC3545"), cell_text(color = "white")),
    locations = cells_body(rows = 4, columns = Category)
  ) |>
  tab_footnote(
    footnote = "Categories based on distribution from validation cohort (Mean 21.2, SD 16.3)",
    locations = cells_column_labels(columns = `TSS Range`)
  )
Table 2.4: Suggested TSS Severity Categories
TSS Severity Categories
Clinical interpretation guide
Category TSS Range1 Percentile (approx.) Clinical Implication
Low 0-10 <25th Minimal symptoms; monitor
Moderate 11-25 25th-50th Moderate burden; consider symptom-directed therapy
High 26-44 50th-85th Significant burden; treatment indicated
Very High ≥45 >85th Severe burden; urgent intervention needed
1 Categories based on distribution from validation cohort (Mean 21.2, SD 16.3)

2.4.3 Interpreting a TSS of 50

A TSS of 50 represents:

  • ~1.8 standard deviations above the mean (Mean 21.2 + 1.8 × SD 16.3 ≈ 50)
  • Approximately the top 5-10% of symptom burden in MPN patients
  • Classified as “Very High” severity
  • Patient likely experiencing substantial impairment in quality of life
Show code 
# Simulate distribution based on published parameters
set.seed(42)
tss_dist <- tibble(
  tss = pmax(0, pmin(100, rnorm(10000, mean = 21.2, sd = 16.3)))
)

ggplot(tss_dist, aes(x = tss)) +
  geom_histogram(aes(y = after_stat(density)), bins = 50, 
                 fill = "#4A90A4", alpha = 0.7, color = "white") +
  geom_density(linewidth = 1, color = "#2C5F72") +
  geom_vline(xintercept = 50, color = "#DC3545", linewidth = 1.5, linetype = "dashed") +
  geom_vline(xintercept = 21.2, color = "#28A745", linewidth = 1, linetype = "solid") +
  annotate("text", x = 52, y = 0.035, label = "TSS = 50\n(Very High)", 
           hjust = 0, color = "#DC3545", fontface = "bold", size = 3.5) +
  annotate("text", x = 23, y = 0.04, label = "Mean = 21.2", 
           hjust = 0, color = "#28A745", fontface = "bold", size = 3.5) +
  annotate("rect", xmin = 45, xmax = 100, ymin = 0, ymax = 0.045, 
           alpha = 0.15, fill = "#DC3545") +
  scale_x_continuous(breaks = seq(0, 100, 10), limits = c(0, 100)) +
  labs(
    x = "Total Symptom Score (TSS)",
    y = "Density",
    title = "MPN-SAF TSS Distribution"
  ) +
  theme_minimal(base_size = 12) +
  theme(
    plot.title = element_text(face = "bold"),
    panel.grid.minor = element_blank()
  )
Figure 2.1: Distribution of TSS Scores with Reference Point at TSS = 50

2.5 Response Criteria

2.5.1 ≥50% TSS Reduction

In clinical trials, a ≥50% reduction in TSS is commonly used as a meaningful response criterion.

Show code 
#' Assess TSS Response
#'
#' @param baseline_tss Numeric baseline TSS
#' @param followup_tss Numeric follow-up TSS
#' @return List with percent change and response status
assess_tss_response <- function(baseline_tss, followup_tss) {
  
  pct_change <- ((followup_tss - baseline_tss) / baseline_tss) * 100
  
  response <- case_when(
    pct_change <= -50 ~ "Response (≥50% reduction)",
    pct_change < 0 ~ "Improvement (<50% reduction)",
    pct_change == 0 ~ "Stable",
    pct_change > 0 ~ "Worsening"
  )
  
  list(
    baseline = baseline_tss,
    followup = followup_tss,
    absolute_change = followup_tss - baseline_tss,
    percent_change = round(pct_change, 1),
    response = response
  )
}

# Example: Patient with baseline TSS of 50
example_response <- assess_tss_response(baseline_tss = 50, followup_tss = 22)

Example: A patient with baseline TSS of 50 who improves to TSS of 22:

  • Absolute change: -28 points
  • Percent change: -56%
  • Classification: Response (≥50% reduction)

2.6 Correlation with Quality of Life

The MPN-SAF TSS demonstrates strong correlation with quality of life measures:

Show code 
correlation_df <- tibble(
  Scale = c(
    "Global Quality of Life",
    "Physical Functioning",
    "Role Functioning",
    "Emotional Functioning",
    "Cognitive Functioning",
    "Social Functioning"
  ),
  `Correlation (r)` = c(-0.59, -0.56, -0.56, -0.50, -0.50, -0.48),
  Interpretation = c(
    "Strong negative correlation",
    "Moderate-strong negative",
    "Moderate-strong negative", 
    "Moderate negative",
    "Moderate negative",
    "Moderate negative"
  )
)

correlation_df |>
  gt() |>
  tab_header(
    title = "TSS vs. Quality of Life Correlations",
    subtitle = "Higher TSS associated with worse QoL"
  ) |>
  fmt_number(columns = `Correlation (r)`, decimals = 2) |>
  cols_align(align = "center", columns = `Correlation (r)`) |>
  data_color(
    columns = `Correlation (r)`,
    palette = c("#DC3545", "#FFC107", "#28A745"),
    domain = c(-0.6, -0.45)
  ) |>
  tab_footnote(
    footnote = "All correlations p < 0.001",
    locations = cells_column_labels(columns = `Correlation (r)`)
  )
Table 2.5: TSS Correlation with EORTC QLQ-C30 Functional Scales
TSS vs. Quality of Life Correlations
Higher TSS associated with worse QoL
Scale Correlation (r)1 Interpretation
Global Quality of Life −0.59 Strong negative correlation
Physical Functioning −0.56 Moderate-strong negative
Role Functioning −0.56 Moderate-strong negative
Emotional Functioning −0.50 Moderate negative
Cognitive Functioning −0.50 Moderate negative
Social Functioning −0.48 Moderate negative
1 All correlations p < 0.001

2.7 Psychometric Properties

The MPN-SAF TSS demonstrates excellent psychometric properties:

  • Internal consistency: Cronbach’s α = 0.83 (excellent)
  • Factor structure: Single underlying construct (unidimensional)
  • Convergent validity: Strong correlations with QoL measures
  • Known-groups validity: Discriminates between disease subtypes

2.8 References

  1. Emanuel RM, et al. Myeloproliferative neoplasm (MPN) symptom assessment form total symptom score: prospective international assessment of an abbreviated symptom burden scoring system among patients with MPNs. J Clin Oncol. 2012;30(33):4098-103. DOI: 10.1200/JCO.2012.42.3863

  2. Langlais BT, et al. Validation of a Modified Version of the Myeloproliferative Neoplasm Symptom Assessment Form Total Symptom Score. J Hematol. 2021;10(5):207-211. DOI: 10.14740/jh914

  3. Scotch AH, et al. Symptom burden profile in myelofibrosis patients with thrombocytopenia: Lessons and unmet needs. Leuk Res. 2017;63:34-40. DOI: 10.1016/j.leukres.2017.10.002

Document generated: 2026-08-19